Applying a Therapeutic Jurisprudence Perspective to the DOJ's Use of Deferred Prosecution Agreements for Corporate Offenders


Material Information

Applying a Therapeutic Jurisprudence Perspective to the DOJ's Use of Deferred Prosecution Agreements for Corporate Offenders
Physical Description:
1 online resource (184 p.)
Austell, Lynn Langton
University of Florida
Place of Publication:
Gainesville, Fla.
Publication Date:

Thesis/Dissertation Information

Doctorate ( Ph.D.)
Degree Grantor:
University of Florida
Degree Disciplines:
Criminology, Law, and Society, Sociology and Criminology & Law
Committee Chair:
Hollinger, Richard C
Committee Members:
Borg, Marian J
Lanza-Kaduce, Lonn M
Matheny, Albert R


Subjects / Keywords:
corporations -- deferred -- jurisprudence -- prosecution -- therapeutic
Sociology and Criminology & Law -- Dissertations, Academic -- UF
Criminology, Law, and Society thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation


Over the past two decades, the Department of Justice has steadily increased its use of deferred (DPA) and non-prosecution (NPA) agreements in criminal cases against corporations.  While the legalities of these agreements have been discussed extensively in numerous major law journals, the shift in the handling of corporate offenders has largely escaped the attention of the criminology community. This paper seeks to assess whether DPAs are more effective in reducing the collateral consequences of corporate punishment and preventing reoffending among corporate offenders than traditional criminal sanctions.  The assessment utilizes the therapeutic jurisprudence perspective—an approach to studying criminal justice practices and policies which focuses on the extent to which a justice-related action produces therapeutic or anti-therapeutic consequences for the offender and other persons impacted by the outcome. The therapeutic jurisprudence approach, which has never previously been applied to a policy related to corporate offending, first involves an examination of existing research in related areas to develop hypotheses as to whether the justice-related action - here the use of DPAs - is therapeutic. The non-empirical analysis is then followed by an empirical analysis testing these same assumptions. An examination of criminological theory and research related to the various components of DPAs, including deterrence, rational choice, organizational theories, procedural justice, and restorative justice, suggests that this practice is therapeutic on several levels.  However, the empirical analysis, which assesses whether DPAs are related to a lower likelihood of subsequent criminal charges than criminal sanctions, finds no relationship between the two types of case disposition and recidivism. Discussion of the limitations of the research and suggestions for future research are provided.
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Includes bibliographical references.
Source of Description:
Description based on online resource; title from PDF title page.
Source of Description:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility:
by Lynn Langton Austell.
Thesis (Ph.D.)--University of Florida, 2012.
Adviser: Hollinger, Richard C.
Electronic Access:

Record Information

Source Institution:
Rights Management:
Applicable rights reserved.
lcc - LD1780 2012
System ID:

This item is only available as the following downloads:

Full Text




2 2012 Lynn Langton Austell


3 ACKNOWLEDGMENTS This dissertation is dedicated to my parents and sister, John, Andrea, and Victoria Langton. Without their constant love, support, and encouragement I would never be where I am today. I am eternally grateful for the doors they opened and the opportunities they provided me. It is dedicated to my husband, Dan Austell, who has been endlessly loving, understanding, supportive, and patient through this process I am incredibly thankful to him for keeping me sane, always lending a listening ear for venting, providing much needed distractions, and giving me a bright spot in every day. Finally, it is dedicated to my dissertation committee, Drs. Richard Hollinger, Marian Borg, Lonn L anza Kaduce, and Albert Matheny, who have gone above and beyond in providing wisdom, guidance and advice over the many years of this process I am so thankful to them fo r their support and for not giving up on the idea or expectation that I would eventually finish someday.


4 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ ............... 3 LIST O F TABLES ................................ ................................ ................................ ........................... 7 LIST OF FIGURES ................................ ................................ ................................ ......................... 8 ABSTRACT ................................ ................................ ................................ ................................ ..... 9 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .................. 11 Why Formal Sanctions Are Not Effective Against Corporate Offenders .............................. 17 Overview of the Examination of DPAs through the Therapeu tic Jurisprudence Lens ........... 22 Voluntary Disclosure and Cooperation ................................ ................................ ........... 25 Compliance Monitor ................................ ................................ ................................ ........ 26 Monetary Penalties and Restitution ................................ ................................ ................. 26 Business Reforms and Compliance Programs ................................ ................................ 27 2 THERAPEUTIC JURISPRU DENCE ................................ ................................ .................... 28 Review of the Literature ................................ ................................ ................................ ......... 28 The Application of Therapeutic Jurisprudence in Extant Research ................................ ....... 33 3 THE DEVELOPMENT AND CURRENT APPLICATION OF DEFERRED PROSECUTION AND NON PROSECUTION AGREEMENTS ................................ ......... 38 Developing Guidelines for Charging and Sentencing Co rporations ................................ ...... 39 US Sentencing Commission Organizational Guidelines ................................ ................. 39 The Holder Memorandum ................................ ................................ ............................... 41 Corporate Offending at the Turn of the Century ................................ ............................. 43 Sarbanes Oxley ................................ ................................ ................................ ........ 44 Corporate Fraud Task Force ................................ ................................ ..................... 45 The Thompson Memorandum ................................ ................................ ......................... 46 The McCallum Memorandum ................................ ................................ ......................... 48 The McNulty Memorandum ................................ ................................ ............................ 50 Post McNulty DPA Changes ................................ ................................ ........................... 52 The Features of DPAs and NPAs ................................ ................................ ........................... 53 Voluntary Disclosure and Cooperation ................................ ................................ ........... 55 Compliance Monitor ................................ ................................ ................................ ........ 60 Business Reforms and Compliance Programs ................................ ................................ 64 Monetary Penalties and Restitution ................................ ................................ ................. 68


5 4 NON EMPIRICAL ANALYSIS OF THE THERAPEUTIC VALUE OF DEFERRED PROSECUTORIAL AGREEMENTS ................................ ................................ .................... 72 Voluntary Disclosure and Cooperation ................................ ................................ .................. 76 Business Reforms and Compliance Programs ................................ ................................ ........ 80 Business Reforms ................................ ................................ ................................ ..... 81 Compliance and Ethics Programs ................................ ................................ ............ 85 Monetary Penalties and Restitution ................................ ................................ ........................ 87 Compliance Monitors ................................ ................................ ................................ ............. 94 5 CASE STUDY: BRISTOL MYERS SQUIBB ................................ ................................ .... 101 Terms of the BMS DPA ................................ ................................ ................................ ....... 104 Voluntary Disclosure and Cooperation ................................ ................................ ......... 104 Business Reforms and Compliance Programs ................................ ............................... 106 Monetary Penalties and Restitution ................................ ................................ ............... 108 Compliance Monitor ................................ ................................ ................................ ...... 109 DPA Aftermath ................................ ................................ ................................ ..................... 111 Subsequent Criminal and Other Transgressions ................................ ............................ 111 Collateral Consequences ................................ ................................ ............................... 115 Case Study Concl usion ................................ ................................ ................................ ......... 119 6 EMPIRICAL ANALYSIS: DATA AND METHODOLOGY ................................ ............. 121 Current Focus ................................ ................................ ................................ ........................ 121 Data ................................ ................................ ................................ ................................ ....... 122 Hypothesis 1 Variables ................................ ................................ ................................ ......... 12 8 Dependent Variables ................................ ................................ ................................ ..... 128 Independent Variables ................................ ................................ ................................ ... 130 Compliance Monitor ................................ ................................ .............................. 131 Business Reforms and Compliance Programs ................................ ....................... 131 Monetary Penalties and Restitution ................................ ................................ ........ 132 Voluntary Disclosure and Cooperation ................................ ................................ .. 132 Controls ................................ ................................ ................................ .................. 133 Hypothesis 2 Variables ................................ ................................ ................................ ......... 134 Analytic Plan ................................ ................................ ................................ ........................ 134 7 EMPIRICAL ANALY SIS: FINDINGS ................................ ................................ ............... 136 General Descriptive Statistics ................................ ................................ ............................... 136 Hypothesis 1 Findings ................................ ................................ ................................ .......... 141 Hypothesis 2 Findings ................................ ................................ ................................ .......... 147 8 CONCLUSIONS, LIMITATIONS, AND FUTURE RESEARCH ................................ ..... 156 Discussion and Conclusion ................................ ................................ ................................ ... 156 Limitations and Suggestions for Future Research ................................ ................................ 161


6 LIST OF REFERENCES ................................ ................................ ................................ ............. 167 BIOGRAPHICAL SKETCH ................................ ................................ ................................ ....... 184


7 LIST OF TABLES Table page 7 1 Number and availability of corporate non prosecution and deferred prosecution agreem ents, by year, 1992 2009 ................................ ................................ ...................... 136 7 2 U.S. Attorneys' offices and Department of Justice litigating divisions overseeing more than one deferred (DPA) or non prosecution agreement (NPA) from 19 92 through 2009 ................................ ................................ ................................ .................... 138 7 3 Industries represented by corporations that entered into deferred (DPA) and non prosecution (NPA) agreements from 1993 through 2009 ................................ ................ 139 7 4 Offenses committed by corporations that entered into deferred (DPA) or non prosecution (NPA) agreements from 1992 through 2009 ................................ ................ 140 7 5 Desc riptive statistics for variables included in the model testing hypothesis 1 ............... 142 7 6 Components included in the deferred (DPA) and non prosecution (NPA) agreements of corporations which h ad subsequent criminal sanctions following the agreement ....... 144 7 7 Results of logistic regression predicting likelihood that a corporation which received a deferred or nonprosecution agree ment will recidivate given the terms of the agreement ................................ ................................ ................................ ......................... 147 7 8 Comparison of the offenses committed by corporations receiving deferred or non prosecution agreements and formal criminal s anctions, by year ................................ ..... 149 7 9 Comparison of punishment, prior criminal record, length of time from offense until disposition, and recidivism among corporations that received deferred and non prosecution agreements and formal criminal sanctions ................................ ................... 152 7 10 Descriptive statistics for variables included in the model testing Hypothesis 2 .............. 154 7 11 Results of logistic regression predicting the likelihood of corporate recidivism given the disposition of the case (deferred prosecution agreement versus criminal sanctions) ................................ ................................ ................................ ......................... 155


8 LIST OF FI GURES Figure page 5 1 Share prices for Bristol Myers Squibb (BMS) and competitors at fiscal year close, 2002 2011 ................................ ................................ ................................ ........................ 116


9 Abstract of Dissertat ion Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy APPLYING A THERAPEUT IC JURISPRUDENCE PER S USE OF DEFERRED PROSECUTION AGRE EMENTS FOR CORPORATE OFFENDERS By Lynn Langton Austell December 2012 Chair: Richard Hollinger Major: Criminology, Law and Society Over the past two decades, the Department of Justice has steadily increased its use of deferred (DPA) and non prosecutio n (NPA) agreements in criminal cases against corporations While the legalities of these agreements have been discussed extensively in numerous major law journals, the shift in the handling of corporate offenders has largely escaped the attention of the c riminology community. This paper seeks to assess whether DPAs are more effective in reducing the collateral consequences of corporate punishment and preventing reoffending among corporate offenders than traditional criminal sanctions. The assessment utili zes the therapeutic jurisprudence perspective an approach to studying criminal justice practices and policies which focuses on the extent to which a justice related action produces therapeutic or anti therapeutic consequences for the offender and other per sons impacted by the outcome. The therapeutic jurisprudence approach which has never previously been applied to a policy related to corporate offending, first involves a n examination of exist ing research in related areas to develop hypotheses as to whethe r the justice related action here the use of DPAs is therapeutic. The non empirical analysis is then followed by an empirical analysis testing these same assumptions. An examination of criminological theory and research related to the various component s of


10 DPAs, including deterrence, rational choice, organizational theories, procedural justice, and restorative justice, suggests that this practice is therapeutic on several levels. However, t he empirical analysis, which assesses whether DPAs are related to a lower likelihood of subsequent criminal charges than criminal sanctions, finds no relationship between the two types of case disposition and recidivism. Discussion of the limitations of the research and suggestions for future research are provided.


11 CHAPTER 1 INTRODUCTION (Braithwaite, 1984 : 6 ; Simpson, 2002 : 6 7 ). This concept of corporate offending or crime in relation to business was first introduced by Edwin Sutherland (1940) over half a century ago. Sutherland (1945) broadly defined corporate crime as a type of white collar crime that includes any socially injurious behavior committed by corporations and their agents, whether the behavior carries a criminal charge or not. Though criminologists have long debated whether an individual or corporation that has not been criminally charged with violating the law should be classified as a criminal (see Tappan, 1947), the concept of corporate crime today generally encompasses any acts of malfeasance that could be addressed through civil, regulatory, or criminal action. It is also distinguished from other economic or white collar crime by the fact that the direct motivation for corporate offending is organizational rather than individual benef it. Prosecutors tasked with the handling of corporate offenses must weigh and balance a number of complex yet pivotal issues in their determination of the best recourse for such conduct. One of the first issues for consideration revolves around corporate accountability and punishment. The discussion over the rights and liabilities of corporate entities and whether juristic pers ons can be held accountable and punished for criminal acts dates back to the early 1900s (see Canfield, 1914; see also, Coleman 1974). B ecause corporations are juristic persons, they can do many things that natural persons cannot do, including buy and sell each other (Cressey, 1995; Braithwaite and Fisse 1995). On the flip side 1995:414). That said, the age old questions raised regardin g whether a n entity that cannot have a


12 guilty state of mind can be blameworthy ( see Fisse and Braithwaite, 1993) W hile it is now widel y accepted that corporations should be legally treated as persons with legal and ethical responsibilities more severe methods of punishment such as inca rceration, will never be feasible for corporate entities. Some, like Ralph Nader, have long argued for the federal chartering of corporations to make corporate death penalty a more readily available punishment option ( Nader et al. 1976; see also Russell and Gilbert, 1999) However, at this point in time, such an option is not available for the majority of corporations, and thus c be dammed and n o body to be kick ed ( Edward, First Baron Thurlow, 1844 ; see also, Coffee, 1980). Therefore punishing a corporate entity for a crime is significantly different from punishing a natural person. Of equal importance for prosecutorial consideration corpora tions and corporate deviance have a major economic imp act on our society This means that deterrence is of particular importance wit h corporate punishment and that corporate punishments can have far reaching collateral consequences. Despite the challenges of punishing an inanimate entity, b ecause corporations have great influence on the economy and society (Clinard and Yeag e r, 1980; Nader et al., 1976 ; Green, 1973 ), scholars and other members of the general public are quick to criticize the Justice Departm ent for any perceived disparate or lax treatment of corporate tr ansgressions Failure to effectively punish and deter corporate offending has serious economic consequences and is seen as weakening the moral fiber of society by creating an atmosphere of di strust and noxious, profit driven competition (Geis, 1973) And while individual offenders must be held accountable (Fisse and Braithwaite, 1993) it is not enough to punish individuals within the corporation (Nader et al., 1976) As former Deputy Attorne y General Larry Thompson said in The organization itself must be held accountable for the culture and the conduct it


13 promote s Only by prosecuting the corporation itself can we insure systemic reform (Mokhiber, 2005) For these reasons there is no shortage of scholars advocating for the criminal prosecution of corporate offenders (see, for example, Bucy, 2007 ; Russell and Gilbert, 1999 ). On the opposite end of the spectrum, when the Justice Department is overaggressive in seeking corporate indictm ents and handing down harsh sentences, it faces substantial criticism and backlash as a result of the collateral consequences of corporate criminal prosecution. Corporations facing criminal indictment and prosecution are likely to experience a decline in stock market share value, bankruptcy, a weakening of employee morale leading to a mass exodus of employees, an increase in civil charges filed against the company, an overturn of upper management, shortage s of time and resources as both are taken up by def ense and the investigation, a prohibition on receiving government contracts, possible license revocation, and general reputational harm which repels potential business partners and client ( Cooper, 2005; Bohrer and Trencher, 2007 ; Bucy, 2007 ). Thus, crimin al prosecution affects more than just the corporation often h aving a significant, negative impact on the market, innocent employees shareholders, and tax payers. The point is most notably demonstrated by the demise of accounting firm Arthur Anderson fo Cooper, 2005; United States Government Accountability Office, 2009; Robinson et al., 2005). In the aftermath of the case, Arthur Anderson lost a significant portion of their business, saw stock prices nosedive and downsize d from nearly 58,000 employees worldwide to about 200 located in the Illinois offices ). In short, federal prosecutors face the challengin g task of penalizing a corporation in a therapeutic manner which prevents future offending, provides restitution to victims, holds


14 individual offenders accountable, and reforms a corrupted corporate culture, all while simultaneously trying to reduce the ne gative effects of punishment on innocent employees, shareholders, the general public, and the economy. In making the decision regarding how a case should be handled, federal prosecutors have historically had several sentencing options at their disposal an d a number of guidelines from which to base their decision (Christie and Hanna, 2006 ; Ramirez, 2005 ). More recently however, the Department of Justice (DOJ) has steadily encouraged and increased the use of pre trial diversion s such as non prosecution agr eements (NPAs) and deferred prosec ution agreements (DPAs), for accused corporations (Finder and McConnell, 2006; Horowitz and Oliver, 2006; Orland, 2006 ; United States Government Accountability Office, 2009 ). In the nearly two decades since the Departmen t of Justice began to c onsider deferred or non prosecutorial agreements as one of the preferred means of handling corporate offender s there has been a substantial amount of literature produced on the subject, both offering acclamation and criticism DPA s and NPAs are type s of pretrial diversion in which a corporate defendant admits, in a formal signed document, to significant facts of the case and agrees to meet a specified set of conditions, geared towards ensuring that the corporation provides some res titution to victims and takes steps to keep the offense from being committed again, for an agreed upon period of time (Robinson, Urofsky, and Pantel, 2005). Proponents of DPAs 1 argue that they are an effective means of reforming corporate behavio r while r educing the collateral consequences of corporate prosecution ( Christie and Hanna, 2006 ) Opponents, on the other hand, criticize DPAs for being arbitrary and subject to overreaching prosecutorial discretion 1 between a DPA and an NPA is that when a ca se is settled with an NPA the prosecutor has agreed not to criminally prosecute the corporation in exchange for the corporation meeting the terms of the agreement, whereas with a DPA the prosecutor files criminal charges against the corporation but consent s to drop them if the corporation meets the terms of the agreement.


15 ( Garrett, 2007; Martz, 2005) ; for being too easy on corporate offenders and failing to deter corporate crime ( United States Government Accountability Offices, 2009; see also Spivack and Raman, 2008) ; for encouraging gove rnment interference in business; and for pushing the limits of constitutionality (Beh rer and Thrasher, 2007 ) Neither opponents n or proponents, however, tend to examine all aspects of DPAs, systematically weigh their positives against their negatives and assess whether they are a better solution than the traditional criminal, civil, and regulatory punishment options This paper seeks to apply a theoretical perspective which has never been used in the context of corporate crime policy to assess whether DPAs prevent future offending with fewer collateral consequences than traditional cr iminal sanctions. The therapeutic jurisprudence perspective is offered as a new approach to wholly examining DPAs and their therapeutic and anti therapeutic which the focus of puni shment is on restoring both the victim and the offender (Braithwaite, 1989; Makkai and Braithwaite, 1994), the therapeutic jurisprudence perspective advocates that legal decision making should take into account the economic and therapeutic consequences for those being punished as well as any institution or person affected by the particular criminal justice system outcome. Both reintegrative shaming/restorative justice and therapeutic jurisprudence also agree that punishments designed only to stigmatize the offender will not be successful. The therapeutic jurisprudence perspective, however, goes a step beyond reintegrative shaming and spells out a means for assessing whether tools beyond shame can also advance the victim and offender rehabilitation and resto ration process. While therapeutic jurisprudence has typically been applied to such justice related topics as mental health law, drug courts, and juvenile justice, its originators acknowledge that the


16 ider than has been explored to date. As the perspective is applied here, it uses criminological research to examine the extent to which corporate deferred prosecution agreements: 1. promote the rehabilitation and health of the corporation, as well as the well being of the economic and social networks affected by the corporation; 2. attain these ends more successfully than formal sanctions can or do; and 3. are effective without a significant sacrifice to fairness, deterrence, and public safety. The base a rgument from which the research begins is that DPAs are therapeutic agents which rehabilitate the corporation and promote general economic well being. The goal of this article is to assess whether DPAs are actually a more therapeutic approach to punishing corporate offenders than other available criminal sentencing options. Before examining DPAs from the therapeutic jurisprudence perspective, however, the first step is to demonstrate that regular corporate indictment and prosecution has anti therapeutic ef fects. The following section introduce s the problems with formal legal sanctions and the need for an alternative, therapeutic approach to handing corporate offenders. Research on the collateral consequences of prosecution is examined, including the burden on prosecutors to build a successful case, the effects of disrupted business due to prosecution and or corporate death sentences, and the lack of evidence that criminal sanctions deter corporate offending (Coffee, 1981; Geis, 1973; Ramirez, 2005; Simpson, 2002; Simpson and Koper, 1992; Vaughan, 1996). Literature suggesting that corporate and white collar offenders have a greater deterrent susceptibility and a lower threshold for punishment effectiveness are also discussed (Benson and Cullen, 1998; Clinard, 1983; Cullen and Dubeck, 1985; Cullen, Maakested, and Cavender, 1987; Simpson, 2002).


17 Why Formal Sanctions Are Not Effective Against Corporate Offenders Deterrence has most commonly been the meter by which to measure and evaluate corporate crime laws (G eis and DiMento, 2002). As the term is and has been histor ically used, general deterrence focuses on ensuring that punishments are severe enough and consistent enough so as to reinforce the threat of law (Andenaes, 1975). From federal judges to local pros ecutors, there is consensus among members of the courtroom workgroup that general deterrence is the primary goal in the punishment of corporate and white collar offenders (Benson and Cullen, 1998; Mann et al., 1980; Wheeler et al., 1988). P articularly in the case of corporate offenses which individually have a major econo mic impact, specific deterrence, or the suppression of any future criminal activity within the corporation, is also of great importance though M embers of the legal community tend to be lieve th at corporate offenders are more generally and specifically deterrable through formal criminal sanctions than are other offenders (Benson and Cullen, 1998). Both practitioners and criminal justice scholars have pointed out that corporate offenders are rational, future oriented planners (Benson and Cullen, 1998; Cullen et al., 1987; Simpson, 2002). As such, it is assumed that they weigh the costs and benefits of committing an offense prior to doing so (Simon, 1976) Research and experience also sug g est that corporate executives have a particularly strong aversion to criminal punishment because of their high social status and the fact that they have more to lose (Benson and Cullen, 1998; Clinard, 1983; Cullen and Dubeck, 1985; Mann et al., 1980; Nage l and Hagan, 1982 ). Despite the widely held belief that corporate offenders should be more responsive to the threat of punishment, the prevailing sense in corporate crime literature is that, in actuality, corporate offenders are largely undeterred by tradi tional punishment mechanisms (Coffee, 1981; Geis, 1973; The Yale Law Journal, 1979; Ramirez, 2005; Simpson, 2002). Part of the problem,


18 as Australian criminologists, Fisse and Braithwaite, have noted, is that corporate liability often displaces individual accountability (1993:1). However, even just looking at the punishments of the corporate entities, current sanctions do not serve their intended purpose. As Christopher Stone (1975: 93) note d ke into account a whole host of reasons why the threat of legal sanction is apt to lack the desired effects when corporate b ehavior is its target Scholars who argue that typical criminal sanctions will not deter corporate offenders, point out that the t raditional deterrence model is undermined by the structure and culture of many corporations. From a structural perspective, the deterrent effect of punishment is counteracted by the fact that individuals within a corporation are interchangeable and one emp loyee can easily be replaced by another employee who was not involved in the sanctioning experience. Therefore, any deterrent effect that can be achieved through incapacitation and fear of punishment will have only a temporary impact on the corporation. Co rporate decisions also tend to be made by groups which mitigate individual perceptions of risk and sanction costs (The Yale Law Journal, 1979; Simpson, 2002). On top of this, particularly in larger companies, the number of employees and the hierarchical s tructure may mean less oversight and individual accountability for actions her words, The culture of a corporation can also make it impervious to deterrence efforts. Once an employee joins a corporation, he or she learns the norms and practices of the corporation through formal and informal socialization and the behavior of top management (Appelbaum et al., 2005; Hollinger and Clark, 1982, 1983). Within corporations a culture of competition often develops


19 that encourages employees to strive to get ahead at any cost (Coleman, 1995). When this competition is combined with business pressure to reduce costs and maximize profits (Welles, 199 8) as well as perceptions of managerial leniency (Hollinger and Clark, 1983) the result is a corp orate culture in which offending behavior is normal and rationalized to the point that it may stop being considered illegal by those overseeing the offense (Simpson, 2002; Vaughan, 1996). In addition, because corporate offenders cannot be incarcerated and are relatively unaffected by non monetary deterrents, such as shame and loss of social status (Coffee, 1981), fines are used to punish corporate offenders. According to rational choice principles, in order for fines to deter, they must present more cost to the company than could potentially be gained from the criminal act (Cullen, Maakestad, and Cavender, 1987). As John Coffee (1981) explains, though, this is often an impossible notion. If a corporation cannot afford to pay a fine beyond a particular thr eshold, say $5 million, increasing the amount of the fine beyond that point will have no added deterrent value because in either scenario the corporation will pay only what it can. e justice system is essentially unable to set a fine amount which will cost the corporation more than the expected gains from the offense (Coffee, 1981). In most cases, though, fine amounts fall well below the maximum amount a company could be expected to pay. Fines in antitrust cases have historically been particularly low. General Electric was famously slapped with a fine of less than $500,000 for its role in the 1961 heavy electrical equipment antitrust conspiracy. As Gilbert Geis pointed out, this fi ne was the equivalent of giving a $3 parking ticket to a person making $175,000 a year (Geis, 2002). Mark Cohen also examined the fines handed down to 288 non antitrust corporate offenders in U.S. District Courts in the mid to late 1980s and found that t he average fine was about $54,000


20 (Cohen, 1989). Data show that average fine amounts increased after the sentencing guidelines went into effect (Simpson, 2002), but still equate to a minor slap on the wrist in many instances. those corporations that elect to adhere to regulations and statutory prohibitions are arguably at a competitive disadvantage relative to those businesses that do not follow the law but, instead, gamble on the likelihood that they either will not be caught or case of the late 1970s the prosecution revealed evidence that Ford consciously chose the monetary savings from leaving a potentially harmful part in the Pinto over the estimated cost that could result from any ensuing litigation (Swigert and Farrell, 1980 81). Essentially, in this case and in others, the decision to engage in criminal behavior is no different from any other business decision in which economic g ains are pitted against potential losses (Cullen, Maakestad, and Cavender, 1987). With any potential deterrent effect of formal sanctions competing against a strong drive t o get ahead at any cost, l enient punishments stand little chance against the pull o f corporate offending. While there is some debate over the extent to which punishment certainty and celerity are related to deterrence (Nagin and Pogarsky, 2001; Parker and Grasmick, 1979; Paternoster and Piquero, 1995; Pratt and Cullen, 2005; Pratt et al., 2006; Yu, 1994), corporate offending can often go on for years without detection and when it is discovered the investigations are lengthy and drawn out. Though scholars are unable to even estimate the amount of corporate crime that goes undetected, it has been assumed since Sutherland first began discussing corporate and white collar crime that the likelihood of discovering corporate offenses is even lower than it is for conventional crimes (Clinard and Yeager, 1980; Simpson, 2002; Sutherland, 1949). C orporate criminal events are often multifaceted, occurring over an extended period, and


21 involving multiple actors. Furthermore, the government agencies tasked with policing these corporations are often strapped for money and resources and may not be adequ ately trained to filter through the complex cover ups put in place by the corporation (Simpson, 2002; Stotland, 1982; Yeager, 1991). Criminal cases against corporate offenders also take a long time to dispose. Because of the complex nature of the crime s, the investigation process alone often requires extensive time and effort. Simpson (2002) notes that the corporate crime cases that are actually prosecuted by the Justice Department tend to be among the least complex, due in part to the to the limited r esources of the Department for investigating and putting together a solid case against the bigger, more complex offenses (see also Saltzburg, 1991). Corporate attorneys often try to drag out the disposition process in the hopes that the prosecution will b ecome too costly to continue (Simpson, 2002). If the case actually reaches the trial stages, attorneys on both sides struggle to empanel a jury that will be able to comprehend the complexities of the case. The empirical studies that have examined recidivi sm among convicted corporate offenders tend to validate the deterrence theory perspective that criminal sanctions will not deter corporations. Simpson and Koper (1992), conducted a study of recidivism among 38 corporations accused of antitrust offenses and found that, compared to corporations handled through civil or regulatory routes, those sanctioned criminally had an increased likelihood of reoffending. Braithwaite and Makkai (1991), in a study of offending behavior among nursing homes, found that neith er the perceived certainty nor severity of punishment had significant effects on legal compliance. Beyond the evidence that current legal sanctions may not deter corporate offenders, another issue with criminal sanctions for corporations is the potential impact on those not


22 engaged in or benefiting from the offense. problem of corporate punishment seems perversely insoluble: moderate fines do not deter, while severe punishments flow through the corporate she (Coffee, 1980: 386 387). At a minimum, the process of prosecution is likely to disrupt business operations. In more severe instances, prosecution may indefinitely prevent the corporation from being involved with c ertain business clients or operations or may cause the corporation to go under entirely (Benson and Cullen, 1998; Wheeler, 1981). Thus, the punishment falls sharply on shareholders and the innocent employees who lose their jobs. In other words, the tradit ional solution to corporate offending that involves a focus on deterrence through severe sanctioning has been at best ineffective and at worst socially injurious. C riminal sanctions have proven to be anti therapeutic in their inability to reform the corpor ate culture and prevent future offending and in their associated collateral consequences. The remainder of the paper is geared towards utilizing the therapeutic jurisprudence deferred prosecutorial agreements are a more therapeutic punishment option for corporate offenders. Overview of the Examination of DPAs through the Therapeutic Jurisprudence Lens Taking a therapeutic jurisprudence approach to examining a criminal justice practice, policy or procedure involves a two stage methodology that uses empirical research in the field to fully assess the therapeutic or anti therapeutic nature of the par ticular topic or issue. The first step the non empirical analysis, involves con duct ing a n in depth review of relevant research related to the topic with an eye towards developing hypotheses about the expected therapeuti c or anti therapeutic effects of various aspects of the policy or procedure T he empi rical step to the analysis th en involves testing a research based measure of therapeutic outcome using a model


23 with independent variables derived from various aspects of the po licy or procedure examined in the non empirical analysis For this study four major components or clauses o f DPAs a re examined using criminological literature and theory to develop hypotheses about their therapeutic or anti therapeutic nature : the requirement of voluntary disclosure and cooperation; the introduction of a compliance monitor; monetary fines and r estitution; and the establishment of business reforms and /or compliance programs For the non empirical analysis, which assists in developing hypotheses about the therapeutic or anti therapeutic implications of DPAs, several principal bodies of literature from the criminological and legal fields will be drawn upon. The major bodies of literature to be referenced are those pertaining to deterrence theory and preventive effects (Geis and DiMento, 2002), organizational theories of corporate offending (Baysinge r, 1991; Coleman, J.S., 2002; Ermann and Lundman, 2002; Simpson, 2002; Wheeler and Rothman, 1982), reintegrative shaming and restorative justice ( Braithwaite, 1989; Makkai and Braithwaite, 1994 ), opportunity and motivations for offending (Colman, J.W., 199 5), procedural justice, rational choice (Paternoster and Simpson, Ramirez, 2005; Swigert and Farrell, 1980 81), and recidivism (Clinard and Yeager, 2005) Each of these bodies of literature contributes in important ways to developing the distinction betwee n a therapeutic and an anti therapeutic punishment. Literature on deterrence theory suggests that in order for DPAs to be therapeutic agents the y must prevent future offending (Benson and Cullen, 1998; Mann et al., 1980; Wheeler et al., 1988). Organizatio nal theory is important because it suggests that the noxious environment within a corporation that can lead to offending must be improved before the corporation is truly healt hy and rehabilitated (Simpson, 2002 ; Coleman, J.S., 2002). Procedural justice li terature indicates that for DPAs to be


24 therapeutic they must also have a positive effect on the legitimacy afforded to the criminal justice system. Rational choice literature suggests that any punishment must outweigh the benefits of the crime in order to be effective and thus therapeutic. Restorative justice provides evidence that in order for DPAs to be a therapeutic punishment they must afford offenders a degree of decency and respect in the punishment process (Braithwaite, 1989). Finally, work by Nade r and colleagues on federal incorporation is a reminder that a number of scholars and politicians have long contended that federal involvement in corporations would be beneficial to the health and welfare of society. The empirical analysis was conducted u sing a database containing variables derived from the 143 public ly available DPAs and NPAs entered into by US Attorneys and vario us corporate offenders from 1993 through 20 09 as well as any cr iminal action taken against the corporation proceeding or follo wing the signing of the DPA. The database also contains information regarding the charges, sentencing, and any prior and subsequent criminal sanctions of a matched sample of 128 corporations charged criminally during the same period. Building on the hypo theses developed through the non empirical analysis, the empirical analysis addresses two main questions: 1. Among corporations that enter into DPAs, whi ch of the components of the DPA contribute most to the probability that the corporation will not recid ivate; and 2. Are corporations that enter into DPAs less likely to recidivate than corporations that receive criminal sanctions? In other words, the first part of the empirical analysis tests whether each of the components is therapeutic in that they cont ribute positively to the likelihood that a corporation will not recidivate The second part of the empirical analyses then examines the relative effectiveness and therapeutic nature of DPAs versus traditional criminal sanctions.


25 The remainder of Chapter 1 provide s an overview of the various components of DPAs that will be examined from the therapeutic jurisprudence perspective and tested for their therapeutic or anti therapeutic effects using t he previously noted bodies of legal and criminological researc h. Voluntary D isclosure an d C ooperation One of the central elements of DPAs is the focus on cooperation between the federal government and th e corporation during the investigation Coope ration is not unique to DPAs and has always been a factor for conside ration in general f ederal charging decisions (Fe deral Sentencing Guidelines, 2010 ) as well as Securities and Exchange Commission (SEC) enforcement decisions (Securities and Exchange Commission, 2001). However, it became an integral component of corporate charging policies with the Thompson Memo in 2003 (Behrer and Thrasher, 2007) In the cover letter to his memo Principles of Federal Prosecution of Business Organizations the then revision s [to the Holder memo] is increased emphasis on and scrutiny of the authenticity of a a Department investigation, in fact take steps to impede the quick and ef fective exposure of the : 1 ). In an effort to ensure timely and voluntary disclosure of wrongdoi ng, willingness to waive attorney client and work product privileges in order to assist in the investigation, and efforts to encourage individual employees to work with and provide information to prosecutors (Bucy, 2007 ; Duggin, 2008 ). From one perspective then, cooperation is intended to reduce the investigation time and effort required of federal prosecutors, leading to more swift justice. Thus, literature on deterrence theory and the relationship between deterrence and punishment celerity will be exami ned (see, for example, Nagin and Pogarsky, 2004) to assess the therapeutic value of swift


26 case resolution. Literature in the area of procedural justice has suggested a correlation between cooperation and the legitimacy afforded to the criminal justice sys tem and deterrence (see Fagan and Tyler, 2004), and this research will be explored as well. Compliance M onitor Closely tied to the concept of cooperation between the federal government and the cor poration is t he requirement that the corporation take on a n independent compliance monitor either appointed or approved by the Justice Department. The compliance monitor is akin in theory to a pretrial services officer (PSO) or a probation officer, tasked with ensuring that the corporation fulfills their end of t he agreement, remains in good favor, and makes strides towards becoming an active, contributing part of society (Robinson, et al., 2005). However, unlike a PSO or probation officer who is assigned to multiple defendants, the ratio of compliance monitors t o corporation s is at least one to one. Individual and day to day involvement of the corporate compliance monitor is both touted as an effective mechanism for instituting corporate change from the inside (Morford, 2008), as well as criticized as the govern ment taking an Thus, in order to assess the therapeutic or anti therapeutic consequences of the compliance monitor in DPAs, criminological literature on PSOs and probation officers will be explored (for example, Kingsnorth, et al. 1999) as will lite rature on organizational theory and the environment of corruption that can be attributed to corporate offending (Baysinger, 1991; Coleman, 1995; Simpson and Koper, 1992; Wheeler and Rothman, 198 2) Legal and business related literature on the effects of government intervention on corporate affairs will also be examined. Monetary P enalties and R estitution Many DPAs have monetary penalties or restitution payments associated with them and often ti mes these amounts are larger than would be imposed on a corporation upon conviction


27 (McPhee, 2006) In the evaluation of the therapeutic or anti therapeutic outcome of high monetary penalties and r estitution, several major bodies of literature will be exa mined. The first is literature on the relationship between fines and deterrence. Some of the research in this area suggests that increasing the severity of punishment, specifically fines has a deterrent effect ( Yu 1994 ; Staltzburg, 1991; Wood and Grasm ick, 1999 ) while other research find that fines do not have a significant impact on reoffending (Moffatt and Poynton, 2007) Next rational choice literature, which suggests that offenders weigh the costs and benefits of offending (Lee and Ermann, 2002; Sw igert and Farrell, 1980 81 ), will be examined. Finally, literature on restitution will be examined. Business R eforms and Compliance P rograms Another commonality among many DPAs is the requirement that the corporation develops or revise s an existing cor porate compliance and ethics program Along the same lines, DPAs may also require that the corporation institute other types of business reforms, such as restruct uring the corporate governance or bringing in new executive leadership (McPhee, 2006) The go al of these requirement s is to chang e the corporate environment, alter practices that promote offending and improve employee accountability, without the government having to micro manage or become too involved in the daily corporate bu siness (Christie and Hanna, 2006 ). Here again, the literature on organizational theory and how the environment within the corporation is related to the propensity for offending will be examined (Baysinger, 1991; Coleman, 1995; Simpson and Koper, 1992; Wheeler and Rothman, 19 82). Additionally, literature on the effect of compliance programs on employee behavior will also be addressed ( Greenburg and Barling, 1996 ).


28 CHAPTER 2 THERAPEUTIC JURISPRU DENCE Review of the Literature The therapeutic jurisprudence perspective was first introduced by Wexler and Wi nick in the 1990s as a means of analyzing mental health law. The theory is founded on the idea that legal decision making should take into account economic consequences, public safety, and the therapeutic implications of legal rules and their alternatives for those coming into contact with or affected by the criminal justice system. It uses social and behavioral science to study the extent to which policies, legal procedures, and the courtroom workgroup produce therapeutic or a nti therapeutic consequences. Early on Wexler (1995) and others (see Slobogin, 1995) stressed that therapeutic jurisprudence should never be just about achieving desired outcomes in general but must maintain its distinctive focus on the psychological an d physical well being of those impacted by the law. However, from the foundations and subsequent developments of therapeutic Thus, the theory can be more generally conceptualized as a tool for examining whether one legal policy or its alternatives is better able to promote the health and rehabilitation of offenders, victims, and society at large. Although it originally began wi th a very specific focus on mental health law, Wexler and Winick encouraged broader applications, and it has since been applied to a range of areas (see Wexler and Winick, 1996) including domestic violence (Simon, 1996), sex offenders (Birgden, 2000; 2007) appeals (Ronner, 2000; Wexler, 2000), family law, and drug treatment courts (Senjo and Leip, 2001). Moreover, it has application both for examination of whether existing laws have therapeutic or anti therapeutic


29 consequences and for examination of how e xisting laws could be applied in a more therapeutic manner (Wexler, 1996). The concept of therapeutic jurisprudence arose from a study conducted in 1989 that looked at the relationship between low levels of serotonin and blood glucose of criminal offend ers and their likelihood of recidivism (Wexler, 1990). The study found that parolees with low levels of serotonin were at high risk for recidivism when alcohol consumption was involved. It was suggested that for these offenders abstinence from alcohol sh ould be a condition of parole, 1990: 44). In other words, therapeutic jurisprudence theory is founded on the idea of using empirical research to assess how legal pol icies and practices can better rehabilitate offenders. Wexler (1990) then built off this argument and presented the idea that if acknowledged violence prone individuals fail to take the medication that prevents them from becoming violent, they should be punished for reckless endangerment. The threat of punishment, according to deterrence principles, should prevent these violence prone individuals from not taking their medication. Thus, such a punishment policy will have a therapeutic impact by helping p otential offenders not to engage in violence while simultaneously maintaining the health and safety of potential victims. In their further development of therapeutic jurisprudence, Wexler and Winick (1991) laid out the formula for how the theory should be used for conducting policy analysis. First, there is a non empirical element to therapeutic jurisprudence. In this stage a researcher conducts an in depth review of relevant social and behavioral science literature and research to develop hypotheses regarding the expected therapeutic or anti therapeutic consequences of a legal policy.


30 The aim is to relate a body of literature to a body of law and examine how well the law matches the approach that the extant research suggests will have therapeutic res ults. After this stage is complete, there is also an empirical dimension to the application of the theory. In the empirical analysis, various aspects of the particular law or policy become independent variables in a model testing a research based measu re of therapeutic outcome. One of the more important parts of the empirical analysis, then, is determining the most appropriate measure of therapeutic outcome. Although Wexler and Winick originally related therapeutic jurisprudence to policies regardi inmates or other criminal defendants found incompetent to be executed or stand trial, from their earliest statements about the theory, they note that the relevance of the concerns the therapeutic implications of various legal rules and practices. Is a particular legal rule, either presently in effect or proposed, therapeutic or anti therapeutic to patients (and Winick acknowledged from the beginning that there is a general societal benefit in looking at the the rapeutic implications of laws and the legal process (see also Wexler, 1995). They went on to explicitly point out that therapeutic jurisprudence should also eventually be adopted for the legal (Wexler and Winick, 1991:982). Following a number applications of therapeutic jurisprudence to tort law (Shuman, 1992), contract law (Harrison, 1994), and corrections (Cohen and Dvoskin, 1993), Slobogin (1995) further extended and clarified therapeutic jurisprudence theory. He identified five issues or


31 dilemmas that therapeutic jurisprudence theory needed to address in order to have sustained viability: the identity dilemma, the definitional dilemma, the dilemma of emp irical indeterminism, the rule of law dilemma, and the balancing dilemma (Slobogin 1995: 195). Each delineation, and according to Slobogin can be resolved through a mor e precise explanation of the scope and application of the theory. First, in order to ensure that theoretical jurisprudence remains distinct from any other jurisprudence or legal realism perspective, Slobogin proposed ficial in the sense of improving the psychological or physical well psychological a nd physical well Slobogin also points out that tests of therapeutic jurisprudence must take into consideration the fact that what is defined as therapeutic from one perspective may not be how therapeutic is d efined from another perspective; that because the theory relies heavily on social science data it must be understood that all conclusions are subject to some degree of vagary and bias; that what has therapeutic results for the majority of the population ma y not have the same effect for a particular individual; and that researchers suggesting a more therapeutic approach to a particular legal process or policy must determine the appropriate level of consideration to give to general principles of justice, comm unity safety, and economic cost as well (see Roderick and Krumholz, 2006 for similar concerns regarding therapeutic jurisprudence theory). Slobogin is careful to point out that although these issues are unresolved aspects of therapeutic jurisprudence tha t should be weighed in research, they do not significantly detract from the overall value of the


32 have also done their part to provide specification For example, to the argument that the law and legal procedures serve a number of ends among which therapeutic outcomes should not necessarily be given priority, Winick explains that his n of a therapeutic or anti therapeutic outcome does not have to be an argument for legal change and that the theory allows for the possibility that other outcomes of a law or procedure outweigh therapeutic effects. Of particular relevance to the current research, in the process of clarifying the scope of the theory and noting that it will not have bearing for all legal areas, Wexler (1995) expressly provides some encouragement for the application of therapeutic jurisprudence to white collar and corporate crime. Wexler states, Perhaps therapeutic consequences will be particularly difficult to find in areas such as antitrust law, securities law, or water law. On the other hand, if such consequences are actually found in those areas, those issues may be ext remely exciting to therapeutic jurisprudence scholars. Once one starts approaching legal areas with the use of the therapeutic jurisprudence lens, who can tell what one will find? (1995: 228). It is clear from this statement that as long as laws and polic ies can be shown to have consequences for the psychological and physical well being of those affected, white collar and corporate crime is not beyond the realm of therapeutic jurisprudence. This idea is not unique to Dr. Wexler either. A speech made by f ormer US Deputy Attorney General James Comey in 2004, further suggests that prosecutors also consider the health of the corporation in making What we try to focus on is is this [the corporat ion] an entity that is recidivist or that is so sick its culture is so sick that it's


33 going to re offend, that it has to be put down and that we have to suffer the collateral (Mokhiber, 2005). Thus, taking i nto account the wide range of those impacted by the handling of corporate offenders from victims, to employees, to shareholders, to tax payers combined with the knowledge that prosecutors have loosely taken tenants of the perspective into consideration f or years, it seems that examination of corporate punishment policies from the therapeutic jurisprudence lens is both informative and prudent. might be involved with or affected by the law, the focus has largely been on the individual. Of course there is no need for the focus to be so constrained; it could relate to the family, the e that a corporation is not an individual whose well being can be considered (despite the fact that corporations are treated as legal persons by the justice system), it should be clear that therapeutic outcomes of the punishment of corporate offenders can appropriately be explored from the vantage point of those who are affected by such punishment. The Application of Therapeutic Jurisprudence in Extant Research Although the therapeutic jurisprudence perspective has been in existence for only about two de cades a bibliography put together by the International Network on Therapeutic Jurisprudence contains over 1,800 references to articles that utilize the perspective ( n.d.) A sizable portion of these articles explore such topics as mental health law, juvenile justice and mental health, the insanity defense (Perlin, 1994; Wexler, 1991) civil commitment (Tyler, 1992), appellate court processes, family law, sex offender s, clinical and legal education, and guardianship (see intj/ for the complete bibliography). While these contribute significantly to the development and understandin g of therapeutic jurisprudence, the research addressing criminal case processing and


34 alternative means for handling and rehabilitating criminal offenders is most pertinent to the current theoretical application. This research ranges from very broad applic ation to very specific empirical tests. Jeffries (2005), for one, used therapeutic jurisprudence theory in a broad manner to examine changes taking place in Australian criminal court policies and procedures. Utilizing this perspective, she highlighted the positive and negative expected effects of such changes and identified areas in which further research was needed to determine the implications of these changes. Jeffries specifically noted that there has been a juxtaposition of changes in Australian c riminal courts as political figures have faced increased pressure to reduce rising crime rates by becoming tough on crime, while the privatization of criminal justice systems functions has lead to a demand for more effective and economical services. Furth ermore, there has been a general societal push to better take into consideration the well being of those processed through the criminal justice system. She explained that the therapeutic jurisprudence movement, which led to the development of various type s of problem solving courts and forums for restorative justice, from an adversarial model of justice to a problem solving system. However, after reviewing the re search on problem solving courts and restorative justice, she determined that both could have anti therapeutic effects and that more research, driven by the therapeutic jurisprudence perspective, would be necessary for a complete assessment. A number o f scholars have also recognized that therapeutic jurisprudence can be used to examine whether a law or procedure can be applied, particularly by lawyers, judges, and other members of the courtroom workgroup, in a manner which would increase its therapeutic impact (Casey and Rottman, 2000; Stolle and Wexler, 1997; Stolle et al., 1997; Wexler, 1996; Winick,


35 1997). Those accused of an offense, for example, generally meet with an attorney, and these attorneys, in turn, can incorporate therapeutic jurisprudence principles by being cognizant of ways to improve the psychological well being of clients though attorney client interactions. Along these lines, Stolle and Wexler (1997) and Stolle and colleagues (1997) examined the and therapeutic jurisprudence. Both sets of authors noted that through the use of legal check ups, in which the attorney continues to touch base with the client after the initial situation is resolved, an attorney may be able to help the individual to avo id future legal problems before they occur. Thus, the attorney would be approaching a routine attorney client interaction in a therapeutic manner. Beyond applying therapeutic jurisprudence to general criminal court policies and procedures and manners o f implementation, the theory has also been used to examine alternative approaches to dealing with criminal offenders. One of the earliest and more common such areas in which therapeutic jurisprudence principles have been applied to alternative forms of ha ndling criminal offenders is in studying drug treatment courts (see, for example, Clark, 2001; Hora et al, 1999; Senjo and Leip, 2001; Skinner, 2006). Hora and colleagues (1999) initially saw therapeutic jurisprudence as a means for bringing a much needed academic perspective to the drug treatment court movement to assist with analyzing the effectiveness of various aspect of the [drug treatment court] movement represents a significant step in the evo lution of therapeutic jurisprudence the evolutionary step inadvertently been applying therapeutic jurisprudence principles by attempting to instill procedures th at would deter, incapacitate, and rehabilitate. Their non empirical analysis effectively demonstrated additional ways in which therapeutic jurisprudence could be used to


36 inform debates over issues such as the heightened requirements of DTCs over tradition al probation, the effectiveness of legally coerced treatment, the waiving of participants legal rights, Clark (2001) also relied on existing behavioral science research to shape his discussion of a strategy for improving the effectiveness of drug treatment courts as a catalyst for positive behavioral change. After conducting a meta analysis of therapy outcome studies, he identified four factors common to all effective models of treat ment. These four factors were then used to develop a research informed strategy for adapting a more therapeutic approach to handling drug court participants. Most importantly, Clark stressed that the approach was not geared towards treatment professional s, but rather intended to be put into action by court personnel, judges, attorneys, and probation officers. In other words, he used empirical research to develop simple ways in which criminal justice practitioners could actively improve the therapeutic ou tcome of drug treatment courts. Senjo and Leip (2001a, 2001b) advanced the application of therapeutic jurisprudence to drug treatment courts by empirically testing a theoretically designed model for its therapeutic and anti therapeutic effects. Specifi cally, the authors used a cohort of offenders from one drug court in Florida to test the relationship between court monitoring, drug treatment, and court participati on. Based on their non empirical analysis of extant research, the authors hypothesized that: 1. supportive rather than adversarial court monitoring should produce therapeutic results, 2. that the administration of multiple treatment practices should produ ce therapeutic results, and 3. that immediacy of treatment following arrest is also related to offender rehabilitation.


37 The dependent variable for their empirical analysis was the ratio of u r inalysis tests passed to tests taken. Data on the first indep endent variable, supportive court monitoring practices, was probation officers during status hearings. Data on the number of treatment sessions that an offen The authors also controlled for demographic variables, age, gender race, and education level, and for the type of charge and the length of time the offender was in the program. The authors found support for both the supportive court monitoring and increased treatment hypotheses. Moreover, the model including both of these variables with the demographic and case specific variables suggested that the type of court monitoring and variations in the number of treatments best accounted for whether the offender experienced a therapeutic or anti therapeutic outcome. In othe r words, the authors were able to use therapeutic provide a model for using therapeutic jurisprudence theory to analyze the effectiveness of justice system pol icies and procedure for producing therapeutic results. The current research follows from this model. Next, attention is turned to providing background on the particular justice sys tem policy to which the therapeutic jurisprudence perspective will be appli ed, deferred prosecution agreements.


38 CHAPTER 3 THE DEVELOPMENT AND CURRENT APPLICATION OF DEFERRED PROSECUT ION AND NON PROSECUTION AGREEMEN TS It has long been noted that corporations offend. More than 60 years ago Sutherland first detailed the transgre ssions of the 70 largest corporations at the time. He found that each of the corporations had at least one violation on record, with a total of 980 violations across all 70 companies (Sutherland, 1949). These violations included charges such as infringemen t, restraint of trade, false advertising, and unfair labor practices. About 30 years later Marshall Clinard and Peter Yeager (1980) likewise found that three fifths of the Fortune 500 companies they examined had at least one regulatory action taken again st them, and 42% had multiple actions against them in a one year period. of corporate offending the manner by which the Department of Justice (DOJ) has opted to handle corporate crime has shifted dra matical ly over time; from very little legal action in the 1970s (Clinard and Yeager, 1980) to increased prosecution in the 1980s following the savings and loan crisis (Ramirez, 2005), to the development of the organizational sentencing guidelines but less attent ion paid to corporate malfeasance in the 1990s then to the full force crackdown following the corporate financial scandals of the early 2000s. The new trend over the past decade has been for the DOJ to steadily encourage and increase the use of pre tri al diversion, such as non prosecutorial agreements (NPAs) and deferred prosecutorial agreements (DPAs), for accused corporations (Finder and McConnell, 2006; Horowitz and Oliver, 2006 ; Orland 2006 ). As former US Attorney s Christie and Hanna descri be the use of DPAs and NPAs it appears as though the DOJ may have finally settled on a compromise solution that works: [D]eferred prosecution agreements allow prosecutors and companies to


39 work together in creative and flexible ways to remedy past problems and se t the corporation on the road of good corporate citizenship. They also permit us [the Department of Justice ] to achieve more than we could through court imposed fines or restitution alone. These agreements, with their broad range of reform tools permit rem edies beyond the scope of what a court could achieve after a criminal conviction ( 2008: 1043) Thus, it is no surprise that from 1992 to 2006, DOJ disposed of at least 44 corporate offending cases via DPAs and NPAs (Orland, 2006). In 2007 alone, as corpo rate diversion programs became better understood and further promoted, estimates suggest t here were at least 37 publicly known DPAs and NPAs entered into by the DOJ (Spivack and Raman, 2008) Chapter 3 examines the development, theoretical underpinnings, a nd current application of Developing Guidelines for Charging and Sentencing Corporations At the basic level, American jurisprudence and the Moral Penal Code (MPC) of the American Law Institute provide two slightly differ ing standards for determining corporate criminal liability. Under the doctrine of respondent superior a corporation is liable for any employee, agent, director, or officer, who commits an offense with the goal of benefiting the corporation during the term of their employment (Finder and McConnell, 2006). The MPC, on the other hand, specifies that a corporation is only liable for the upper echelon of employees (i.e., managers, directors, and officers). Each of these standards have faced criticism for thei r inability to fully take criminal intent into consideration, and f or, respectively, too broadly and too narrowly imposing liability. US Sentencing Commission Organizational Guidelines One of the first attempts at providing more detailed and standardize d guidance for the prosecutors and courts responsible for dealing with corporate offenders came from the United States Sent encing Commission (USSC) In 1991, t he USSC introduced Chapter Eight,


40 colloquially known a s the Organi zational Guidelines as a supplement to the Federal Sentencing Guidelines. This document essentially presented the court with criteria similar to that which had existed for individuals since 1987, for determining how a bus iness entity should be sentenced Its most notable feature was its emphasis on cooperation with the federal government as a condition for leniency and reduced fines More specifically, these guidelines noted that if the corporation demonstrated an imm ediate willingness to assist the go vernment by waiving legal and work order protections so as not to impede the investigation, providing full disclosure to US and DOJ attorneys, and conduct ing all internal investigations in a timely manner, it would be a candidate for a lighter sentence. Another important component of the USSC organizational sentencing guidelines was the forcing of corporations to have or to develop internal compliance programs in order to receive mitigating credits for sentence reduction (Orland, 2006) As Paula Desio, Deputy General Council of the US Sentencing Commission explained, hen the [US Sentencing] Commission promulgated the organizational guidelines, it attempted to alleviate the harshest aspects of this institutional vulnerability by incorporati ng into the sentencing structure the preventive and deterrent aspects of systematic compliance programs The language of Chapter 8 is also such that any convicted business entity with 50 or more employees and no compliance program is to be p ut on probation until such a program is developed and implemented. As terms of the probation, the corporation would additionally be subject to outside monitoring to ensure that the conditions of probation were being met and that the corporation was engagi ng in ethical and legitimate budgeting, record keeping and reporting practices. In essence, as Finder and treatment, the Organizational Guidelines laid the groundwork for the explicit DOJ prosecutorial


41 policy that considered both the impact of cooperation and a compliance monitor in corporate The Holder Memorandum In 1999, then O rganizational Guidelines to prosecutorial decisions in a memorandum to DOJ a ttorneys titled Against conceived as a set of guidelines for determining whether a corporation should be formally charged in a particular case. The idea behind the Holder Memo, as it is commonly referred, was that, due to their unique form, corporations should not be treated the same as individuals but still had to be held accountable (Finder and McConnell, 2006). The Holder Memo noted that the typical considerations involved in sound prosecutorial decisions should be present in corporate cha rging decision as well However, d ue to the unique nature of corporations seve ral additional issues should als o be regarded as important in the handling of accused corporate offenders E ight factors were laid out as additional items for consideration. These eight factors in brief include : the nature and seriousness of the offense; the pervasiveness of wrongdoing within the corporation, with special attention paid to the role of upper management; the corporate history of similar misconduct; the e of wrong doing and willingness to cooperate in criminal investigations of affiliated individuals; the existence and adequacy of a corporate compliance program; the company most directly responsible;


42 the collateral consequences of formal prosecution; the adequacy of non criminal enforcement of a regulatory or civil nature. After laying out the eight factors, the memo went on to further detail how each factor specifica lly In general, the Holder memo highlighted the importance of corporate cooperation in charging decisions. willingness on the part of the corporate authorities to assist the government in its investigation through complete disclosure and a waiving of work order and attorney client protections when required. It also set the stage f or the later DPA provision that an independent monitor is necessary to truly evaluate whether any existing corporate compliance program is up to par and has been operating effectively. Of equal importance for the current assessment, however, is that the Holder memo gives credence to a focus on restitution, remediation, refor m and avoiding collateral consequences for third parties to the greatest degree possible. As will be discussed later in greater detail, with these element s of focus, the Holder memo laid the foundation for a charging policy that conforms to the fundament als of therapeutic jurisprudence that legal decision making should take into account the psychological and physical well being of those coming into contact with or impacted by the law or policy. While the Holder memo did not directly address pre trial diversion, it did suggest that prosecutors could use the pre trial diversion procedures detailed in the US Attorney Manual as an incentive to encourage corporate cooperation. The pre trial diversion section of the Manual was geared towards the prosecutio n of individuals rather than corporate entities but the general ideas and procedures were still relevant in a corporate context. Echoing the goals of therapeutic jurisprudence, the objectives of pre trial diversion were essentially to deter offenders in a manner which would preserve prosecutorial and judicial resources, provide restitution to victims, and


43 allow the corporate entities to avoid the collateral consequences associated with formal indictment (see Finder and McConnell, 2006). From 1999 to 2003 the time period in which the Holder Memo provided the key source of guidance for US Attorneys handling cases of corporate offending the Department of Justice entered into a total of four corporate pre trial agreements. Though elements of the Holder Me mo are currently visible in the guidelines that US Attorneys follow when entering into deferred prosecutorial agreements with corporate offenders, in 1999 the Holder memo received comparably little attention. Then, from the time the memo was released unti l the time its successor, the Thompson memo, was drafted in 2003, a series of major events in the corporate and federal prosecution worlds elevated the attention given to Corporate Offending at the Turn o f the C entury I n the first years of the 21 st century several of major incidents of corporate offending raised economic and social concerns about the ability of corporations to seemingly operate above the law in a relatively unchecked fashion. The financ ial account fraud boom began with Enron in October of 2001 when the company resulting in losses of over $600 million in one quarter and a $1.2 billion reduction in shareholder equity (Brickey, 2003) Less than a year after the fall of Enron, telecom giant, WorldCom, announced that nearly $4 billion in expenses had been improperly reported as investments and the company declared bankruptcy shortly after, leaving shareholders and a substantial portion o f In response to the impact on Point Plan to Improve Corporate Responsibility and Protect American Shareholde spring of 2002. The plan was focused on increasing corporate accountability and financial transparency


44 and estab lishing auditor independence. T wo major strategies for implementation were put into place: first, President Bush called on Congress to pass Sarbanes O xley legislation, and second, he established the Corporate Fraud Task Force through Executive Order 13271. Sarbanes Oxley The Public Company Accounting Reform and Investor Protection Act of 2002, also known as the Sarbanes Oxley Act (SO X), was passed in July of 2002 The original intent of SOX was to increase accountability and oversight requirements for publicly traded companies and to provide enhanced sentences for those companies which failed to comply or engaged in a range of financ ial crimes (Coates, 2007) T he Act established the non governmental Public Company Accounting Oversight Board to oversee audits of all public companies subject to securities laws It established more stringent guidelines to ensure auditor independence, re quired companies to submit regular, detailed financial reports, and criminalized the falsification of financial reports and the failure to maintai n appropriate audit records. M ost importantly for this review, SOX mandated that the Sentencing Commission re visit and increase economic crime sentences for all offenders charged with fraud (Bowman, 2003; Corporate Fraud Task Force, 2003). Almost immediately, the Justice Department proposed across the board sentence increases for corporate crimes and threatened to impose additional increases if the original increases were not approved by the Commission (Bowman, 2003). Arguably SOX fell short of being the regulatory pow erhouse it was intended to be. Critics have suggested that the Act was drafted hurriedly, wit hout regard to existing research on the implications of the changes, and under frenzied political and economic conditions (Perino, 2002; Romano, 2005). Additional criticisms provide an important contrast to the later terms and conditions of DPAs, which in many ways, account for the problems of SOX. For starters, t h e lan guage of Sarbanes Oxley refers to individual offenders rather than corporate entities, so the


45 sentencing increases are geared more towa rds corporate executives and do not address the handli ng of corporations themselves (Orland, 2006). T he Act also focuses on increasing penalties for corporate offenses without real regard for the difficulties of proving criminal intent in the first place and the potential need for enhanced investigative and prosecutorial resources in order to build a solid case (Bowman, 2003; Perino, 2002). On top of this even if a corporation were to be sentenced and rece ive the enhanced penalties, economic theories of deterrence raise questions as to whether such increase s actually have a deterrent impact (Perino, 2002). Finally, other than the questionably effective changes to the corporate audit requirements, SOX does little to add ress the underlying factors that create an environment ripe for corporate offending. As law professor and editor of the Federal Sentencing Reported, Frank Bowman, somewhat prison terms on a few crooked or unlucky CEOs was preferable to seriously revisin g the way in W hether this was the intention of some or not, the point holds that the Act provided little by way of serious business reform s Corporate Fraud Task Force At approximately the same time as Congress was in the process of passing the Sarbanes Oxley Act of 2002, President Bush was also creating the Corporate Fraud Task Force, with the al, State, and local specific offenses on which the Task Force was to focus were securities fraud, accounting fraud, mail and wire fraud, money laundering, tax fraud, and other corporate financial crimes. The members, including the Deputy Attorney General, Assistant Attorney Generals from the DOJ Criminal and Tax Divisions, the Director of the Federal Bureau of Investigation (FBI), and


46 several US Attorneys from districts with a relatively high number of corporate crime prosecutions, were also tasked with making recommendations for DOJ policies and resource allocation that would best address the handling of specific financial crimes. In its first year of exist ence the Task Force claimed responsibility for over 250 corporate fraud convictions and estimated that 75 percent of the defendants in these cases were sentenced to incarceration (Corporate Fraud Task Force, 2003). The Task Force also stated in the First Year Report to the President that federal prosecutors working in connection with the Task Force had been awarded $2.5 billion in fines, forfeiture, and restitution. Thus, in the early years of the 21 st century, corporate crime was a hot button issue and th e preferred approach for dealing with corporate offenders was of aggressively seeking them out and focusing on criminal prosecution and conviction. The criminal prosecution of corporate entities also began to receive some negative coverage late in 2002. In March of 2002, the Justice Department announced the indictment of accounting firm Arthur Anderson for obstruction of justice related to the Enron investigation. Arthur Anderson was the auditor for Enron and during the federal investigation into Enron, the book company had to give up its Certified Public Accountant (CPA) ro le and in essence the vast majority of its United States clientele. Approximately 85,000 jobs were lost The Thompson Memorandum Despite the fact that the Thompson Memorandum only slightly added to the already existing Holder Memo ( Spivack and Raman, 200 8) the environment into which it was released in 2003 led to it receiving significantly more attention than its predecessor. The Thompson Memo, formally titled Principles of Federal Prosecution of Business Organizations was


47 disseminated on January 20, 20 03 It included a s trong introductory statement from Larry Thompson, the first Deputy Attorney General under President George W. Bush. In addition to noting the contributions of the Corporate Fraud Task Force and three years of experience working with th e Holder Memo, Thompson wrote, The main focus of the revisions is increased emphasis on and scrutiny of the authenticity of a corporation's cooperation. Too often business organizations, while purporting to cooperate with a Department investigation, in fa ct take steps to impede the quick and effective exposure of the complete scope of wrongdoing under investigation. The revisions make clear that such conduct should weigh in favor of a corporate prosecution. The revisions also address the efficacy of the co rporate governance mechanisms in place within a corporation, to ensure that these measures are truly effective rather than mere paper programs (Thompson, 2003: 1) Though much of t he Thompson memo echoed the Holder memo (Doyle, 2007), it focused on making it clear that superficial shows of coop eration would not be tolerated or feasible, by client and wo (Thompson, 2003 : VI.A ). However, it also openly noted that a non prosecutorial agreement could be considered an appropriate outcome to some criminal investigations, particularly those involving genuine cooperation (Finder and McConn ell, 2006 ). In other words, the Thompson memo to a large degree reflected both the tough on corporate crime mentality of the time and also the fallout from the Arthur Anderson fiasco. It seemed to send the message that the DOJ would not to lerate corporat e misbehavior but that those corporations which assisted the government fully in bringing responsible individuals to justice would be rewarded rather than face a potential corporate death sentence. While the same message was arguably present in the Holder memo as well, certain aspects had particular relevance for the time period as the sting from Arthur Anderson should also be aware of non penal sanctions that may accompany a criminal cha rge such as


48 potential suspension or debarment from eligibility for government contracts or federal funded 1999: IX.B; Thompson 2003: IX.B). In light of Arthur Anderson, this statement had a particular poignancy in the Thompson Memo. Like the Hold er Memo, the Thompson memo stressed cooperation and specified a series of unique mirror ed those laid out by his predecessor. His li st also include d a ninth i tem of consideration as well as new specifications that attorneys should consider from the prosecut Thompson, 2003: II.A.9 ) in their charging determinations Moreover, whereas the Holder Memo spoke primarily of the possibility of granting amnesty or immunity in exchange for cooperation ( Holder, 1999: VI b), the Thomps on Memo more explicitly includ ed the option of pre trial diversion. T hus, the Thompson Memo is generally considered to be the catalyst for the rise in corporate deferred prosecutorial and non prosecutorial agreements (McPhee, 2006). After the Thompson Memo was introduced, there was a significant increase in the number of corporations with which the DOJ entered into DPAs or NPAs (Finder and McConnell, 2006; McPhee, 2006) In fact, from 2003 to 2006, all but one major federal case involving corporate misconduct was resolved without the filing of an indictment against the corporation (Orland, 2006). The McCallum Memorandum In 2005, a third memorandum, further promulgating the Principles of Federal Prosecution of Business Organizations was issued by Acting Deputy Attorn ey General Robert McCallum, Jr. The McCallum Memo (2005) Client and Work Product Protection s timely and voluntary disclosure


49 of wrongdoing and its willingness to cooperate in the investigation of its agents, including, if necessary, the waiver of corporate attorney client and work product protection older, 1999: II, A.4.; Thompson, 200 3: II.A .4 ). At focus was information gathered by corporate counsel during internal investigations which typically would be protected under the attorney client privilege and attorney work product doctrine. In an attemp t to speed up investigations and overturn al l relevant facts in the case, US Attorneys built a waiver of these attorney client and work product protection s into many DPAs. The agreement of a corporation to accept such a condition was seen as a genuine display of the ly cooperate with the federal government ( Spivack and Raman, 2008). However, others argued that the DOJ was exerting undo influence by leveraging the threat of a corporate death sentence much like Arthur Anderson had expe rienced ( Spivack and Raman, 2008) Fifth Amendment right against self incrimination ( Bohrer and Trencher, 2007; Duggin, 2008) While the McCallum Memo was a relatively short, two page document, it both refle cted and foreshadowed issues that had arisen and would continue to arise with the waiver of attorney client and work product protections. The Memo established a uniform review process in which federal prosecutors from all US Attorneys Offices would be req uired to seek approval from a supervisory official prior to requesting a corporate waiver of attorney client or work product protections. Among the critics who continued to protest the unconstitutionality of waivers, however, the McCallum memo did little to improve the problem Some, in fact, viewed it as a merely a


50 The McNulty Memorandum In May of 2006, a federal district court in New York ruled in the case of United States v. S tein that the policies of the DOJ regarding attorney client and work product protection waivers violated the KPMG and rights against self incrimination, as well as Sixth Amendment rights to assistance of counsel (Doyle, 2007 ; Duggin, 2008 Stein ot be said that payment of legal fees for the benefit of employees and former employees necessarily or even usually is indicative of an unwillingness to corporatio process that the constitution adopted as a mode of determining guilt or innocence in a criminal Senate Judiciary Committees held client and work product privilege waivers. The House hearing was titled White Collar Enforcement: Attorney Client Privilege and Corporate Waivers: H earing Before the Subcommittee on Crime, Terrorism, and Homeland Security of the House Committee on the Judiciary was titled Investigations: Hearings Before the Senate C ommittee on the Judiciary During the hearings, witnesses testified that these waivers were virtually unheard of prior to the Thompson Memo; that government agencies had subsequently come to expect companies under investigation to broadly waive employee pr otections fees ;


51 that employees would cease cooperation with in house counsel if they were not protected; and that c ompanies found not guilty of criminal or regulatory wrong doing were vulnerable to third party civil suits because of the disclosure of privileged information (Doyle, 2007 ; Duggin, 2008 ). In response, in December of 2006 D eputy A ttorney G eneral Paul J. McN ulty released a memo geared towards further specifying and restricting the factors considered in assessing corporate cooperation. First, the McNulty memo provided additional guidance and regulation regarding attorney client and work product waivers (Pauls on, 2007) The memo clarified that neither waiver should be considered a prerequisite for cooperation. Furthermore, it instructed client or work product protections where there is a legitimate need fo r the privileged information to fulfill their law enforcement ). Among the factors to be considered in determining whether there is a legitimate need are the degree to which the information is beneficial to the investigat ion, the time and resources saved by requesting a waiver versus attaining the information through other means, the completeness of information provided through voluntary memo also distinguished between requests for Category I information, such as key factual documents, witness statements, and counsel documents containing factual information, an d Category II information, which includes non factual attorney notes or reports containing mental impressions and legal advice given to the corporation by counsel. While requests for both types of information still had to be approved by an Assistant Attorney General as specified by the McCallum memo, category II information could be requested only in rare circumstances when category I information was insufficient for a thorough investigation (Duggin, 2008)


52 Another change from the Thompson memorandum to the McNulty memorandum was in the provisions relating to the advancement of attorn cated in the wrongdoing. T he Thompson memo expressly stated, on for their misconduct, or through providing information to the employees about the government's investigation pursuant to a joint defense agreement, may be considered by the prosecutor in weighing the extent and value of a corporation's cooperation (Thom pson, 2003: V I.B ) However, the McNulty memo instructed prosecutors not to take into account whether es are advanced (McNulty, 2006: VII.3 ). The McNulty memo acknowledged that many corporate charters or bylaws wer e worded such that the corpor ation was contractually discredited for following contractual obligations. Post McNulty DPA C hanges Despite the changes instituted through the McNulty Memo, Senator Arlen Spect e r brought bills before the 109 th and 110 th of waivers. The Attorney Client Privilege Protection Act passed the U.S House of Representatives in November of 2007 (see 153 Congressi onal Record H13562 13564, November 17, 2007), but then died in the Senate. In August of 2008, Deputy Attorney General Mark Filip released DOJ revisions to the Principle s of Federal Prosecution of Business Organizations While the Principles remained large encouraging corporate protection waivers The revised Principles advised that federal prosecutors should only take into consideration the disclosure of relevant facts a nd not the waiver of attorney cooperation.


53 t forbid federal prosecutors from requesting the disclosure of non factual attorney client communications and work product s except under two conditions which were already well established in existing law (Department of Justice, 2008). Concurrent to the re lease of the revisions, the U.S Court of Appeals for the Second Circuit issued an opinion upholding the decision in the Stein case. The Court opined that the dismissal of the indictment against the KPMG officers was warranted on the grounds that the govern ment violated the Sixth Amendment by unjustifiably interfering counsel and their ability to mount a defense ( United States v. Stein 2008 ) The Featu res of DPAs and NPAs Deferred prosecutorial agreements and non prosecuto rial agreements function in a similar manner for all corporations to which they are offered. Both are forms of pre trial diversion. With a DPA, the US Attorney assigned to the case drafts an agreement with a set of terms to which the corporation must com ply. If the corporation accepts the terms and honors the agreement, the criminal complaint filed by DOJ against the company will be dismissed after an established period of time. NPAs involve a similar agreement but the difference is that no charges are actually filed against the corporation if the terms of the agreement are accepted and met. In general, both DPA s and NPAs allow corporations to avoid criminal punishment by instead being held accountable to the government. However, in both instances, if the corporation fails to meet any of its obligations to the DOJ or engage s in additional misconduct while the agreement is active the corporation faces almost certain prosecution and conviction (Martz, 2005). Despite the four memorandums issued to provi de guidance to federal prosecutors handling corporate offenders, there are no formal standards or regulations governing the terms of


54 DPAs and NPAs ( Garrett, 2007; Martz, 2005 ; Warin and Boutros, 2007 ). As Acting Deputy Attorney General Morford noted in 20 Essentially the terms of the agreement are at the discretion of the US Attorney. In theory this lack of specification allows the agreement to be personalized based on the specific characteristics of the corporation, the circumstances surrounding the case, and an evaluation of the need for restitution and rehabilitati on. However, it also means DPAs can suffer from abuse of prosecutorial discretion (Garrett, 2007; Paulson, 2007). T he specific terms of DPAs and NPAs can vary widely from one case to the next and have occasionally been the subject of intense scrutiny (se e McLucas et al., 2006 ). As a Project Director for the National Association of Criminal Defense declared he terms of a Deferred Prosecution Agreement range from the absurd to, if not the sublime, the acceptable in light of 5:45). Despite the variability in the agreements, most do share som e of the same notable features. As Orland (2006) notes, by increasing the use of deferred prosecution agreements, the DOJ wrongdoing by corporate executives must promptly accept full responsibility, discipline wrongdoers, institute serious Orland, 2006: 61). Thus, to achieve this goal, most DPAs and NPAs contain some expression of at least one or two of the following: a statement of general acceptance of responsibility for acts of malfeasance, an emphasis on cooperation individuals to justice increased transparency of disclosure and financial reporting, changes in corporate governance and accountability, the assignment of an outside compliance monitor to the


55 corporation, the replacement of guilty parties within the corporation, a waiver of statute s of l i mitation during the agreement, an element of restitution and the imposition of monetary fines. Voluntary D isclosure and C ooperation From the first iteration of the Organizational Guidelines, the degree to which a corporation cooperated with the governm s investigation has been a factor in DOJ charging and sentencing decisions. The H older Memo provided the initial justification for the emphasis on cooperation, arguing that the nature of corporations is such that investigatio ns into misconduct are comp lex r eturn for corporate cooperation (Couden, 2005: 411). The importance of cooperation as a component of DPAs was then formally acknowledged with both the Holder and T hompson Memorandums listing voluntary disclosure and cooperation as the fourth factor for prosecutorial consideration. Cooperation became increasingly import ant following the Thompson Memo in which the government acknowledged that superficial displays of cooperation were actually impeding investigation s (Finder and McConnell, 2006) The Thompson Memo stressed the need for gaining authentic cooperation rather than allowing companies to merely pay lip service to the idea ( Finder and McConnell, 2006; Spivack and Raman, 2008) The Memo stated that in order to gauge f cooperation the prosecutor could willingness to identify the culprits within the corporation, including senior executives; to make witnesses ava ilable; to disclose the complete results of its internal investigation; and to waive attorney client and work product protection s 2003: VI.A ). Prosecutors were to cooperation would truly assist investigators in br ing ing guilty individuals to justice reducing the time and resources that prosecutors had to devote to case and preventing future offending Additionally, as noted by Michael Elston, former US Attorney and


56 counselor to Deputy Attorney General Paul McNult y, cooperation should be considered a priority because Often the DPA both specifies the terms of cooperation in effect through out the agreement period and details the ways in which the corporation cooperated with prosecutors at early stages of the investigation, contributing to the decision to defer filing in the first place For instance, the DPA drafted in 2000 between prosecu tors in the US DOJ Criminal Division and AGA Med ical Corporation read s: The Department [of Justice] enters into this Agreement based on the individual facts and circumstances presented by this case and AGA. Among the facts considered were: (a) AGA voluntar thorough internal investigation of that misconduct; (c) AGA reported all of its findings to (U.S. Dep artment of Justice Criminal Division, Fraud Section, 2008 : 3 ) Other DPAs and the Canadian Imperial Bank of Commerce, are even more specific, detailing for example, members of the corporate governance already removed from their positions, restitution payments paid, and changes instituted in the governance structure ( U.S. Department of Justice, Enron Task Force, 2003). The degree whi ch subsequent cooperation is emphasized in DPA s varies to a greater deg ree depending on the agreement. Essentially there are two major categories of clauses related to cooperation; one involving the efforts of the corporation to both facilitate individual employee cooperation and testimony and reveal any known misconduct of i ndividual employees and the documents and pieces of information needed to conduct a thorough investigation (Couden, 2005).


57 Of the publicly available corpo rate DP As entered into from 1993 through 2009, about 81%, or 116 out of 143 agreements, contain ed a clause pertaining to the corporation using best efforts to ensure the cooperation of individual employees An example of such a clause is seen in the Agreement be tween Chevron and the US Attorney for the District Cou rt of the Southern District of New York (SDNY) which states among other stipulations that in order to be in compliance of the Agreement Chevron S hall [New York County District At request, use its best efforts promptly to secure the attendance and truthful statements or testimony of any officer, agency or employee at any meeting or interview or before the grand jury or at any conduct by or criminal investigations of CHEVRON or its senior managerial employees well as any administrative proceeding o r civil action brought by any governmental authority that alleges fraud by or against CHEVRON (emphasis included) ( U S. Attorney for the Southern District of New York, 2007 : 2 ) The second aspect of cooperation, the willingness to provide prosecutors wit h access to all necessary information including that which would typically be protected by attorney client and work product privilege, forces the corporation to makes its actions completely transparent. It gives the government full access to the materials it needs for the investigation and allows investigators to assemble the facts of the case without having to jump through hoops at every stage of information gathering (Finder and McConnell, 2006) In theory it allows for a more complete and accurate inves tigation and permits prosecutors to devote less time and resources to the case. T he agreement between PNC I Fraud Section contains a clause exemplifying this aspect of cooperation: PNCICLC agrees that its coop : (a) Completely and truthfully disclosing all information with respect to the activities of PNCICLC and its affiliates, and its present and former officers, agents, and employees, concerning all m atters inquired into by the Fraud Section [Criminal Division, DOJ] as may be requested by the Fraud Section; (b) Assembling, organizing and providing on request possess ion, custody, or control; (c) Not asserting a claim of attorney client or work


58 product privilege as to any documents, information, or testimony requested by the Fraud Section related to factual internal investigations or contemporaneous advice given to PNC ICLC concerning the conduct at issue ( U.S. Department of Justice, Criminal Division, Fraud Section, 2003 : 3 ). In short, DPAs t treatment. In the post Thompson Memorandum days the burden to prove genuine cooperation was arduous, often involving an allowance of total access to documents and records, privilege waivers, vigorous encouragement of individual employee cooperation, and the removal of individuals involved in th e transgression under investigation. The benefits of such cooperation for prosecutors are clear. The corporation essentially helps the prosecutor build cases against individual employees and significantly reduces the time and resources that would typically be required to get to the bottom of what are often complex, multi layered, multifaceted cases. Despite the perceived value of attaini ng corporate cooperation, the emphasis on cooperation has also raised some cr iticism regarding the potential impact on individual employee rights (Bohrer and Trencher, 2007; Duggin, 2008). In regards to the condition that the corporation will ake witnesses available synonymous with the threat of job or benefits termination if the employee refuses to cooperate. Thus, some employees will be in the untenable position of either waiving their Fifth Amendment right against self incrimination or losing their jobs (Bohrer an d Trencher, 2007). In other cases KPMG for example, rather than threatening job loss, corporations have revoked the payment of legal fees for any employee who refused to cooperate with investigators regardless of longstanding polices regarding the remit tance of such fees (Duggin, 2008). In the United States v. Stein (2006) however, the court declared that the government had acted inappropriately by forcing KPMG to coerce employees into cooperation. I n response, then


59 Deputy Attorney General Paul McNult y circulated his Memorandum in which he addressed the (McNulty, 2006: V II.3 ) He furth er clarified though that the exceptions to this guideline included such instances as when the advancement of fees was done explicitly to impede the investigation. The privilege waiver aspect of many DPAs and NPAs is another of the more con troversial components of these agreements. The need for such a waiver was originally seen by DOJ as a Spivack and Ram an, 2008). However, the issue of whether privileged informatio n should be turned over to the self incrimination. Potentially an employee may have spoken candidly with in house attorneys during any internal investigation under t he assumption that such conversations were protected by attorney client privileges. By then waiving this privilege, the corporation essentially waives the incrimination. Moreover, as pointed out by Thomas Donohue, President a nd CEO of the U.S. Chamber of Commerce, in testimony before the House Judicial The const itutionality of such waivers was addressed through both the ruling in the Stein case that the government cannot force a corporation to coerce employees into providing testimony and through House and Senate Judiciary Committee hearings of the 109 th United S tates Congress (Doyle, 2007). In the hearing on the House side, former Attorney General


60 is reasonable and appropriate to them to expect a company under investiga (Doyle, 2007:22). While, Associate Attorney General McCallum countered with his point that privilege, and the societal benefits of rigorous enforcem ent of the laws supporting ethical Ultimately, the 109 th Congress concluded its last days with legislation introduced by Senator Arlen Specter to prohibit federal authorities from requesting a waiver of attorney clie nt or work product protections. To avoid having this legislation passed, the McNulty Memo attempted to address the waiver issue by specifying that waivers could be requested only unde r exceptional circumstances The new Memo directed prosecutors that onl was established, on the grounds that the privileged information was central to the investigation and could not be attained through other sources, could the government request s uch a waiver (McNulty, 2006:VII.B.2 ). In other words, whi le cooperation and voluntary disclosure continue to remai n central components of DPA s, the burden on the corporation to demonstrate genuine cooperation at the expense of employee civil liberties has been relaxe d since the stringent DPAs drafted under the T hompson Memo guidance. Compliance M onitor A compliance monitor is an independent and neutral individual appointed by the US Attorney to essentially sit on the corporate board and serve as both an advisor to the corporation and as the eyes and ears to the government. The overarching goal of the monitor ship is to practice (Ford and Hess, 2009). The monitor is typically expected to attend corporate meetings and is pri vy to financial and operating records for an established period of time (Finder and


61 McConnell, 2006) The idea is that he or she observes where the processes and operations of the corporation lend to uneth ical and unlawful practices He or she then repor ts to the DOJ on how well the terms of the agreement are being followed and works with the corporate board to ensure that preventative measures are put in place to avert future acts of malfeasance and institute change from the inside The concept of a co mpliance monitor is not unique to the DOJ. The Securities and Exchange Commission (SEC) has been known to seek the appointment of a compliance monitor as ancillary relief to supplement injunctions (Ford and Hess, 2009) Such use of monitors has led to cr iticism that t he SEC is attempting to regulate the future behavior of corporate offenders n from the court to require a compliance monitor in high profile cases such as WorldCom and even to expand on the From the perspe is to assess and m The monitor benefits the DOJ by helpi ng to ensure that the corporation is keeping up its end of the bargain and not engaging in further malfeasance To this be highly irresponsible to allow a corporation whose prosecution is being deferred t o go unsupervised during the deferral period (2008: 1054). On the other hand, the monitor also benefits the corporation by theoretically improving its internal controls and corporate governance structure (Spivack and Raman, 2008).


62 A monitor was first u sed in the 1994 DPA negotiated with Prudential Se curities Incorporated ( U.S. Attorney for the Southern District of New York, 1994; Robinson et al., 2005 ). From 1994 to 2009 of the 141 DPAs, 47 (33 %) contain ed the requirement that the corporation retain a n outside compliance monito r. Nearly three quarters ( 72 % ) of these monitorships were implemented during the five year period from 2005 through 2009 ( see also, Ford and Hess, 2009). Depending on the terms of the agreement, the corporation has varying degre es of involvement in the selection and retention of the monitor. Moreover, the monitor also has varying degrees of responsibility and authority within the corporation. In terms of selection and retention, m ost often the corporation must foot the bill to retain the compliance monitor, an expense which can cost the company millions of dollars (Spivack and Raman, 2008). While some agreements note that the selection of a monit or will be jointly made by the prosecutors and corporate counsel, other s make no su ch mention, but instead suggest that the US Attorney has already independently deter mined who will serve the role. On the other end of the spectrum, occasionally the agreements allow for an already appointed company consultant to serve as the monitor (Robi nson et al., 2005) Regardless, because the federal prosecutors have a major if not arguably have limited experience or qualification to make business decisions (Spi vack and Raman, 2008). While monitor s typically have broad authority, the the corporation varies from on e DPA to the next On one end of the spectrum, in the DPA between Zimmer Inc. and the US Attorney for the Distr ict of New Jersey, the agreement specified that the monitor could review and make recommendations on a wide range of company


63 policies, and could hire consulta nts and accountants as desired on the company dime ( U.S. Attorney for the District of New Jersey, 2007). The DPA negotiated with Pru dential Securities specified that the corporation had 30 days to retain a mutually agreed upon compliance monitor appointed to serve on the Board of Directors and submit progress reports every three months (U.S. Attorney for the District of Massachusetts, 2006). In other instances however, the monitor ority is more restricted and he or she is tasked with overseeing compliance with a ween InVision and is one such example, with language that reads, procedures related to FCPA [Foreign U.S. Department of Justice Criminal Division, Fraud Section, 2004: 17 ). As demonstrated through the InVision DPA, most compliance monitors do not actua lly sit on the corporate board. However, most still retain the authority to make recommendations that the corporation must adopt, barring dissent from both the corporation and the USAO ( Christie and Hanna, 2006 ). These recommendations are based on the inside information that the monito r gathers from corporate documents and reports, interviews with employees, and general observations from meetings and day to day activities. In addition to the authority to make recommendations, the monitor is also responsible for scheduling regular progr ess meetings with all necessary parties from the corporation and for submitting regular progress reports to the corporate executives and the DOJ. In 2008, the DOJ faced backlash from critics who argued that the choice and appointment of monitors was based rather than the


64 best interest of the company or the competency of the monitor (Ford and Hess, 2009). Part of the impetus for this accusation was the appointment of former Attorney General John Ashcro firm as the corporate compliance monito r in a lucrative deal earn ing the firm $52 million (Lichtblau, 2008) Facing cries of partiality, the DOJ issued a memorandum to provide more specific regulation of the selection, authority, and length of app ointment of corporate compliance monitors. However, even in the process of issuing more stringent guidelines for US Attorneys, the DOJ made it abundantly clear that the terms of use and appointment should be flexible and ultimately derived from the nuance s and unique circumstances of each corporation and case (Morford, 2008). Business Reforms and Compliance Pr ograms Another ce ntral component of most DPAs is the requirement th at the corporation institute business reforms and compliance program reforms desig ned to make the corpora tion more self regulating and prevent future offending. T h e Holder and Thompson Memoranda advise d that both the remedial action taken by the corpo ration to correct the harms caused by the offense and the existence of a voluntary com pliance program are important factors of consideration in sentencing decisions (Couden, 2005). As specified in the terms of the DPA s, however, these two components can take on a variety of requirements Prior to offering a corporation a DPA prosecutors f avorably weigh whether the organization has a compliance program that allows it to be a self policing entity. The emphasis on compliance programs began in the 1960s when antitru st regulators began considering the presence of a corporate compliance program as a factor in the determination about whether a violation was deliberate versus inadvertent (Ford and Hess, 2009). Then, u nder the Foreign Corrupt Practices Act of 1977 and Sarbanes Oxley, public companies were required to have internal control mechanism s in place (Bucy, 2007 ; Ford and Hess, 2009 ). T he Holder Memo


65 went one step further to say that merely having a compliance program was not sufficient (Finder and McConnell, 2006) While there are no specific rules regarding the content of the compliance maximum effectiveness in preventing and detecting [misconduct] by employees and 1999: VII, B.6). As with cooperation, the Memo suggested that prosecutors should have no tolerance for compliance programs that exist in name only (Finder and McConnell, 2006) While the presence of a compliance programs and the ability of the corporation to uncover misconduct and disclose it to the federal authorities are often citied as reasons for a DPA rather than the filing of the indictment, 54% of DPA s from 1992 through 2009 also include d a requirement that the corporation institute further changes to their co mpliance or ethics program. Columbia and Monsanto Company which reads, MONSANTO COMPANY represents that it has implemented a compliance and ethics program designed to detect and prevent violations of the Foreign Corrupt Practices Act and and implement by March 1, 2005, additional specific new policies and procedures relating to the preventio n and detection of corrupt practices (emphasis included) (U.S. Department of Justice, Criminal Division, Fraud Section, 2005: 6). The Appendix then goes on to further detail these policies and procedures, specifying, among other requirements, the assignmen t of a senior official to provide oversight of compliance with the terms of the DPA, the establishment of a review committee to vet the retention of agents or contractors working with or for foreign governments, procedures to ensure that the corporation do es business with reputable agents and contractors on foreign soil, disciplinary measures for employees who fail to detect violations of the laws or of the compliance policies, and periodic, five year reviews of the corporate policies and compliance progra ms related to the


66 Foreign Corrupt Practices Act ( U.S. Department of Justice, Criminal Division, Fraud Section, 2005 : 22). In relation to the comprehensive corporate reforms pushed by prosecutors, s cholars have noted that Enron world, DOJ officia ls appear to believe that the principal role of corporate criminal enforcement is to reform corrupt corporate cultures that is, to effect widespread structural reform (Spivack and Raman, 2008:161). Cer tainly the emphasis by federal prosecutors on ensuring that corporations adopt internal compliance measures to ensure that previous violations will be prevented in the future reinforces this point. Along similar lines of reforming corrupt corporate culture 73 % of available DPAs from 1992 through 2009 included some type of business ref orm requirement These business reforms range from restructuring the corporate leadership or executive board, to replacing all or certain i mplicated corporate executives, to revising policies related to a specific type of business transaction or ceasing to engage in certain types of transactions One of the best documented examples of the types of business reforms required of DPAs and the logic behind the reforms comes from t he writings of the USAO for the District of New Jersey, which opted to defer prosecution in the Bristol Myers Squibb securities fraud case. Regarding the general negotiations between USAO and Bristol Myers, Christie and Hanna, the U.S. Attorney and Assist ant U.S. Attorney for the District, state that the general objectives for the DPA were to : [Give] n otice to the corporate community, deterrence, full disclosure to the investing public, calibrated reform of a corrupted corporate culture, and restitution t o the victim its law abiding past and present employees, and its current shareholders ( Christie and Hanna, 2006 :1049). The authors then discuss the specific condition built into the DPA to incite fundamental change in the corrupt way business had been conducted at Bristol Myers. For one, the DPA


67 contained conditions designed to not only encourage the company to divulge certain pieces of information to shareholders but also to address corporate attitudes that favored secrecy over transparency in public reporting and disclosure. The agreement required Bristol Myers to utilize external auditors for advice on matters related to discloser and accounting and to meet with tho se auditors on at least a quarterly basis ( Christie and Hanna, 2006 ). which had failed to detect and prevent the fraudulent practices that occurred without excess government intru sion into business decisions, was a more complicated matter for the prosecutors drafting the Bristol Myers Agreement. Ultimately the USAO worked with corporate counsel and developed a condition obligating the company to create a non executive Chairman of the Board position in order that the CEO was not serving as both CEO and Chairman. T he condition reflects the interest of the DOJ to have a system of internal checks and balance within the corporation. As tability should go a long way towards 2008: 1054). There have been concerns about the extent to which the busi ness and compliance program reforms that ar e a condition of so many DPA s are an intrusion of g overnment into busi ness decisions and operations. Former US Attorney Mary Jo White, for example, remarked ( Corporate Crime Reporter, 2005a ) implying that such a r ole may be beyond the realm of their job description and qualification (Spivack and Raman, 2008:186). Another attorney similarly only for government but also for th e corporation and its stakeholders, excessive government intrusion into the affairs of the corporation poses a unique set of risks and business


68 though, business r eform s and corpora te compliance reforms remain among the least controversial, most widely utilized aspects of corporate diversion agreements. Monetary Penalties and R estitution Under the Holder and Thompson Memoranda, prosecutors are instructed that sente ncing decisions should harms inflicted (see Holder, 1999: VIII. B.). This emphasis on restitution translates into monetary penalties and restitution payments frequently inco rporated into the terms of DPAs In restitution payments or monetary penalties than would be imposed upon prosecution and There are a varie ty of ways in which monetary penalties and restitution are bui lt into the terms of the DPA First, because remedial action take n by the corporation prior to any court order provides prosecutors with justification for a deferment rather than an indictment, a number of DPAs contain reference to specific payments made by the corporation prior to the agreement in an attempt to rectify wrongdoings. The DPA negotiated between Aetna Capital Management, Incorporated and Aetna Financial Services, Incorporated and the USAO for the District of Massachusetts, for example, states that among other factors, the deferment has been granted approximately $9.5 million to its affected custome U.S. Attorney for the District of Massachusetts, 1993 : 2 ). In a few cases, the agreements further note that due to the monetary payments already made by the corporation, no further fines or penalties will be assessed as a conditio n of the DPA. For instance, in light of the $1.1 billion settlement paid by Royal Ahold N.V. and U.S. Foodservices, Incorporated, prosecutors determined that no further


69 monetary payment or fine was necessary within the terms of the agreement (U.S. Attorney for the Southern District of New York, 2006). In addition to the reference to previous fines paid by the corporation, a numbe r of DPA s do also contain punitive fines and penalties which are thought to provide a measure of deterrence ( Christie and Hanna, 2 006 ). These fines range from $20,000, a civil penalty required in the terms of the agreement with Armour of America, to the over $700 million that Aldelphia agreed to pay in order to avoid formal indictment (Finder and McConnell, 2006). In some instances such as the case with both Prudential Equity Group and HealthSouth, these fines are designated to be paid into a general federal fund to assist victims of fraud, such as the Postal Inspection Consumer Fraud Fund or the Securities and Exchange Fair Fund In other DPA s at least a portion of the required monetary payment is labeled as strictly a criminal penalty. America Online, for example, agreed to establish a $150 million fund to be used for the settlement of shareholder securities law litigation and t o pay $60 million to the US Treasury as a monetary penalty ( U.S Attorney for the Eastern District of Virginia, 2004 ). The US Attorneys responsible for the drafting of DPAs regard monetary penalties and restitution payments as very different things. Of the 110 DPA s from 1992 through 2009 that contain ed payment requirement s 25% include d some type of restitution requirement The Bristol Myers Squibb DPA obligated the corporation to pay an additional $300 million to victimized shareholders, above the $539 mi llion the company had previously shelled out in restitution payments, but did not require any other fines or penalties ( Christie and Hanna, 2006 ). Former US Attorneys Christie and Hanna explained the reasoning of the USAO seeking restitution only: The cri minal complaint and terms of the agreement itself provide adequate deterrent effect;


70 en aiding the victims of the crime and preventing any future criminal conduct. If we were to err on one side, we chose to err towards restitution to the victims and a solid company in the aftermath (2008: 1059 60). Other DPA s stipulate a monetary payment that resembles a form of community service more than a criminal fine or restitution payment. Often the community service feature is linked to the specific conduct of the corporation, but this is not always the case (Finder and McConnell, 2006). Some of t he more notable examples of the co mmunity service requirement are contained within the DPAs for FirstEnergy, Roger Williams Medical Center, the New York Racing Association, and Bristol Myers Squibb. FirstEnergy had been implicated in violations of various environmental laws and the DPA included $4.35 million designated to go towards protection of the environment in the Northern District of Ohio, the building of energy efficient homes for Habitat for Humanity, and the development of energy efficient technol ogies at the University of Toledo. In a DPA related to a public corruption case, Roger Williams Medical Center agreed to provide $4 million worth of free, non tax deductible medical services. Along slightly different lines, the agreement negotiated betwee n the USAO for the Eastern District of New York and the New York Racing Association (NYRA) included a requirement that the NYRA install video lottery slot machines at its venues to help the state of New York raise money to fund public education (Spivack an d Raman, 2008). Finally, in one of the more notorious of the community service requirements, Bristol Myers agreed to endow a chair at Seton Hall University School of Law, dedicated to the teaching of business ethics and of U.S. Attorney for the District of New Jersey, 2005 : 6). While, in general, the restitution, co mmunity service, and monetary f ines required of DPA s are well accepted, some observers have suggested that the


71 community service aspects are an example of federal prosecutors overstepping their bounds of prosecutorial discretion (Coffee, 2005; Finder and M cConnell, 2006). In the Bristol Myers case, the US Attorney was an alumnus of the Seton Hall Law School receiving the endowed chair position, a connection which raised concerns that prosecutors were using the flexible guidelines for DPAs to advance person al causes. On the other hand, the NYRA case raised allegation s that Perhaps mindful of these crit icisms, the more recently drafted DPAs and NPAs from 2007 tend to include only the more standard terms for monetary payments and restitution (Spivack and Raman, 2008). In order to begin to assess whether the major components of DPAs contribute to deferment being a therapeutic approach to punishing corporate offender, Chapter 4 takes a closer look at the expected implications of a punishment that includes an independent monitor, business and compliance program reforms, monetary penalties and restitution, and a requirement of cooperation with federal prosecutors.


72 CHAPTER 4 NON EMPIRICAL ANALYSIS O F THE THERAPEUTIC VA LUE OF DEFERRED PROSECUTORIAL AGREEM ENTS The goal of the non empirical analysis is to examine research that relates to the major components of DPA s The therapeutic jurisprudence perspective science to study the extent to which a legal rule or practice promotes the psychological and physical well being of the peopl : 193) Corresponding ly, Chapter 4 will explore wh at social science literature suggest s regarding the therapeutic or anti therapeutic qualities of each of the major component s of DPAs Prior research has demonstrated that traditional criminal sanctions against corporations ar e anti therapeutic in that they unsuccessfully prevent future offending and instead can have devastating collateral consequences for innocent bystanders. As noted in Chapter 1 there are thre e overriding problems with the use of traditional criminal sancti ons to punish corporate offenders: 1. criminal investigations and the disposition of corporate cases are lengthy and burdensome processes that rely on punishment threats and fear to deter and do not meet either the certainty or celerity conditions of deter rence; 2. traditional criminal sanctions do not change the corporate environment which facilitated the offending behavior; and 3. when criminal sanctions are severe enough to have a meaningful effect on the corporation, this effect is often felt by innocen t employees and shareholders. Thus, before examining the therapeutic value of each major component of DPAs, it is necessary to address how DPAs in general, offer a solution to the three aforementioned problems. For starters, w hen preventive effects are ac hieved though socialization and education rather than incapacitation and punitive action this is a central consideration for the therapeutic jurisprudence perspective Therapeutic jurisprudence postulates that the most effecti ve punishment is not the most punitiv e punishment but the punishment that best prevents future


73 offending Likewise, DPAs are generally considered to be less punitive than formal crimina l sanctions and focus on socializing and rehabilitating corporat ions rather than on strictly punishi ng them. G oing back to Beccaria in the 1960s, researchers have found that certainty of punishment is a more crucial factor in the calculation of the costs and benefits associated with illegal behavior than the other dimensions of deterrence (Blumstein et a l., 1978; Erickson et al., 1977; Grasmick and Bryjack, 1980; Paternoster, 1987; Williams and Hawkins, 1986) DPAs also remove much of the uncertainty surrounding whether a corporation will be charged and convicted for an offense. Once a corporate offense i s discovered, much of the uncertainty with the case stems from the complex nature of corporate crimes and the reality that the prosecution may not be able to build a strong enough case against the corporation to file charges and successfully convict. DPAs remove this level of uncertainty by giving prosecutors a middle ground between filing and not filing charges. They allow corporations to be punished without the ilt beyond a reasonable doubt. Thus, with DPAs as a punishment option, the perceived certainty of corporate punishment should increase. Research on deterrence also suggests that punishment celerity, the swiftness with which a punishment occurs after the of fense, reduces the likelihood of recidivism ( Ross, 1987 ). Likewise, DPAs allow a corporate punishment to be imposed more expeditiously than it would be if formal charges were filed in court In speaking about DPAs, Columbia Law School professor John Coffe e noted: You are getting about what you would get at sentencing. You are getting it quicker. The corporation is forced to enter into this plea bargain and admit its guilt at the outset, rather corporation against its


74 former officers and against its current officers by firing them you get immediate reform (Corporate Crime Reporter, 2005 b ). In other words, since DPAs do not require the same burden of proof that criminal sanctions require and they allow for the bypassing of lengthy criminal p rocedures, t he punishment is presumably carried out more swiftly In terms of being better able to rehabilitate a corporation than traditional fines, DPAs also have business reform and cooperation elements built into them that will again be discussed in detail later in Chapter 4 Unlike fines and other traditional corporate punishment s which do not require any significant change to the corporate behavior DPAs are focused on genuine change and cooperation w ith the Department of Justice. In fact, the Thompson and subsequent memos state that in order to be eligible for deferment of prosecution the corporation should have independently tak en remedial actions to right any harm caused by the malfeasance The m emos specify that these remedial actions in clude implementing an effective compliance program, r eplacing accountable management, disciplining or terminating guilty parties, paying restitution, and cooperating with government agenc ies (see Thompson 2003; M cNulty, 2006 ). This pervasive idea of rehabilitation and of making a corporation a partner in their own punishme nt, so to speak, is also aligned with the foundational views of the therapeutic jurisprudence perspective. Finally, DPAs are therapeutic because they are thought to lessen the collateral consequences of criminal sanctions against corporations. DPAs do still require most corporations to pay some type of fine or restitution and these costs trickle down to shareholders to some degree. However, histor ically the biggest impact to innocent employees and shareholders has occurred when a corporation is forbidden from engaging in certain types of business transactions or working wit h certain clients because of criminal char g es.


75 The once powerful accounting firm Arthur Andersen exemplifies this point. Following the Enron scandal, many states revoked Anderse Accountants (C PAs) and eventually the company voluntarily surrendered its CPA licenses entirely. W ith the accounting segment practices were sold off to various accounting and consulting companies. A rthur Andersen de clined from 85,000 employees worldwide in 2002 to about 200 employees in 2007. Presumably, many of the nearly 85,00 0 employees who lost their jobs with the firm were not involved in illegal activities and innocent shareholders saw the price of their stocks plu mmet. Had Andersen enter ed into a DPA with the Justice Department rather than face criminal conviction, they would have avoided at least some of these losses to innocent bystanders Additionally, not only do innocent individuals often suffer from the collateral consequences of criminal sanctions imposed on corporations, the culpable individuals within the corporation are sometimes able to hide behind the shield of the corporation and avoid personal punishment. With DPAs though, the corporation agrees t o cooperate with the DOJ by turning over relevant documents pertaining to the case rather than serving as a road block to prevent these documents from being obtained. This allows the DOJ to build a strong case against guilty individuals within the company rather than risking the collateral consequences of corporate charges. On the surface, DPAs appear to be more therapeutic than other punishment options for corporate offenders. However, to get a true sense of the elements of DPAs that are therapeutic versu s anti therapeutic the remainder Chapter 4 will examine social science research findings that can be applied to each of the maj or components of DPAs. The chapter will be used to make


76 predictions about the therapeutic qualities of DPAs based on these resea rch findings in other areas. Voluntary Disclosure and Cooperation Cooperation is a central component of DPAs and arguably one of the more apparent example s of the therapeutic nature of DPAs On the surface the concept of cooperation is synonymous with the rapeutic jurisprudence because it suggests that all parties are working together for the best possible outcome. The corporation is able to avoid criminal charges and the negative reputational and eco nomic consequences that can result from an indictment Th e government saves valuable time and money investigating and building the case because the corporation makes relevant documents and materials available rather than trying to bury or destroy them. Society benefits from any preventative and deterrent effect s on the particular corporation and other corporations that is achieved by swiftly bringing an offending company to justice. Finally, victims and innocent member of the public are not subjected to the negative collateral consequences of corporate indictment However, cooperation and voluntary disclosure policies have been some of the most criticized aspects of DPAs (see Bohrer and Trencher, 2007). As noted previously, until the McNulty Memorandum was issued in 2006 a corporation was essentially expected to waive attorney client privileges and work order protections in order to demonstrate cooperation. Legal scholars rightfully ar gued that such a waiver violated the rights of employees ( Bohrer and Trencher, 2007; Griffin, 2007 ), and because of this violation of rights the cooperation clauses made DPAs a less therapeutic punishment option. By 2008 the Principle s of Federal Prosecution of Business Organizations wer e revised to instruct attorneys that cooperation should not rest on the waiver of privileges or the advancement fees, but should simply refer to a corporation working with the federal government


77 to ensure that relevant facts about the case were disclosed. The revision of the Principles did not lessen the importance of c ooperation in DPA s, though D ue to the weight placed on cooperation and the removal of its most controversial and anti therapeutic aspects an assessment the therapeutic nature of DPAs and the likelihood of a corporation reoffendi ng cannot ignore cooperation. S ocial scien ce research has not specifically examined whether the concept of cooperation between an offender and criminal justice authorities has an impact on recidivism. However, an examination of literature on p rocedural justice and recidivism prevention sheds light on the expected outcome of including cooperation in DPAs. Both bodies of research suggest that cooperation should have positive, therapeutic effects on the victims, the offenders, and society at large. Procedural justice pertains to the perceived fairne ss of the criminal justice process as well as the perceived treatment received from the decision maker (e.g. law enforcement officer, prosecutor, judge) (Murphy, 2009; Tyler, 2006). According to extant literature, p rocedural justice increase s the legitim acy of the law for those exposed to the justice system; thereby increasing the likelihood that the individual (or corporation) will have more respect for the law in the future (Tyler, 1999, 2006). While much of the work done in this area has focused on pro cedural fairness related to law enforcement practices, the general themes are applicable to the handling of corporate offenders as well. Research on the factors affecting perceptions of procedural justice demonstrates both that cooperation is important in establishing the legitimacy of the law and that cooperation increases as perceptions of fairness and legitimacy increase. A 1992 study conducted by Tyler and Lind, revealed that perceptions of fairness and respect for criminal justice authorit ies a re


78 related to whether justice system authorities demonstrated sufficient consideration for their arguments or views. In other words, when people felt that they were given a voice in the criminal justice process and were treated with respect, they in t urn had more respect for the criminal justice system. By definition, cooperation implies that the offender has a say in the process and the outcome. Procedural justice literature also finds that sanctions perceived as fair legitimate, and not overly puni tive are better able to reform and rehabilitate (Braithwaite, 1989; Makkai and Braithwaite, 1994; Sherman, 1993). A study of nursing home compliance in Australia found that when regu lators inspecting the nursing homes had a disrespectful or stigmatizing a ttitude, compliance with the inspection standards declined in the period following the visit. By contrast, when inspectors tried to convey disapproval without causing humiliation, compliance with the standards significantly improved following the inspecti on (Makkai and Braithwaite, 1994). While these findings do not directly address DPAs, the focus on cooperation implies that the federal government and the corporation are working together to punish and rehabilitate the corporation. Therefore, a n offending corporation that enters a DPA is more likely to perceive the punishment as a fair and legitimate than a corporation that receives criminal sanctioning after a long, drawn out litigation process. While corporations cannot exactly respond to unfair sanctions legitimacy of the law and authority permeates a corporation, the probability of re offending is reduced. Not only is cooperation linked to perceptions of procedu ral fairness, the voluntary disclosure and cooperation components of DPAs will also speed up the criminal case investigation and resolution. On ce a corporation has demonstrated a willingness to cooperate the


79 DOJ can quickly en ter into a DPA with the compa ny instead of going through years of criminal investigation. How quickly an offender is punished is important from the perspective of criminological theory which specifies that celerity, or swiftness, of punishment is one of the three factors predicting bo th general and specific deterrence. F indings regarding the independent effects of punishment celerity on recidivism are somewhat mixed though S ome stud ies have found that the preventive eff ects of celerity are washed out by factors such as punishment certainty and sev erity (Simpson and Koper, 1992). O thers have found at least some support for the notion that swift punishment reduces the probability of reoffending (Ross, 1987) For example, one study evaluated a program implemented in New Mexico in whi ch law (Ross, 1987). An interrupted time series analysis of data on Driving While Intoxicated (DWI) arrests in two New Mexico counties revealed substantial de clines in the number of drunk driving arrests following the implementation of the administrative license revocation law. A similar analysis of data on fatal motor vehicle accidents in New Mexico also revealed a ten percent reduction in the percentage of cr ashes in which the driver had illegal blood alcohol concentrations. viewed as emphasizing swiftness of punishment, these results testify to the potential of the previously unexamin Another study examining the records of nearly 600,000 drivers in New York State convicted of DWIs, looked at the relationship between high fines, time from arrest to conviction, the interaction be tween the two, and drunk driving recidivism (Yu, 1994). The study found that while celerity had no measurable effect as an independent factor, the interaction between celerity


80 and high fines did have a statistically significant impact on recidivism. The au thor further noted that celerity may not have been a strong independent factor in the model because very few of the convicted offenders actually had a short period of time between arrest and conviction. In other words, while research has not conclusively d emonstrate d that swifter punishments prevent future offending there is evidence to suggest that celerity does have preventive effects Taken together, research on punishment celerity and procedural justi ce suggest that the strong focus of DPAs on coopera tion and voluntary disclosure contribute to the therapeutic nature of the punishment. With the cooperation focus, DPAs are expected to increase the legitimac y afforded to the law, thereby increasing the likelihood that the corporation is rehabilitated. Bus iness Reforms and Compliance Programs The majority of DPAs (54%) from 1992 through 2009 require d either some form o f business reform, or both In some instances the reforms were a prerequisite fo r the agreement; in others they were required as a term of the agreement. Examples of the types of business reforms and compliance program changes addressed within DPAs include: restructuring of the corporate board, the replacement of certain or all corpo rate executives, the revision of policies related to a specific type of business transaction, the implementation of a new compliance or ethics program, additions to an existing compliance program, such as a hotline for employee concerns, or a change in the frequency and intensity of employee participation in an existing compliance program. Research in criminology and business has long suggested that business reforms, particularly a change in corporate management, and the presence of a strong compliance and ethics program, are necessary for true corporate rehabilitation and the prevention of future offending i.e. a therapeutic outcome. An article in the Yale Law Journal in 1979 states,


81 procedures be restructured in such a way as to foster future compliance with the law; institutional elements that facilitated the commission of an offense must be modified so that they operate subsequently to prevent : 361). Thus, just on the surface, the fact that DPAs contain business reforms at all suggests that DPAs may be a more therapeutic corporate punishment option than punishments that do not address the rehabilitation of the corporation. Business R eforms In 1979 the aforementioned research note in The Yale Law Journal proposed the idea of corporate probation as an alternative to the fines that were most commonly doled out as corporate punishment (The Yale Law Journal, 1979). The concept was based upon the b elief so as to reduce the probability of future vio 1979: 361). Thus each probationary term would center on judicial ly mandated c or porate restructuring targeted at the internal operations and p rocedures that had facilitated the offending behavior The forced corporate reforms would serve deterrent and retributive purposes, further rehabilitative goals, and lessen the burden of punis hment imposed on innocent shareholders and consum ers. Essentially, the research note proposed that corporate probation involving judicially mandated business reforms would be a more therapeutic approach to handling corporate offenders than the commonly ut ilized sanctioning methods. The research note further laid out more specific reasons why effective corporate punishments should incorporate business reforms. First, courts had previously been successful in implementing similar organizational reforms wit hin public i nstitutions such as prisons ( Yale Law Journal, 1979: 365). Since reforms of entire public institutions were more complicated than would be required to reform certain operations or procedures of corporations, success in the


82 public realm implied that similar reforms could be successfully implemented within offending corporations. Next, from a deterrence perspective, forced corporate compliance from an outside authority imposes a burden on impacted managers that may serve as a deterrent for other corporate leaders. Finally, corporate offending is often attributed to institutional forces rather than particular individuals so unless those forces are changed, the offending behavior will continue despite any replacement of directly responsible parties The latter explanation for why corporate punishments that include business reforms are more effective than a traditional fine is further elucidated in corporate crime research on the criminogenic aspects of the corporate environment and organization. O rganizational theories of offending suggest that the market internal competition and lax managerial oversight create an atmosphere in which corporate employees are inclined to do whatever it takes to increase personal and company success and reduce the o dd of failure (Coleman, 1995; Paine, 1994; Simpson and Piquero, 2002 ; Sims, 1992 ). The structural and organizational characteristics of a corporation can then serve to either diffuse or facilitate the urge to seek effect ive but illegitimate means for gett ing ahead and increasing success. In a corporate setting with lax oversight standards (Vaughn, 1996), unreasonable goals for employees (Clinard, 1983; Reed and Yeager, 1996), and minimal efforts at instituting compliance and ethics training and programs ( Braithwaite, 1984, 1989; Kram et al., 1989 ; Paine, 1994 ) the structure of the organization effectively encourages malfeasance. As the behavior continues, it becomes a normalized, inherent part of the corporation that is manifested in the attitudes and act ions of the employees (Coleman, 1987). At this point, traditional punishments have very little deterrent power against an institutional environment that is conducive to offending. New employees will continually evolve into products of the environment and i t will only be a matter of time before recidivism


83 occurs. Thus, the way to break the cycle of offending is to correct an institutional structure and environment that is favorable for malfeasance. As organizational theories and other research have shown, ma nagers are largely responsible for the ethical climate within a corporation (Hollinger and Cl ark, 1982, 1983 ; Moohr, 2007 ) Top managers establish and reinforce the culture in which employees learn to behave (Appelbaum et al., 2005). In interviews with 64 retired, middle managers from 51 Fortune 500 companies, Clinard (1983), found that the most common explanations for illegal behavior were the character of and standards set by top managers. Hollinger and Clark (1983) reached a similar conclusion approach ing their research from the perspective of employees rather than managers. They conducted a study of 9,000 employees that examined, among other things, the relationship between employee deviance and employee perceptions of the sanctions im posed by manageme nt. Both t sanctioning highly correlated with employee misbehavior. Both were also stronger predictor s of likelihood for workplace deviance than were other personal attr ibutes, such as personal financial need and job dissatisfaction (Hollinger and Clark, 1983). Their study supports the notion that employees workplace deviance mimic those attitudes that they perceive in their managers and coworkers. A l ater study, conducted by Kamp and Brooks in 1991 replicated and extended Hollinger and more lenient or more strict than the norm for management tends to be accompanied by employee attitudes that are more pro theft or anti theft t Kamp and Brooks, 1991: 455).


84 Likewise, a study of intentions to offend among about 100 bu siness executives in an MBA program, found a significant decline in intentions to engage in corporate offenses when another employee had previously been fired for a similar behavior (Simpson and Piquero, 2002). In other words, when management demonstrates that illegal behavior will not be tolerated, these actions and attitude appear to have a preventive effect on other employees. While these studies refer to employee offending that is directed against the company rather than for the benefit of the company there is agreement that managers are responsible for creating an ethical climate within a company and that the development of such a climate involves more than simply avoiding illegal conduct. While managers may not necessarily intend for employees to co mmit offenses, they establish a culture with extreme performance pressure without tools in place for regulating whether the process for meeting the performance standards is legitimate ( Cohen, 1993; Ford and Hess, 2007; Paine, 1994). Therefore, replacing th e top management of a corporation as many DPAs require, is an integral step towards changing the corporate culture and preventing continued corporate offending. Other empirical studies have demonstrated that not only the corporate culture, as established by management, but also the structure of the corporation impact illegal corporate behavior. Baucus and Near (1991), for example, conducted a 19 year event history analysis of Fortune 500 firms and found that both the economic environment and the size of th e company increased the likelihood of corporate offending. While company size and the economic environment are not factors that can be altered through the terms of a DPA, there are multiple other unfixed corporate policies that have been found to increase the likelihood of illegal behavior on the part of employees. Such policies that have been discussed in extant research include: an emphasis on meeting performance goals with


85 out sufficient monitoring to ens ure that the goals are achieved ; excessive push fo r profit by the corporation ( Clinard and Yeager, 1980; Passas, 1990 ); generous stock options and compensation for executives that encourage aggressive measures to stay ahead at any cost ( Bratton, 2002; Clinard and Yeager, 1980; Dallas, 2003; Moohr, 2007 ); and a corporate board that is not in touch with regular business operat ions All of these types of policies have been shown to contribute to a corporate culture that fosters illegality. Thus, DPA terms that require a replacement of existing management, a c hange or additional oversight for the corporate board, and other related business reforms make for a more therapeutic punishment. These business reforms reduce the probability of future offending by forcing the corporation to improve the corporate climate that facilitated the offending in the first place. Compliance and Ethics P rograms Research in the area of organizational theory and the impact of corporate culture on corporate offending also suggests that corporations with effective compliance and ethics programs are less likely to engage in malfeasance. strong evidence to believe that properly implemented compliance programs can improve ethical behavior in organizations and reduce the high levels of fraud tha t currently exist (Hess, 2007: 1804). Simply having a compli ance and ethics program on the books is not enough, though. Various s tudies and guidelines have identified specific features of these programs that are necessary for effectiveness; many of which a re also addressed in the terms of DPAs. For starters, the Organizational Sentencing Guidelines (OSG) specify that in order to be considered effective in the eyes of the United States Sentencing Commission, a compliance program must, at a minimum: establis h standards of conduce and internal controls to prevent and


86 governing authority; make all reasonable efforts to avoid hiring personnel who have behaved in a manne r inconsistent; regularly communicate its standards and procedures; engage in auditing and monitoring to detect criminal conduct; actively promote and enforce compliance with appropriate incentives and disciplinary action; and respond immediately to any cr iminal conduct that is discovered and take action to prevent a similar incident from occurring again to that prom oted by the compliance program (USSC, 2010, §8b2.1.). The idea is to create a corporate climate that reduces opportunities for employees to eng age in misconduct and holds managers and those in high ranking positions culpable for the actions of low level employees (Ferrell et al., 1998). A study of over 4,000 companies conducted by KPMG Forensic in 2004 and 2005 found that employees in companies w ith complete compliance programs under the OSG guidelines were significantly more likely to believe that misconduct was not tolerated, that there were limited opportunities to engage in misconduct, and that people in the corporation were motivated include all the elements (Hess, 2007). This finding suggests that when a DPA requires a out by the OSG, the compliance program will have a positive effect on the ethical climate of the corporation. Research has also demonstrated that t he management and structure of a corporation is r elated to the effectiveness of its compliance and ethics pr ogram in discouraging deviant behavior. As Professor of Law, Lynne Dallas pointed out in a 2003 Rutgers Law Journal article, with ethical decision making and b ehavior. Codes of ethics can contribute, however, to an ethical


87 Dallas, 2003: 32). Likewise, a study conducted in the late 1980s to early 1990s indicates that compliance programs may not s ufficiently impact corporate behavior without a change in upper management. The study examined whether the compliance programs advocated by the 1991 Federal Sentencing Guidelines reduced corporate illegality. Using TOBIT regression analysis to assess the relationship between the self reported compliance programs of 108 major manufacturing companies and data on the corporate violations of these companies from the Occupational Safety and Health Administration (OSHA), the authors found no significant relationship between Sentencing Guideline compliant ethical program s and fewer legal ethics programs are to be developed, the y must be supported by top management; this factor is far more important than external controls such as 2002: 379). Thus the most therapeutic punishment in terms of rehabilitation and preventative effects require s both a change to the compliance and ethics program as well as a change in the upper management of t he corporation. As discussed in the previous section, most DPAs require a change in upper management. Those that also include the adoption or improvement of a compliance and ethics program should have a therapeutic impact. Thus, when DPAs contain clause s t o reform business practices particularly involving a change the upper management and structure of the corporation as well as to develop or improve compliance progra ms, they are expected to have a longer lasting, more therapeutic effect on the corporation than a typical criminal punishment might. Monetary Penalties and Restitution Assuming that fines and restitution are determined to be a therapeutic form of punishment for corporate offenders, the in corporation of monetary penalties in to DPAs seemingly br idges a gap between those who argue for and those who argue against the use of criminal sanctions for


88 corporate offenders. As noted previously, many scholars have argued against the use of criminal sanctions against corporations because of the collateral consequences of formal charging and sentencing (Bohr and Trencher, 2007; Bucy, 2007) and because of a lack of empirical evidence that criminal san ctions reduce the likelihood of reoffending (Geis and DiMento, 2002; Simpson, 2002). However, there are also a substantial number of scholars who have argued from political, social, economic, a nd criminological perspectives that there is a need to punish corporate offenders through formal criminal sanctions (Bucy, 2007; Friedman, 2000; Saltzburg, 1991; Ulen, 1996) These scholars tend to agree both that criminal law is the most effective means for influen cing behavior (see Bucy, 2007) and that since corporations cannot be imprisoned, monetary fines are one of the most severe and effective penalties that can be levi ed against a corporate offender (Clinard and Yeager, 1980; Ulen, 1996). General criminological literature focusing on the use of monetary penalties tends to support the notion that fines are an effective sanctioning mechanism and can have a general preven tative effect for all types of offenders. While fines are not typically used as stand alone pu nishments in the United States they have been widely implemented in a number of European countries such as England, Sweden, and Germany, and are the preferred ty pe of punishment in these areas (Tonry and Lynch, 1996 ; Gorden and Glaser, 1991 ). Based on evidence from abroad as well as various research studies many scholars have argued that fines offer sufficient deterrent, retributive, and rehabilitative advantage s to also be the primary form of punishment in the United States (see, for example, Morris and Tonry, 1990). Gordon and Gla ser (1991) for example, conducted a study which examined the effect of monetary fines on post sentencing outcomes for over 800 offen ders in California in the early 1980s. The offen ders had been convicted of either assault, burglary, drug crimes, driving under


89 the influence (DUI), or theft and all were sentence d to probation or probation pl us another sanction(s) After controlling for o ffender characteristics and the type of offense, the researchers found that individuals given a sentence that included a fine were less likely to be rearrested in the two years following the sentencing than those sentenced to just probation or a sentence t hat included jail time. Financial penalties as a condition of probation were also associated with a lower likelihood of probation revocation. The amount of the fine however, had little effect on post sentencing recidivism. Gorden and Glaser light of the cost benefits The preventive effect of fines has also been examin ed briefly in the context of driving related offenses. The two studies in this category, however, revealed mixed findings regarding the use of fines as an effective punishment. A study examining drun k driving recidivism among of sample of almost 14,000 dru nk drivers in New York found that the probability of recidivism was negatively correlated with fine amounts and that high fines were more of a deterrent than license revocation (Yu, 1994). On the o ther hand, a study conducted in New South Wales, Australia examining the court appearance records of between 7,000 and 15,000 individuals accused of a range of traffic offenses, failed to find evidence of a significant relationship between fine amount and the likelihood of recidivism (Moffatt and Poynton, 2007). O verall, the aforementioned studies have mostly supported the use of fines as a punishment option for general criminal offenses. E conomists, such as Tom Ulen, have argue d further that for business organizations monetary sanctions are in fact, preferable to non monetary sanctions (1996). Ulen argues that employees, shareholders, and customers shoulder the losses from increased corporate offending and that fines are one way of ensuring that the


90 corporate offender feels these losses as well. Other experts simp ly point out that b ecause co rporations cannot be imprisoned fines are among the most severe penalties that can be levied against a corporate offender Thus, fines should have the greatest deterrent poten tial of all sanctions against corporate offenders According to the economist model for understanding deterrence, effective deterrence a multiplier to counterbalance the possibility that a violation would esc 1985: 447). In other words, a fine can deter a corporate offender if the expected cost of the punishment is greater than the ex pected benefit of the crime, p lus an estimation of the likelihood of getting away with the offense ( Ulen 1996). F or fines to be effective they must be based on the expected gains from an offense rather than a measure of the harm caused by the offense. One of the more widely known examples of the cost/benefit considerations that go into corporate behavior w as the 1970 decision by executives at Ford Motor Company to keep faulty parts in the Pinto despite the knowledge that these parts could be deadly for Pinto drivers. Ford executives calculated that while the replace ment of the parts would cost $137 million dollars, the most they would be required to pay in a wrongful death civil lawsuit was closer to $50 million dollars. Thus, Ford decided not to replace the faulty parts and 27 deaths were later attributed to this decision (Lee and Ermann, 2002). The Ford Pi nto case is frequently cited as evidence that if fines are imposed on corporate offenders, the amounts must reflect the magnitude of the offense. Historically though, th e fines imposed for most corporate offense s have tend ed to be small in comparison to th e profits made through the illegal behavior (Clinard and Yeager, 1980; Geis, 1973). In 1979 the Harvard Law Review argued for ution as


91 I n one of the first large cases with a fine levied against a corporation for a violation of anti pollution laws, Chevr on was forced to pay $1 million. This represented a mere income (Shostake, 1974). Data on federal prosecutions against corporate offenders in the mid to late 1980s showed that about 90% of corporate offenders received fines averaging around $50,000 (Simpson, 2002). While mo re recent statistics suggest that fine amounts increased after the Federal Sentencing Guidelines were implemented, the average is still just over $200,000. The opposing consideration to the need for increasing fines is that shareholders not the manager s who are likely to be making any cost benefit decision calculations ultimately bear the cost of every fine imposed (Kennedy, 1985). Every dollar of a fine directly reduces the value (Kennedy, 1985) Howe ver, Gilbert Geis (1973) would seem that rousing fines against offending corporations will at least lead to stockholder retaliations against lax or offendi ng managerial personnel, and will forewarn officials in other (1973:193). Similarly, Christopher Kennedy (1985) points out that while most stockholders have no a bility to directly control or monitor corporate management shareholder satisfaction is a major consideration for managers In other words, the potential drop in stocks from the imposition of a large fine should, at least indirectly, impact organizational decision making. Moreover, if larger fines were consistently imposed against corporate offenders the market would eventually necessitate that the increased risk of loss from fine s would correspondingly result in lower priced shares (Kennedy, 1985). Thus, the innocent shareholders would benefit from reduced price


92 shares but would, as per usual, have to consider all of the risks of investing with the corporation. crim In addition to fines imposed for the purpose of punishing the offender, monetary penalties can also be imposed for victim restitution. The goal of a restitution payment is to have the offender compensate the victims for economic losses suffe red as a result of the offense. In 1982 as part of the Victim and Witness Protection Act, the United States Congress declared that: Restitution is an integral part of virtually every formal system of criminal ju stice, of every culture of every time. It hol ds that, whatever else the sanctioning power of society does to punish its wrongdoers, it should also insure that the wrongdoer is required to the degree possible to restore the victim to his or her prior state of well being (U.S. Code Congress and Adminis trative News, 1982: 2515 2516) In the federal court system, about 90% of fines are designated for victims and deposited in the Crime Victims Fund (Ruback and Bergstrom, 2006). Restitution penalties can also be designated for specific victims in specific a mounts. While the benefits of restitution for victims are apparent, scholars have also argued that restitution is therapeutic for offenders because it allows the offender to make amends for wrong doings (Cohen, 1944; Outlaw and Ruback, 1999; Sims, 2000). As Outlaw and Ruback explain, 1999: 850 851). Similarly, several studies have suggested that offenders required to pay restitution have lower recidivism rat e s than those sentenced to incarceration or probation alone. A series of studies conducted among adult offenders in the 1970s, for example, found that offenders requ ired to pay restitution had lower recidivism rates than incarcerated offenders (Heinz, Galaway, and Hudson, 1977) and were less likely to reappear in court than offenders on standard probation (Hudson and Chesney, 1977). Another study, conducted in 1994 am ong


93 probationers in Alleghany County, Pennsylvania also found that restitution reduced recidivism. The study, however, found that the relationship was dependent on how much of the required payment was actually paid as well as the amount of time that the of fender had to pay the restitution (Outlaw and Ruback, 1999). This finding suggests that DPAs should be an effective tool for imposing restitution since the terms of DPAs require defendants to pay the entire monetary penalty and often provide them with paym ent options to pay the amount over an established period of time. Likewise, s everal of the major criticisms against requiring a restitution payment are moot when the restitution requirement is attached to a DPA. One of the top criticisms is that collectio n rates for restitution Bearden v. Georgia (1983) that imprisonment cannot be imposed as punishment for default of payment ( Outlaw and Ruback, 1999; Ruback and Bergstrom, 2006). When restitution is required in t he context of a DPA, however, this problem is averted. Under the terms of a DPA, a corporation that neglects to pay the restitution payment is eligible to face further criminal charges. Another common criticism is that courts and probation offices are typi cally too overworked to effectively monitor payments and ensure that victims receive their due (Ruback and Bergstrom, 2006; Sims, 2000). In contrast, ensure th at the terms of the agreement are properly carried out. When monetary fines are imposed for the purpose of punishment, there is mixed evidence regarding their deterrent value and some evidence to suggest that fines may not be therapeutic in that they ofte n fall on the shoulders of innocent corporate shareholders. When fines are specifically labeled for restitution, however, there is evidence to suggest that they have a preventive effect. Moreover, studies also suggest that restitution payments are therapeu tic for


94 both the victims and the offenders. Thus, fines marked for restitution rather than punishment alone are expected to contribute to DPAs being a therapeutic punishment option that reduces a Compliance Monitor s Independent monitors have been a feature of DPAs since Prudential Securities entered into the first DOJ corporate agreement in 1994 (Robinson et al., 2006). Most DPAs contain a clause requiring a compliance monitor to be assigned to the corporation and o ften these clauses have similar elements. Typically th e monitor cannot be a direct e mployee of the company. He or she is expected to have unrestricted communication with the federal government regarding the status of the compan y and the agreement, and t his communication is facilitated through the submission of routine reports at pre determined increments of time. Beyond these general commonalities however, the appointment, authority, and responsibility of the monitor can vary significantly from one DPA to office and the company h as very little say in the decision. In other instances the monitor may be already retained a s a consultant for the company, or he or she is selected through n egotiation and authority also depend to some degree on the nature o f the offense. In some DPAs the monitor is tasked with simply e nsuring that the terms of the a greement are carried out W hen the offender behavior is viewed as endemic to the corporate climate or when it is tied into the everyday business of the corporation, however, the monitor may oversee operational and personnel changes and even sit on the co rporate board (see, for example, U.S. Attorney for the Southern District of New York, 2005). T he therapeutic value of imposing an independent monitor on a corporation can be debated. On the one hand, t he monitor can be viewed as somethi ng akin to a probati on officer


95 who is able to provide supervision and one on one support for the offender. In theory, m onitors can be valuable for aiding in corporate reforms and ensuring that the terms of the DPA are met (United States Government Accountability Office, 2009 ). As Christie and Hanna, U.S. Attorneys, explain: whose prosecution is being deferred to go unsupervised during the deferral e on site reminders that compliance with the terms of a deferred prosecution agreement is mandatory, not optional (2006: 1054). The monitor is not only put in place to ensure that the terms of the agreement are met, but also to encourage progress, assi st i n the development of the too ls necessary for meeting the terms of the agreement, and facilitate regular communication between the corporation and the U.S. On the other hand, US Attorneys have been criticized for overuse of compliance m o nitors Critics argue that monitors are indiscriminately added into the terms o f most DPAs rather used for select companies for which they are most appropriate (Ford and Hess, 2009 ; Khanna and Dickson, 2007 ). Professors Ford and Hess (2009), of the Univer sity of British Columbia law school and University of Michigan School of Business, respectively, argue that monitors should be assigned in cases in which the seriousness of the offending warrants the cost of a monitor but should not be used in cases in whi ch the corporation has previously demonstrated an unwillingness to change. Monitors are also viewed as an opportunity for extreme prosecutorial discretion ( Khanna and Dickson, 2007; Paulson, 2007) and as potentially harmful government intervention into pri vately owned corporations. Monitors are often former judges or prosecutors who do not have the business background to fully understand the impact of any changes on the company (Spivack and Raman 2008) Since over half of the monitors appointed through DPA s


96 have prior ex perience working for the DOJ the independence of these monitors has also been called into question (United States Government Accountability Office, 2009). Many of these criticisms of corporate monitorships regarding poten tial overuse, cost to the corporation, and government intervention have been raised by business and legal scholars. An article dating back to 1982 by John Braithwaite, however, suggests that criminologists may not view these aspects of monitorship from a negative perspective In his article, Braithwaite proposes the concept of enforced self regulation as a means for curbin g corporate offending. While this model was proposed as a standard for all corporations, not just those that have already of fended, the basic approach is mi rrored in the use of corporate monitors. compliance standards. These standards would have to then be approved by federal regulators and private citizen groups. Once approved, responsible for ensuring that the compliance standards were met and also for regularly reporting to the federal government on how well the standards were met and if there were any instances in which mana gement over ruled the compliance directives. As Braithwaite explains, this model would be effective at regulating combines the versatility and flexibility of voluntary self regulation but without the inherent weakness of vol 1470). He also points out that another positive aspect of his model is that it would make it easier for government prosecutors to obtain corporate crime convictions against guilty individuals as the compliance reports would provide a deta iled record of any acts of noncompliance. The elements of and ideas behind enforced self regulation model are certainly present in the use of c ompliance monitors in DPAs. Under a DPA the corporation is being forced to self regulate by takin g on a monitor who essentially must become integrated into


97 the daily operations of the corporation as any other compliance staff would. T he compliance monitor must regularly report to the DOJ a federal regulatory agency, and the compliance standards put i n place are geared towards the specific issues that need to be addressed by the corporation. Additionally, as occurs with DPAs, Braithwaite suggests that the corporation should shoulder the cost of regulation. Unlike the critics of DPAs, he believes this t o be a positive thing, them In short, while some legal and business scholars may criticize certain aspects of the use of corporate monitors in DPAs, not all agree that these aspects are necessarily bad. Braithwaite, in fact, proposed a similar model to be applied universally to all corporations and in his model he highlighted as positives some of the very aspects that other scholars have critici zed. This suggests that whether monitors can reduce the likelihood of corporate recidivism is a bigger consideration in determining if they are a therapeutic or anti therapeutic component of punishment than many of the criticisms raised against them. To address the question of whether monitor s can be expected to reduce recidivism the most relevant body of research comes from assessment of a comparable type of criminal justice program intensive supervision probation (ISP) program s ISP programs involve o ffenders placed under close supervision with one probation officer assigned to only a small group of probationers at one time. ISP participants are typically required to undergo regular drug and alcohol testing, participate in work and treatment programs, and perform community service. The programs are designed to rehabilitate by addressing the underlying problems that contributed to the original offending behavior. An additional purpose of ISPs is to change


98 perceptions about the likelihood that criminal behavior will be detected and punished (Turner et al., 1992). Corporate monitorships arguably have a similar structure and similar goals to ISPs A monitor is assigned to work intensely with one corporation and the corporation must open up a ll records and documents for routine scrutiny and validity checks. Like the ISP programs, the corporate monitorships are also geared towards rehabilitation and altering the underlying corporate mentality that contributed to the offending. When corporate mo nitorships are understood as being similar to ISP programs there is some research to suggest that ISPs, and correspondingly, corporate monitorships, can have an impact on recidivism (Lipsey, 1999; Paparozzi and Demichele, 2000; Ross and Gendreau, 1980). F or starters, several studies have found that offenders view probation, in general, to be more punitive and burdensome than prison (Petersilia, 1990; Crouch, 1993). Among a sample of over 1,000 recently processed inmates being sent to a Texas penitentiary, two thirds selected a year in prison as preferable to 10 years on probation when presented with hypothetical punishment scenarios (Crouch, 1993). The study does not examine whether the aversion to probation equates to lower recidivism rates if probation we re a more widely used sentence. However, the findings suggest that reporting to a compliance monitor, like reporting to a probation officer, is likely considered a burdensome punishment, the threat of which may preclude at least some corporations from enga ging in illegal practices. Research on the effectiveness of ISP programs finds some support for reduced recidivism rates among program participants In a meta analysis of about 400 studies on rehabilitation, Lipsey (1999), found that intensive probation s upervision was significantly related to a 10 25% decrease in recidivism among juvenile offenders. Other studies have similarly found that


99 recidivism can be reduced by 20 30% when probation programs offer a combination of surveillance and treatment (Clear a nd Braga, 1995; Petersilia and Turner, 1993 ; Petersilia, 1997). Among the evaluations of ISP programs that do not find support for the notion that intensive supervision reduces recidivism, there is still some suggestion that the ISP approach to handling c orporate offenders (i.e. corporate monitorships) has merit. Two evaluations of ISP programs, conducted in the United States and England, found that when compared to a sample of regular probationers and incarcerated offenders, respectively, the ISP probatio ners had similar levels of recidivism The ISP probationers, however, had more technical violations, in which the probation officer discovered reversions in the underlying behaviors that were targeted for change through the program than new crimes commit ted (Brownlee, 1995; Turner et al., 1992) In the context of monitorships, the terms of the DPA are intended to change the corporate culture underlying the offending, and thus, the discovery any technical violations of a DPA would similarly be crucial to determining whether the corporation was merely paying lip service to change. Another ISP study conducted in the late 1980s in three California counties similarly fou nd tha t after one year in an ISP program, a quarter of serious offenders had no new incid ents, 40% had technical violations (a good number of which were drug violations), and a third had new arrests (Petersilia and Turner, 1990). While the outcome distributions were not significantly different for ISP offenders compared to regular probationers this finding can be explained to some degree by the fact that the ISP probationers were among the more serious offenders and the towards the idea of monitor s as agents for change was that, while participation in programs for employment, education and counseling was generall y low, in two of three counties ISP offenders

PAGE 100

100 were more likely to participate in counseling sessions than regular probationers. For all th ree sites, program participation was inversely related to recidivism. Thus, there is evidence from literature on ISP programs to suggest that the use of corporate compliance monitors could be a therapeutic component of DPAs in that their use is related to a reduced risk of recidivism. In sum, the non empirical analysis reveals a wide range of applicable research and theory in criminology, law and business that can be used to develop hypotheses about the expected contribution of each component to whether D PAs are a therapeutic or anti therapeutic punishment option. From the non empirical analyses the following two hypotheses are developed: 1. The most therapeutic DPAs, which reduce the likelihood of corporate recidivism, contain clauses emphasizing the impo rtance of cooperation without the waiver of employee rights, restitution to victims, a change in corporate structure and improved ethics programs, and the use of a corporate complia nce monitor; and 2. Corporate offenders that enter DPAs with US Attorneys h ave a lower likelihood of recidivism than corporate offenders that received regular criminal sanctioning. The case study and empirical analys is in Chapter 5 and Chapter 6 will test these two assumptions.

PAGE 101

101 C HAPTER 5 CASE STUDY: BRISTOL MYERS SQUIBB On June 15, 2005, Bristol Myers Squibb (BMS) Company a leading producer of pharmaceuticals and health care products, entered into a deferred prosecution agreement with the Office of the US Attorney for the District of New Jersey. The DPA allowed BMS to avoid cri o commit securities fraud in 2000 and 2001. Chapter 5 examines the events leading up to the signing of the DPA, the terms of the agreement, and the consequences of the agreement for BMS, bo th while the terms were in effect and in the years after it ended. BMS is an ide al candidate for a case study on the use DPAs in part because of the high profile, wel l documented nature of the case. The DPA, itself is a lengthy document with highly detail ed terms and descriptions of the case. On top of this, Christopher J. Christie, the former US Attorney for New Je rsey who handled the BMS case, co authored a 2006 a rticle in the American Criminal Law Review describing the process of implementing the BMS DP A and the rationale behind each of the key elements of the agreement. Additionally, BMS is a good candidate for a case study because the expired in June of 2007, which makes it possible to assess their well being and legal compliance whil e the agreement was in place, immediately after the agreement expired and five years after its expiration. Finally, BMS is a large, publicly traded company, and therefore, public documents that can be used to assess the financial health of the corporation and whether the DPA achieved therapeutic outcomes are readily available. Among criminologists who argue that the environment within a corporation ca n facilitate offending, it was no surprise that BMS became the subject of criminal investigation in the ear ly 2000s. From the 1990s to early 2000s, BMS developed a corporate culture in which meeting

PAGE 102

102 increasingly aggressive budget targets and sales goals was a mandatory condition of employment. First, i n 1994, BMS publically announced a goal to double earning an d sales by the year 2000. Achieving this goal required approximately 10% compound annual growth across the 6 year period down budget process, in which corporate executives set targets f or the company that corresponded with the necessary quarterly growth to meet or exceed goals. company had also an 24 stock prices, BMS announced an even more aggressive goal of doubl ing the 2000 sales and earnings in just five years (P R N ewswire, 2000) In order to ac hieve the 15% annual growth arnings and increasingly high levels of excess inventory at the wholesalers, was concealed from the SEC, security analysts, investors, external auditors, and the BMS Board of Directo rs. Furthermore, BMS was expected to maintain sufficient reserves to cover all Medicaid re bates associated with the sales. The Medicaid rebates were expected be paid a short time after the initial transaction, but r ather than pay the require d rebates for the excess inventory, BMS hid the excess inventory on the books and under a ccounted for the required rebates. Meanwhile, throughout 2000 and 2001, BMS continued to announce the meeting or exceeding of performance and budget goals, without revealing the information about the excess inventory and insufficient reserves. Since s uch i nformation would have allowed analysts to make

PAGE 103

103 re performance, BMS was knowingly and intentionall y mislead ing investors and the SEC throughout 2000 and 2001. After reporting sales of about $19 billi on and earnings around $5.2 billion in 2001, the company finally announced in April of 2002 that excess inventory at the warehouses would negatively effect BMS sales and reduce earnings by as much as half compared to the previous year. The news was not wel l received. had stepped down after pressure from the board ( Harris, 2002) and BMS stock prices plummeted from over $50 a share in mid March of 2002 to under $30 a share by the end of April 2002 The New Jersey enough evidence existed to be able to secure a grand jury indictment, prosecutors from the USAO were faced with the challenging decision of determining how to move forward with the case. According to then U.S. Attorney and Assistant U.S. Attorney, Christop her Christie and Robert Hanna (2006) several factors weighed on the decision to proceed with a deferred prosecution agreement. Some of these factors included: remedial actions taken by the corporation prior to any criminal charges brought against them ; th culpable individuals; the fact that BMS counsel was a former U.S. Attorney who had significant lack of prior experience with corporate DPAs (Christie and Hanna, 2006). Ultimately, though, as Christie and Hanna stated, Confident that our goals of notice to the corporate community, deterrence, full disclosure to the investing public, calibrated refo rm of a corrupted corporate culture, and restitution to victim shareholders could be achieved without visiting undue collateral consequences on Bristol Myers' business, its law abiding past and present employees and its current

PAGE 104

104 shareholders, we engaged in extensive discussions and negotiations over a three month period with Bristol Myers management, its board of directors and its counsel, ultimately arriving at the June 15, 2005 deferred prosecution agreement (2006: 1049). The BMS DPA was in place for two y ears, from June of 2005 to June of 2007. While it is more therapeutic effect on the corporation than other criminal sanctions would have, it is possible to dissect the terms of the DPA and use findings from the non empirical analysis to generate hypothesizes about the therapeutic or anti therapeutic nature of the punishment. Current corporate records from 2005 on then allow for comparison of the expected versus actua l outcomes of the DPA First, though, the terms of the BMS DPA are dissected. Terms of the BMS DPA Voluntary Disclosure and Cooperation The non empirical analysis of DPAs suggests that cooperation will improve the legitimacy afforded the law and the puni shment, reducing the likelihood of future offending. The non empirical analysis also reveals that the voluntary disclosure of information and cooperation between the corporation and the federal prosecutors may result in a shorter case resolution time and t hat this reduction in the time from offending to punishment may also reduce the likelihood of re offending. Similarly, cooperation was a central focus of the BMS DPA. For start ers, the decision to offer BMS a deferment rather than file formal criminal cha rges was made after the USAO for the District of New Jersey engaged in extensive consultation and discussion with BMS legal counsel Formulating the details of the agreement then involved an additional three months of negotiation between the BMS board, ma nagement and counsel and the USA O (Christie and Hanna, 2006) For all intents and purpose BMS was actively engaged in developing the terms of their punishment.

PAGE 105

105 Additionally, the BMS case was resolved in a relatively timely manner. Approximately 26 months passed from final signing of the DPA on June 15, 2005 In other words, because of the cooperation between the USAO and BMS, the process of investigating the crime s and carrying out th e punishment took just over two years to complete. It was further determined that two years was an appropriate length of time for the agreement long enough to ensure that the necessary changes in corporate functioning were real and permanent without puttin g undue burden on the corporation. Other terms of the DPA also stress ed the importance of cooperation and voluntary disclosure and are expected to increase the therapeutic nature of the punishment. For example, t he DPA specified that BMS would continue to into responsible individuals by making employees available to attorneys, identifying potential witnesses, and provi ding the USAO with all documents pertinent to the case BMS also agreed not to engage in future criminal conduct and waived all 6 th amendment rights to a speedy trial if the terms of the agreement were broken. The o ne potentially anti therapeutic aspect of the DPA was the requirement that BMS demonstrate cooperation by waiving attorney client and work product privileges for any documents that would be relevant and useful to the investigation b ut would typically be safeguarded because of these protections This waiver meant that if an individual employee of the corporation went prepared legal materials in anticipation of litigation, the details of the meetings and materials could be shared with the federal government despite the fact that they took place under the auspi ces of confidentiality.

PAGE 106

106 Business Reforms and Compliance Programs Findings from the non empirical analysis suggest that the most therapeutic corporate punishment includes both business reforms and the establishment or strengthening of the pliance and ethics program. In order for a corporation to be rehabilitated, the entire corporate environment that helped to facilitate the malfeasance must be transformed into an environment that discourages misconduct The corporation must establish a man agement structure that rewards positive behavior without accepting illegal activity as a means for achieving goals. Finally the strong management and positive environment must be supplemented with a serious compliance and ethics program that makes it clear to employees the types of behavior that are not acceptable. T he U S A O from the District of New Jersey recognized that the corporate culture within BMS in the late 90s and early 2000s facilitated deception and encouraged employees to use any means necessar y to meet aggressive sales and earnings goals. The BMS DPA was, therefore, designed to promote the improvements to the BMS corporate culture and business practices that were necessary for sustained change and total rehabilitation As Christie and Hanna (20 06) contemplating remedial actions under a deferred prosecution arrangement must be concerned and The DPA acknowledged that BMS had independently taken steps to change business practices prior to the signing of the agreement ( U.S. Attorney for the District of New Jersey, 2005) Federal prosecutors credited BMS with dismissing and replace culpable employees in senior management position s, including the Chief Financial Officer, Controller, and Pres ident of an implicated group within the corporation. The agreement noted that BMS had changed its budget process to allow input from lower level employees with a different perspective on the

PAGE 107

107 cor to day a ctivities than the upper level executive s It also noted that BMS had taken several major steps towards improving internal compliance, such as forming a business risk and disclosure group establishing a Chief Compliance Officer posi tion and a position for a securities and disclosure attorney, and developing a confidential employee hotline and email address for reporting suspicious activity. Because BMS had failed to disclose pertinent information regarding excess warehouse inventory to their shareholders, the DPA mandated that BMS take further steps to improve its transparency and reporting behavior shareholder reports include such information as warehouse inventory levels for the top 15 products sold by the company both in the United States and internationally, data on user demand accounts. Additionally, the agreement required quarterly meetings between BMS senior management and independent auditors to ensure that information was being shared and no parties could claim ignorance to the activities of the corporation. blic was also facilitated by a management structure that gave the company Chief Executive Officer (CEO) virtually unchecked power. Because the CEO was both the chief executive and the chairman of the b oard, the individual holding that position was not accountable to anyone else within the corpora tion In order to increase accountability the DPA required the splitting of the CEO and chairman roles into two separate positions. The DPA gave the board the authority to fil l the chairman position. It also required regular communication between the CEO and the chairman and gave the chairman a limited role in preparing public financial reports and statements.

PAGE 108

108 Finally, the DPA specified that BMS would develop a mandatory training and education program f or all BMS employees holding positions related to a cc ounting, financial reporting legal affairs, and upper management ( U.S. Attorney for the District of New Jersey, 2005 ). According to the agreement, the program w ould be geared towards improving integrity and professionalism within the corporation and creat ing a culture of openness, accountability, and compliance. The subjects covered in the training would range from disclosure obligations in federal securities laws to proper accounting practices and what to do if an improper or illegal accounting practice w as discovered. Monetary Penalties and R estitution Findings from the non empirical analysis also suggest that restitution payments are a therapeutic element of DPAs in that they allow the corporation to make amends for wrong doings. Likewise, the BMS DPA r equired the corporation to pay $300 million into a fund for aggrieved shareholders (U.S. Attorney for the District of New Jersey, 2005 ). This payment requirement was in addition to $539 million that BMS had previously paid to shareholders as a result of a related civil settlement. A ccording to Christie and Hanna (2006) the $839 million levied against BMS was not an arbitrary amount, but rather reflected a reasonable calculation of t he total shareholder losses resulting from the BMS rapid decline in stock p rices following the revelation of channel stuffing practices. Not onl y was the USAO able to use the DPA to ensure that victims received full restitution, the federal prosecutors were also cognizant of the need to balance deterrence with the potential coll ateral consequences of extreme monetary penalties. Thus, the DPA required restitution but no other fines or penalties. As Christie and Hanna explained the decision While punitive fines and penalties undoubtedly provide a measure of deterrence, we believe d that form of fiscal punishment was unnecessary to achieve our prosecutorial

PAGE 109

109 deterrent effect; a fine or penalty would punish the company's shareholders and exact a toll o n a company trying to get its corporate house in order (2006: 1059). In addition to the restitution payment, the BMS DPA did contain an additional but unprecedented financial requirement ; the endowment of a business ethics and corporate governance chair a t Seton Hall University School of Law in New Jersey. The agreement required that BMS fu nd a position for a professor to teach a minimum of one seminar a year on business ethics in which BMS executives and executives of other New Je rsey companies could enro ll ( U.S. Attorney for the District of New Jersey, 2005 ). Accord ing to Christie and Hanna BMS counsel proposed the idea of an endowed chair at a local college or university with the belief that this would be another step towards improving the BMS corporate culture ( 2006: 1058). Seton Hall was then selected as the recipient of the endowment because the other New Jersey public law school at Rutgers University already had an endowed business ethics chair. Critics, however, argued that the inclusion of the end owed chair requirement in the DPA represented an egregious abuse of prosecutorial discretion as Seton Hall University was the alma mater of then U.S. Attorney Christie. If BMS in fact perceived this requirement to be an arbitrary addition to the DPA based solely on the interests of the U.S. Attorney, the inclusion of the clause could be viewed as unfair, and therefore, would be an anti therapeutic component of the DPA. Compliance Monitor The non empirical analysis also revealed that compliance monitors sho uld be a therapeutic component of DPAs. Compliance monitors are intended to provide oversight, guidance, and serve as a visible reminder of th e terms of the DPA. The appointment and utilization of a compliance monitor was a central aspect of the BMS

PAGE 110

110 during the deferra The USAO worked with BMS to ensure t hat the corporation had input into who would be appointed compliance monitor. BMS had already retained federal judge and former U.S. Attorney, the Honorable Frederick B. Lacey, to serve as an indep endent advisor, so both sides agreed to expand his role to include the duties of a federal compliance monitor (Christie and Hanna, 2006). In the terms of the DPA, BMS agreed that the monitor would remain independent from the corporation but would have ful l access to corporate records and business proceedings. The Honorable Judge Lacey was expected to submit 400 to 500 page reports each quarter to the USAO, which would then be discussed at a quarterly meeting attended by the BMS CEO, Board irman, and general counsel, as well as the U.S. Attorney and Judge Lacey. Judge Lacey was additionally tasked w activities related to: compliance with federal securities laws; compliance with the terms of the civil litigation against BMS; information received through the employee hotline; new legal issues surfacing across any BMS operating entities; and potential legal or compliance issues that could develop in subsequent quarters. Beyond monitoring and re Lacey was also given authority to make recommendations regarding actions and changes that report submitted by the Monitor to the Office [of the U.S. Attorney for the District of New Jersey] unless BMS objects to the recommendation and the Office agrees that the adoption of the U.S. Attorney for the District of New Jersey, 2005: 4 ).

PAGE 111

111 While invasive intervention into the workings of the corporation could have anti therapeutic effects, the language of the DPA allowed BMS to challenge any recommendation that was considered unreasonable or harmful to the corporation. In fact, in 2006 BMS accepted one of the more expansive recommendations from Judge Lacey, w hich was to dismiss the BMS general counsel and the CEO who had initially retained him (Lattman, 2006). DPA Aftermath With a few exceptions, the BMS DPA contained elements gene rally expected to result in a therapeutic outcome s for the corporation as well as those tangentially impacted by the corporation and its punishment. The remainder of Chapter 5 examines the state of BMS immediately following the expiration of the DPA in 200 7 and five years later in 2012. In order to determine whether the punishment was therapeutic, the major questions to be addressed are: 1. Did the DPA rehabilitate the corporation? In other words, was BMS able to successfully carry out the terms of the agre ement and did the corporation subsequently refrain from engaging in similar or other criminal beh avior, and 2. Did innocent employees and shareholders suffer any collateral consequences resulting from the punishment? While collateral consequences could inv olve a wide range of harms, for the purpose of measuring and quantifying collateral consequences, the size and earnings of the corporation prior to and after the DPA are examined, as are the stock prices prior to and after the DPA. To the extent possible, these will be compared with the growth of similarly situation pharmaceutical corporations in an attempt to isolate the effects of the DPA from other market factors. Subsequent Criminal and Other T ransgressions Assessing whether BMS was able to carry out t he terms of the agreement is fairly straightforward. The day before the expiration the DOJ announced that the criminal charges against BMS would be dropped. In a public DOJ memo, then U.S.

PAGE 112

112 Myer s Squibb has made significant and transformational would help to make BMS a better company. The objective facts prove that goal has been Departmen t of Justice, 2007 a ). Thus, f successfully fulfilled the requirements of their punishment. Assessing whether BMS was able to refrain from similar or other corporate crimes is more complicated because it first requires a deter mination about what constitutes corporate crime. In the U.S, both the criminal and civil justice systems are regularly used to punish corporate offenders. However, t he imposition of criminal sanction s is arguably the clearest delineation between a corporat ion that has committed a crime and one that has not. Criminal sanctions typically carry harsher punishment, more stigma, and greater legal ramifications than their civil counterparts. Additionally, while criminal sanctions are traditionally viewed as a mea ns to right public w rongs, civil sanctions are viewed as a means to right wrongs against private individual s (Blackstone, 1962). From a practical perspective, the civil justice system is also often used to resolve dispute s between corporations such as tho se involving copyrights and patents, and disputes between employees and corporations, some of which may be outside of the realm of corporate crime. For these reasons, examining whether BMS had subsequent criminal prosecutions is the most apparent and clear cut approach to assessing the deterrent effect of the punishment. Ironically, less than a month before the USAO for New Jersey dropped the charges against BMS, BMS was facing unrelated, but somewhat similar criminal charges. On May 30, 2007 the corporati on entered a guilty plea to two counts of filing false statements to the Federal Trade Commission. The charges stemmed from a patent deal for the blood thinning drug, Plavix, in

PAGE 113

11 3 which BMS concealed and lied about an earlier agreement with the company Apote x, Inc., to not develop a generic version of the drug. The Assistant Attorney General overseeing the Antitrust fine (U.S. Department of Justice, 2007 b ) While i subsequent criminal sanctions that the DPA did not effectively convince the corporation of the value of transparency over deception, a review of the civil litigation against BMS in the years following the agreement, gives due dilig ence to the reality that criminal transgressions are often handled through the civil justice arena as well. A search of Federal civil court dockets through Lexis Nexis Courtlink reveals that i n the one year period from June 15, 2007 to June 15, 2008, there were 516 civil case dockets with BMS as a defendant the majority of which were product liability cases stemming from alleged negative side For practical purposes, only those cases which rose to a level of importa From fiscal year 2007 through fiscal year 2011, the majority of civil litigation involving BMS stemmed from the same dispute as the criminal case regarding the manufacturing of a generic form o f the drug Plavix. Many of the cases were intellectual property disputes over the patent for Plavix. Additionally, there were several class action antitrust and securities proceedings involving the cost of Plavix and market control over the drug, as well a s individual lawsuits claiming personal injuries sustained from use of the drug. BMS was also a defendant in several class action cases alleging that a faction of pharmaceutical companies had conspired to fix the prices of drugs internationally and inflate the Average Wholesale Prices (AWPs) within the U.S.. In the case of James Clayworth et al. v. Bristol Myers Squibb et al. which sought damages for alleged price fixing in the Canadian

PAGE 114

114 market, the judgments of the California District Court, California Cou rt of Appeals, and California Superior Court were all in favor of BMS and the other defendants (Bristol Myers Squibb [BMS], 2012) jury found the defendants, including BMS, not liable for fraudulent or negligent misrepresentation of prices (BMS, 2011) Another br oad category of civil cases involved environmental actions Fro m 2007 to 2011, BMS was implicated in several class action cases alleging that a New Brunswick, New Jersey BMS f acility had contaminated soil and groun dwater in surrounding areas, resulting in health and other complications for residents. The corporation was also several class action cases pertai ning to BMS waste disposal site in the New Brunswick area from the 1940 s 1960s. As of December 2011, all cases were still ongoing but not expected to have major financial implicatio ns for BMS (BMS 2012). Beyond these larger categories of civil litigation, BMS was the defendant in a variety of other types of litigation in the One of the biggest civil cases, during the period from 2007 to 2011, was against BMS and other manufacturers of a hormone replac ement therapy that was found to be linked to breast cancer. By December 31, 2010, BMS and fel low defendants had settled with 200 of the approximately 450 plaintiffs that had filed lawsuits pertaining to the hormone drug (BMS 2011) Two other civil cases involved qui tam allegations that BMS had provided kickbacks to pharmacies and other providers The first case involving Omincare, a provider of pharmacy services for the elderly, was voluntarily dismisse d by the plaintiff with prejudice (BMS, 2010 ) whi le the California case was in the early stages in 2011 (BMS, 2012)

PAGE 115

115 Because the nature of the civil cases are vastly different from the channel stuffing practices that assess ing whether the DPA or any other criminal sanctions should or could have deterred the events that resulted in civil litigation is next to impossible. Additionally is not feasible to ascertain whether there would have been additional criminal and civil charges against BMS had the corporation not been issued a DPA. Collateral C onsequences T wo of the negative potential consequences of corporate criminal sanctions are economic losses for innocent shareholders if stock prices decline and the loss of jobs for innocent employees if the corporation is forced to shut down particular operations or d ownsize because of fines or loss of business. The use of a DPA rather than formal criminal sanction is intended to reduce these collateral consequences. stock value and Fortune 500 corporate ranking are examined before and after the DPA. Because there are market and environmental factors that could also impact these measures of corporate e also examined during the same time period. Figure 5 1 shows the share prices at the close of the fiscal years 2002 through 2011 for BMS and three competitors AstraZ eneca, Eli Lilly, and Teva Pharmaceutical Industries Limited. Each of these companies i s relatively similar in size and revenue to BMS, which in 2011 generated $21.2 billion in revenue and employed approximately 27,000 people Inc., 2012b) In 2011, AstraZeneca employed about 57,000 persons with revenue of $33. 6 billion c., 2012a) Eli Lilly had 38,000 employees with $24.3 billion in revenue

PAGE 116

116 Inc., 2012c) and Teva had about 46,000 employees and generated $18.3 billion in reven ue ). 5 1. Share prices for Bristol M yers Squibb (BMS) and competitors at fiscal year close, 2002 2011 [ Annual share price data available from Hoover's I nc., 2012a, 2012b, 2012c, 2012d.] BMS shares closed lower at the end of 2005, the year the DPA was implemented, than the y had in the prev ious year ($22.98 in 2005 compared to $25.62 in 2004) 2012b) However, fiscal year 2006 closed with shares back up to $26.32 At the end of fiscal year 2008, the year after the DPA expired, BMS shares declined again to $23.25 but this decline is likely attributed more to the legal trouble surrounding the drug Plavix, than to the DPA. Overall, i n comparison to the other companies BMS stock prices closed at a relatively stable l evel across the ten year period. None of the other three companies displayed consistent trends in their stock prices to suggest that the

PAGE 117

117 DPA was impacting BMS share prices over normal market trends. While, the analysis does not take into consideration any within year fluctuations, it appears that BMS shareholders did not take a long term hit as a result of the DPA. Another indicator of the health and success of the corporation is their ranking among the Fortune 500 companies. Fortune 500 ranks U.S. compan ies according to their revenue minus excise taxes and f rom 2000 through 2005, BMS was consistently in the top 100 of the Fortune 500 companies, reaching as high as 78 th in 2000 (Fortune, 2000 2001, 2002, 2003, 2004, 2005 ). In 2005, the year the DPA was i mplemented, BMS was ranked 93 rd out of all public and private companies in terms of revenue (Fortune, 2005). However, by 2007 when the DPA was terminated BMS had dropped out of the top 100 to 129 th overall (Fortune, 2007). In 2012, BMS ranked 134 th overal l and sixth among the 12 Fortune 500 pharmaceutical companies (Fortune, 2012). This performance was not typical of all pharmaceutical companies. Eli Lilly, the only of the three comparable companies included in the Fortune 500 list, 1 consistently increased in ranking during this time period going from 170 th in 2000 to 152 nd in 2002 to 149 th in 2007 and up to 119 th place in 2012 (Fortune, 2000, 2002, 2007, 2012). implemented, th e decline in revenue is strongly tied to the fact that from 2000 to 2002, BMS was falsely inflating their revenue through channel stuffing practices and thus, falsely inflating their Fortune 500 ranking Additionally, in 2006, the Canadian drug maker, Apo tex, was able to get around patent restrictions for Plavix and flood the market with a generic version of the drug for several weeks, a move 1 AstraZeneca is primarily based in the United Kingdom and Teva is headquartered in Israel. Since neither company is US based they are not eligible for inclusion in the Fortune 500 listing

PAGE 118

118 Both the inflated revenue in the early 2000s and the losses suffered from patent and legal execution of the DPA. Finally, the question as to whether innocent employees lost their jobs due to the corporate DPA is al so difficult to assess. The number of employees within a company is closely re lated to revenue and sales. Therefore it is no surprise that BMS has done some down sizing in recent years. In 2000 and 2001, BMS had more than 50,000 employees (Fo rtune, 2000, 2001), but by the end of 2002, after the channel surfing and false reporting activities had been disclosed, that number dropped to 46,000 (Fortune, 2002). At the end of 2005, the year the DPA was implemented, BMS had 43,000 employees (BMS, 2006) and by the end of the term of the agreement in 2007, that number dropped even more to 42,000 (BMS, 2008). From 2007 to the end of fiscal year 2011, the number of BMS employees was drastically reduced to 27,000. However, the loss of jobs cannot be attributed to the DPA. In the 2007 annual reduction program that includes workforce reductions and rationalization of some facilities. Specific productivity goals include reducing total headcount by approximately 10 percent The rationale behind the launch of the BMS Productivity Transformation Initiative (PTI) was unrelated to the 2005 corporate DPA. Instead, the PTI was implemented to improve the changing business environment and to take advantage of the diverse opportunities of the Thus, while there were some workforce reductions during the time the DPA was in place, the majority of reductions was unconnected to the punishment and likely would have taken place regardless of the type of sanctioning against the corporation.

PAGE 119

119 Case Study Conclusion I t is ultimately not possible to assess whether the DPA was more or less th erapeutic for BMS and those touched by the corporation than formal cri minal sanctions would have been. BMS was convicted of criminal charges after the DPA was implemented, which suggests that the DPA did not fully meet its goal of rehabilitating a corrupt corporate culture. However, whether formal criminal sanctions would have done a better job at preventi ng BMS from with holding information in the Plav ix cases is unknown. In terms of the collateral consequences of the DPA, the fact that the corporation rem ained relatively stable in terms of share prices, Fortune 500 ranking, and number of employees during the time the DPA was in effect, suggests that the DPA was a therapeutic punishment option that did not cause undue harm to innocent employees and sharehol ders. Unfortunately, the long term implications of the DPA for the well being of the corporation and those associated with it, are difficult to assess because they are confounded by the subsequent criminal sanctions and other market forces. That said, the re is reason to question whether the government would have been able to build a strong enough case ag ainst BMS to get a conviction had they moved forward with criminal charges rather than the DPA In 2010, five years after they were first indicted for frau d and conspiracy related to the channel stuffing case, former BMS executives, Frederick Schiff and attorney, David Zornow, the offer of a DPA came after the government w as unable demonstrate criminal wrongdoing (Corporate Crime Reporter, 2010). In 2008, the U.S. District Court in Newark, New Jersey put the case on hold, ruling that among other issues the prosecutors in the Schiff case had not followed prop er legal procedu re for introducing expert testimony (Voreacus, 2010). In 2010, t he U.S. Third Circuit Court of Appeals in Philadelphia upheld the District

PAGE 120

120 d in a game of musical chairs with the pu In other words, after five grueling and taxing years of attempting to bring criminal charges against the individuals at the center of the BMS case, prosecutors ended up at the same place they were five yea rs earlier with the corporation offering a deferred prosecution agreement. This is not to say that not harm the market and shareholders and that restitution and rehabilitation of the corporate culture were not ju stified and necessary just that the burden of proof and the challenges of corporate prosecutions are often great. It suggests that by offering BMS a DPA from the beginning, at a minimum, US Attorney Christie avoided the loss of countless hours and money p ursuing a case that, by its nature, may have been difficult to win.

PAGE 121

121 CHAPTER 6 EMPIRICAL ANALYSIS: DATA AND METHODOLOGY C urrent F ocus To date, DPAs have not received the type of rigorous, empirical examination that one would expect and desire for a crimin al justic e policy in practice for over two decades. A report from the U.S. Government Accountability Office (GAO) in 2009 titled, Corporate Crime: DOJ Has Taken Steps to Better Track Its Use of Deferred and Non Prosecution Agreements, but Should Evaluate E ffectiveness makes th e point that the DOJ needs to develop performance measures that can be used to assess the effectiveness of DPAs an assessment that is not currently undertaken. The academic community also has an obligation to contribute to the conve rsations and research on the positive and negative consequences of criminal justice policies pertaining to punishment and rehabilitation. While a number of legal scholars have debated the merits and constitutionality of various components of DPAs ( see, for example, Couden, 2005; Duggin, 2008; Finder and McConnell, 2006; Garrett, 2007; Horowitz and Oliver, 2006; Weissman n 2007) the criminology community has remained relatively quiet on this topic. This study presents the first systematic application of a c riminological theory to s use of DPAs to punish corporate offenders Using the framework of the therapeutic jurisprudence per spective and hypotheses developed through the non empirical analysis, the research will examine whether deferment is a the rapeutic or anti therapeutic punishment option for corpor ate offenders. The two hypotheses to be empirical ly tested are: Hypothesis 1 DPAs that contain clauses requiring cooperation without a waiver of employee rights, restitution to victims, a change in corporate structure, an improved compliance and ethics program, and the retention of an independent compliance monitor are the most therapeutic and are related to a reduced likelihood of corporate recidivism ; and Hypothesis 2. Corporate offenders that ente red

PAGE 122

122 DPAs are more likely to be rehabilitated than corporate offenders that were formally sanctioned and as such, have a lower likelihood of recidivism. T he therapeutic jurisprudence perspective helps to identify whether particular components of DPAs are l ikely to contribute to deferment being a therapeutic punishment option in which corporate offenders are deterred from future criminal behavior (or, at least, criminal behavior is not detected by legal officials) and collateral consequences are reduced. Ba sed on findings from the non empirical analysis and as noted in Hypothesis 1 it is anticipated that the most therapeutic DPAs will include the following components: a strong emphasis on cooperation without requiring a corporation to waive attorney c lient or work order protections; a requirement for some type of restitution; a requirement for a change in the corporate structure and management as well as improvements to corporate ethics and compliance programs; and a corporate compliance monitor In general, the non empirical analysis also suggests that DPAs should result in lower levels of recidivism than traditional criminal sanctions. Whether the corporation is rehabilitated is the most important measure of the therapeutic nature of the punishment because if the practice fails to prevent future offending to the same or better degree than criminal sanctions, it is fundamentally anti therapeutic. Data One explanation for why similar assessments of DPAs have not been conducted within the criminological realm is that the DOJ does not currently have a central repository where DPAs are tracked and documented (see also, United States Government Accountability Office, 2009). In 2009, the DOJ implemented a system to internally track new DPAs However, the lack of of ficial tracking and documentation during the first decade and a half of DPA use makes identification of all of the early agreements difficult. A Freedom of Information Act (FOIA) request was submitted to the DOJ in November of 2009, requesting that the Dep artment provide copies of all

PAGE 123

123 corporate DP As valid from 1992 to 2009. The request was denied on grounds that the documents could only be provided if the request include d the specific names, dates, and locations of the DPAs. This response corroborates the s uggestion that the DOJ does not have a means of readily identifying all of the agreements that it has entered into over the years. Because of the lack of previously compiled data on DPAs, for the purpose of this research, a database was created from the te xt of all publically available DPAs initiated between 1992 and 2009. Four primary sources were used to identify and ensure a complete enumeration of the DPAs that were commenced during this time period and also to verify key details of the DPAs. First, i n a 2006 article published in the St. Louis University Law Journal, Lawrence Fi nder and Ryan McConnell detail the characteristics of 70 DPAs and NPAs drafted between 1994 and 2006. ements of each agreement, including whether they contain provisions for a monitor, business reforms, privilege waivers, 6 th amendment waivers, and fines (Finder and McConnell, 2006: 36 39). If an agreement contains any of these provisions, the authors pro vide brief descriptions of the provisions. For example, if a fine was included as par t of a DPA, the fine amount and designa ted recipient are noted in the m atrix. The authors also provide links to the online location of the majority of the 70 examined DPAs Another source with a large compilation of DPAs and NPAs is the Corporate Crime Reporter a legal newsletter in its 24 th year of production, which focuses on corporate crime and corruption (available at www. corporatecrimereporter .com ) An article written by the editor in 2005 summarizes 34 of these agreements (17 NPAs, 17 DPAs) and also provides links to their online locations. Though the majority of cases detailed by the Corporate Crime Reporter ma tch

PAGE 124

124 between the two sources. Though Finder and McConnell purport to cover all non antitrust DPAs and NPAs prior to 2006, there are two NPAs captured by the Corporate C rime Reporter that did not make it into the matrix; one between the U.S. Attorney for the Southern District of New Royal Dutch Petroleum Company and The Shell Tran sport and Trading Company (Shell Oil) Additionally, the Corporate Crime Reporter categorized the 2004 agreement between the DOJ NPA. The third source is the afore mentioned 2009 GAO report on the need for evaluating the effectiveness of DPAs. For the report, GAO official s collected data on all of the DPAs and NPAs negotiated by DOJ prosecutors from 1993 through the end of fiscal year 2009. To compile the list they relied on agreements provided by DOJ to the House Judiciary Committee in 2008 as well as independent web searches and press releases, the Corporate Crime Reporter and several other legal websites (Government Accountability Office, 2009: 5). The report pro vides a breakdown of the number of agreements entered into each year, as well as the division of DOJ or the initiating the agreement. It also compares the number of DPAs entered into from 2004 to 2009 with the total number of corpora te criminal cases prosecuted by DOJ and the U.S. Att orneys during this time period. Unfortunately, though, the GAO report does not identify the corporate parties involved in the DPAs or provide any additional details about the nature of the DPAs and their provisions. The final source, the most comprehensive of the four, is a library collection of federal organization plea agreements developed by Professor Brandon Garrett and Jon Ashley from the University Of Virginia School Of La w (Garrett and Ashley, n.d. a ). The collection is organized

PAGE 125

125 responsible for negotiati ng it, the month and year of the agreement, and whether it was a deferred or non prosecution agreement. Each en try also contains links to the text of the plea agreement where available and also to any related press releases. With 257 companies identified as entering most c omplete database of corporate DPAs. W hile there are multiple sources to assist with the identification of all DPAs, no existing data source co ntains the detailed information about the content of the agreements necessary answer the empirical research quest ions. Therefore, prior to any assessment a data file had to be developed specifically to address the questions at hand. the Depart ment of Justice entered into 163 DPAs and NPAs with offendin g corporations across the country from 1992 through 2009 Office and an individual corporation. In some circumstances, an agreement was signed by both a parent company and the offen ding subsidiary, but if the agreements were identical, they were treated as one agreeme nt. Of the 163 DPAs and NPAs about 20 agreements were not publicly available or the text of the agreement was not accessible through any electronic sources For a numbe r of these unavailable agreements limited information about the agreement and could be obtained through press release Google news searches. To create the data file used for the empirical assessment of the outco me of DPAs, attempts were made to locate all 16 3 NPAs and DPAs and to the extent possible, collect details about the characteristic and components of these agreements, as well as information about the corporation

PAGE 126

126 and the offense to code into SPSS. To ensu re that the coding was accurate and elements of the DPAs were correctly identified, the data file was compared to the more limited data on DPAs complied in the GAO report tables (2009), and a spreadsheet containing details of DPAs drafted by Professors Gar rett and Ash l ey (n.d. a ) For example, t he number of agreements contained in GAO report. The GAO report identified 152 agreements entered into from 2003 throug h September of 2009 and noted that the U.S. Attorney for the Southern District of New York had that most agreements during the period with 23. It was then confirmed that the data file contained 152 agreements during the period, plus an additional two NPAs that were sealed, and that the Southern District of New York was coded as the office responsible for 23 of these agreements. Similarly, the spreadsheet compiled by Professors Garrett and Ashley (n.d. a ) could be used to verify in formation on whether the agr eement was a DPA or an NPA; whether the program, pre agreement remedial measures, or privilege waivers; fine and restitution amounts; and the length of the agreeme nt. A sample of 20 corporations was randomly selected from the data file and checked against the data compiled by Garrett and Ashley to ensure that the agreements had been properly coded in terms of these key elements. The coding of the DPAs for the date f ile was also consistent with the information reco r ded by Garrett and Ashley (n.d. a ). For each of the 163 corporations that entered a DPA from 1992 to 2009 additional data on any subsequent or prior criminal charges against the corporation were coll e cted f rom federal court records using Courtlin k from LexisNexis Because corporations change names or may have parent or subsidiary companies that are not easily identifiable, the history of each corporation was examined using a Dunn & Bradstreet d atabase Hoove r

PAGE 127

127 information on companies including their history, size, and affiliations. Court records for any parent or subsidiary companies of the original corporation were also searched and any criminal charges against these affiliated cor porations were also included in the database. Additionally, in order to compare corporations that entered DPAs to corporations that were formally indicted and prosecuted a sample of other criminally sanctioned corporate offenders was selected for each y ear. Because the goal was to match the two sets of corporations by the offenses committed, prior to beginning the selection, an assess ment of the types of offenses committed by corporat ions with DPAs was conducted, by year, from 1992 to 2009. T o then match the DPA corporations with corporations that were indicted and sanctioned two sources were used: LexisNexis Courtlink and a database of Federal Organization Plea Agreements also compiled by Professor Brandon Garrett and Jon Ashley at the University of Vir ginia School of Law (Garrett and Ashley, n.d.b ). In Courtlink, a search was conducted on all Federal District their name for each year. The search returned lists of organizational defendants, the charges against them, and links to the related federal court docket and case disposition information The plea agreement database which provided data on federal corporate plea agreements entered into from 2000 and on, al lowed for filtering the agreements by year and criminal charges in order to select a sample of corporations that committed similar offenses as the corporations that received DPAs. Using these two sources of data, formally sanctioned corporations with simi lar charges as the corporations that received DPAs each year were randomly selected and added to the database Th e database contains a total of 128 criminally sanctioned corporations. For each year, t he number of formally charged corporations mat ched the n umber of corporations that received

PAGE 128

128 DPAs, up to a maximum of 25 1 The year 2007 was the only year in which the number o f corporations with publicly available DPAs exceeded the number of formally charge corporations in the database (see Table 7 1 for detail on the number of corporations that received DPAs by year) For each corporation that was formally indicted and criminally sanctioned information on the t ype and amount of the sanction was collected through case disposition documents Data on prior and su bsequent criminal charges were also collected for these corporations using the same approach used for collecting data on additional charges among DPA corporation s Hypothesis 1 Variables Depende nt V ariables In addition to the difficulty of identifying all of the DPAs, another challenge to the empirical analysis revolved around the measurement of the effectiveness and therapeutic nature of deferment as a corporate punishment option In the previously mentioned GAO report which recommended that the DOJ begin to measure the effectiveness of DPA, GAO suggested that effectiveness be measured by : 1. whether the company recidivated during and after the period of the agreement, and 2. whether t he company successfully fulfilled the terms and require ments of the agre ement. To this suggestion, the DOJ countered in a written respo nse published in the GAO report that after the terms of the DPA are completed. Therefore, recidivism could only be defined by whether the company engaged in another criminal offense during the period of the agreement but not beyond. The DOJ further argued against the suggestion that recidivism be a measure of 1 The number of formally sanctioned corporations was capped at 25 due to the difficulties in identifying corporations with similar changes as the corporations that received DPAs. The capping was done to avoid including corporations that were formally sa nctioned for very different types of offenses than the corporations with DPAs, which could impact the analyses.

PAGE 129

129 effectiveness on the grounds that in a large corporation there is the possibility that another employee could commit an unrelated offense. This unrelated offense would suggest that the DPA was ineffective at preventing recidivism regardless of whether the DPA effectively deterred the corporation from committing the same offense for which it was originally punished. While the first argument explains why the DOJ does not currently measure the effectiveness of DPAs in the long term, the second argumen t does not provide a valid justification for not using recidivism as a measure of effectiveness. P revailing research suggests that the climate in the workplace that encourages or discourages employees from engaging in malfeasance (Ferrell et al., 1998; Ford and Hess, 2007; Paine, 1994) If DPAs are effective at reformin g the environment that facilitated the initial offending, they should have a similar impact on all type of offending Moreover, among individual offenders, recidivism is not measured by whether the offender committed the same offense following sanctioning, but whether the offender committed any subsequent offense. There is nothing to suggest that recidivism should be measured differently for corporations than for individuals. In other words, t he argument that recidivism should only be measured by whether th e corporation commits the same offense does not hold much weight. DPA the dependent variable for the analysis is a measure of whether or not the corporation was the subject of additional federal criminal action during o r after the DPA was in effect. For the purpose of this analysis, whether a corporation recidivates is essential to determining whether the corporation was rehabilitated and the DPA was therapeutic or anti therapeutic The therapeutic jurisprudence perspective is about considering the ability of one punishment option to rehabilitate the offender and the victims compared to another punishment option. By nature,

PAGE 130

130 DPAs are therapeutic for victims and fo r society because they ci r cumvent some of the more serio us collateral consequences associated with criminal sanctions. The question remains whether DPAs are also able to prevent of future offending an other important aspect of rehabilitation. The dependent variable is a dicho tom ous measure of whether the corporation, its parent company, or one of its subsidiaries was charged with a crime in a federal district court any time following the signing of the DPA (CRIMINAL) Data on whether the corporation experienced any subsequent ac tion were gathered from federal court records using the Lexis Nexis Courtlink system. Courtlink allows for searching and accessing federal court dockets and filings by litigant name Prior to searching for any corporate court records, the Dun and Bradstree t database, Hoover for Business, was used to iden tify whether the corporation had affiliated parent or subsidiary companies or operated under a different name at any po int. Court record searches also include d any affiliated companies or name changes where applicable criminal action s taken against the parent company or a subsidiary company, are also account ed for in the analysis. For companies that entered a DPA in the early 1990s compared to the late 2000s there is a greater likelihood of recidivism simply on the basis of the longer time frame within which the corporation could have recidivated. For the assessment of the degree to which each of the components reduce the likelihood of recidivism, the year of the DPA will be taken into account in the analysis In the assessment of whether DPAs result in an overall lower level of recidivism than criminal sanctions, year should not impact the analysis as the corporations with DPAs and the corporations with criminal sanctions are matched by year. Independent Var iables The independent variables for the first part of the analysis pertain to the various elements of DPAs that were examined in the non empirical analysis.

PAGE 131

131 Compliance Monitor The first independent variable is a dichotomous measure o f whether the terms o f the DPA require that the corporation is assigned a compliance monitor (MONITOR). The variable is coded dichotomously such that a the DPA required the corporation to retain an independent compliance monitor it did not. Business Reforms and Compliance Programs The second series of independent variable s relates to the type s of court required business reforms that were conditions of the agreement. Seven different types of business reforms are included as covariates in the m odel. The first m easure s whether the DPA included a statement regarding business reforms put in to place prior to the DPA (REFORM). The second measures whether the agreement required the dismissal of any employees directly involved in the matter or the repl acement of particular corporate executives (DISMISS). The third relates to a DPA requirement that the corporation create new leadership or oversight positions wi thin the corporation (POSITION). The fourth business reform variable measures whether the DPA m andated additional job training req uirements for employees (TRAIN) T he fifth variable in the group pertains to a requirement that the corporation or the corpo board be restructured to provide greater accountability (RESTRUCTR) and fi nally, the sixth covariate of the group measures whether the DPA required a revision of policies related to a particular busin ess transaction or a restriction on the ability to engage in a particular type of transaction (BUSREFORM ) E ach of these variables is requirement was Another covariate included in the analysis COMPLY, measure s whether the DPA requires the institution of or a change to the COMPLY is

PAGE 132

132 does include such a requirement. Monetary Penalties and Restitution The next set of independent variable s measure s whether a monetary fine or restitution payme nt is required by the DPA The first variable, FINE, measures whether the terms of the DPA require the corporation t o make any type of monetary payment, including a fine, forfeitur e, or restitution The s econd variable, RESTITUTION, measures whether the DPA required a that the D Another variable, TOTALFINE, which measure the total monetary payment required by the DPA, is also included in the analysis but is used as a proxy measure of the seriousne ss of the offense and thus, is discussed in more detail in the section about the control variables. Voluntary Disclosure and Cooperation Cooperation is a central assumption of all DPAs. Therefore, there is not an independent variable included in the analy sis to specifically measure whether the corporation has agreed to cooperate with prosecutors. However, there are four variables that relate to the extent of the a measure of whether the DPA specifies that the corporation will use best efforts to ensure that individual employees provide information and testimony; DISCLOSURE, a measure of whether the corporation has agreed to voluntary disclosure of any future suspic ious activity; and ATTYWAIV E and WORKWAIV E which measure whether the DPA contains two of the more controversial components, a requirement to waive attorney client privileges and /or a requirement to waive work product privileges. Each of these four variabl es are coded dichotomously such

PAGE 133

133 included in the terms of the agreement. Finally, the last of the key independent variables, TIMEFROMCRIME is a measure of the num ber of months that elapsed between the estimated most recent date of the criminal activity and the date of the signing of the DPA. The variable was computed by indictment or criminal information from the date on which the agreement was signed. Because t he celerity with which a punishment follows a crime is indirectly related to cooperation and is presumed to be related to deterrence, the length of time that elapsed between the offense and the agreement is an important aspect of consideration in assessing the therapeutic nature of DPAs Controls In addition to the key covariates that represent the major components of DPAs, the model also controls for the type of agreement whether it was a DPA or an NPA (AGREETYPE); the number of months from the signing of the DPA to the current day to account for the fact that a corporation that entered a DPA in 2000 had more time to reoffend than a corporation that ente red a DPA in 2009 (TIMEFROMDPA); and whether the corporation had any federal criminal charges prior to entering the DPA (PRIORCRIME). The coding of these variables is included in Table 7 5. As mentioned previously, the model also controls for the total mon etary payment required by the agreement which is included as a proxy for the seriousness of the offen se (TOTALFINE). In general, under sentencing guidelines more serious offenses incur higher fines so the amount of the fine is a reasonable proxy for the s everity of the offense. However because DPAs are designed to consider the collateral consequences of punishment, a corporation may received a reduced fine on the grounds that requiring the maximum monetary payment would have

PAGE 134

134 detrimental effects on innocen t employees and shareholders. That said, TOTALFINE is arguably a more reasonable proxy for offense seriousness than other variables in the date file such as the type of offense, which could contain offenses resulting in a wide range of harms within the sam e category. Hypothesis 2 Variables Fewer variables are required for the model testing the second hypothesis that corporate offenders that entered DPAs are better rehabilitated, and thus, have a lower likelihood of recidivism than corporations that were pro secuted and criminally sanctioned. The model included one dependent and one independent variable and three control variables. The dependent variable is the same dependent variable as used in the model testing Hypothesis 1 (CRIMINAL) Again, CRIMINAL is a m easure of whether the corporation, its parent company, or a subsidiary had a federal criminal charge subsequent to the DPA or criminal sanction of interest. Arguably, whether a corporation has been rehabilitated is the most impor tant measure of the therape utic nature of DPAs. If the use of DPAs results in higher rates of recidivism than formal criminal sanctions, the practice is fundamentally anti therapeutic regardless of the other collateral consequences T he key independent variable for the test of Hyp othesis 2 is a measure of whether the case against the corporation was settled through a DPA or through a guilty verdict or judgment or a plea agreement (DISPOSITION). Finally, t he three control variables also included i n the model are TIMEFROMCRIME, TOTAL FINE, and PRIORCRIME. The coding and distribution of these variables is shown in Table 7 10. Analytic Plan This research effort addresses two main questions. First, which components of DPAs are therapeutic in that they are related t o a rehabilitated corpo ration (i.e. one that does not

PAGE 135

135 recidivate) ? Second, which punishment option results in a lower relative risk of recidivism: formal criminal sanctions or DPAs? Given the dichotomous coding of the outcome variable in both questions did the corporation re cidivate or not logistic regression is used to examine the relationship between the components of the punishment and the outcome The first logistic model will test the Hypothesis 1 predictions about the probability that a corporation will recidivate af ter entering into a DPA with certain components The second logistic regression model will test the Hypothesis 2 assumption that DPAs are a more therapeutic punishment option than for mal criminal sanctions and that c orporations which entered a DPA have a l ower likelihood of recidivism than corporation s that were formal ly convicted in a federal criminal court

PAGE 136

136 CHAPTER 7 EMPIRICAL ANALYSIS: FINDINGS General Descriptive Statistics From 1992 through 2009, the federal government entered into 163 deferred and n on prosecution agreements with alleged corporate offenders. As shown in Table 7 1, the use of deferments for corporate offenders increased in recent years, with nearly 80% of the agreements instituted from 2005 through 2009. The peak year for corporate def erments was 2007, during which 42 agreements were signed. Table 7 1. Number and availability of corporate non prosecution and deferred prosecution agreements, by year, 1992 2009 Number of agreements Total number Total number publicly available* Year Non prosecution Deferred prosecution 1992 1 0 1 0 1993 2 1 3 2 1994 1 1 2 2 1995 1 0 1 1 1996 1 1 2 1 1998 1 1 2 1 1999 1 0 1 1 2000 1 0 1 1 2001 2 1 3 2 2002 2 1 3 2 2003 1 4 5 5 2004 4 5 9 9 2005 9 6 15 14 2006 13 11 24 22 2007 20 22 42 40 2008 17 12 29 22 2009 11 9 20 18 Total 88 75 163 143 *Of the 20 agreements that were sealed or otherwise unable to be located, one was a deferred prose cution agreement and the remainder were non prosecution agreements.

PAGE 137

137 The agreements were fairly evenly split between deferred prosecution, in which charges were filed but not pursued, and non prosecution agreements, in which criminal charges were not for mally filed. About 54% of the agreements did not involve the filing of any charges, while in 46% charges were filed but further prosecution was deferred for the length of the agreement. Of the 163 agreements identified from the 1992 through 2009 period, t he text of the agreement was publicly available for 143. While some details for the 20 sealed or otherwise unavailable agreements could often be pieced together from press releases and news articles the remainder of the analysis focuses on the 143 agreemen ts for which complete information was available. 1 As shown in Table 7 2, the DOJ Criminal Division had primary responsibility for the drafting and monitoring of the largest proportion of DPAs entered into from 1993 through 2009. The Criminal Division handl the Southern District of New York, which administered 16% of the total DPAs, and the U.S. ices covering the United States and Puerto Rico, 33 offices entered into one or more corporate DPA from 1993 through 2009. Additionally, the DOJ Criminal Division, Antitrust Division, Tax Division, and Civil Rights Division each handled at least one DPA du ring the period. Table 7 3 provides a breakdown of the industries represented by the 143 corporations that entered into publicly available DPAs from 1993 through 2009. The classification of corporations is based on the standard North American Industry Clas sification System (NAICS). 1 Of the 20 sealed or unavailable agreements, 19 were NPAs. Two of the corporations (10%) had subsequent criminal charges.

PAGE 138

138 Table 7 2. U.S. Attorneys' offices and Department of Justice litigating divisions overseeing more than one deferred (DPA) or non prosecution agreement (NPA) from 1992 through 2009 Office or division DPAs and NPAs administered Number Percent of total Total 143 100 % DOJ Criminal Division 36 25 New York Southern District 23 16 Massachusetts 12 8 New York Eastern District 8 6 New Jersey 7 5 California Central District 5 3 Vi r ginia Western District 4 3 California Northern District 3 2 DOJ Antitrust Division 3 2 Texas Southern District 3 2 Connecticut 2 1 Illinois Southern District 2 1 Mississippi Southern D istrict 2 1 Missouri Eastern District 2 1 Fl orida Northern District 2 1 Alabama Northern District 2 1 Rhode Island 2 1 Pennsylvania Western District 2 1 Georgia Northern District 2 1 District of Columbia 2 1 New York Northern District 2 1 DOJ Tax Division 2 1 All additional offices 15 10 Over a third of the corporations that entered into a DPA were from the manufacturing sector. Within manufacturing, a bout 43% of the DPAs (15% of the total DPAs) involved corporations that manufactured pharmaceutic als, medicines, and medical supplies. Nearly a quarter of the DPAs were entered into with financial services and insurance companies and 10 %

PAGE 139

139 involved corporations that provide professional, scientific or technical services, including accounting and tax pr eparation, legal, and software support services. Table 7 3. Industries represented by corporations that entered into deferred (DPA) and non prosecution (NPA) agreements from 1993 through 2009 Corporations that entered into DPAs and NPAs Industry Number Percent of total Total 143 100 % Manufacturing 49 34 Pharmaceutical and medicine m anufacturing 21 15 Transportation equipment m anufacturing 7 5 Food m anufacturing 5 3 Computer and electronics m anufacturing 4 3 Petroleum and coal pr oducts m anufacturing 2 1 Finance and i nsurance 31 22 Professional, s cientific and technical s ervices 14 10 Accounting, tax preparation, b ookkeeping and p ayroll s ervices 5 3 Mining, q uarrying and o il and g as e xtraction 8 6 Information 7 5 Utili ties 6 4 Retail t rade 6 4 Arts, en tertainment, and r ecreation 5 3 Transportation and w arehousing 4 3 Health care and social a ssistance 3 2 Construction 3 2 Administrative and support and waste management and remediation s ervices 2 1 Wholesale trade 1 1 Real Estate and rental and l easing 1 1 Accommodation and food s ervices 1 1 Educational s ervices 1 1 Other services (except p ublic a dministration) 1 1

PAGE 140

140 Finally, the last table (Table 7 4) that provides a general overview of the charact eristics of DPAs and the corporations that entered i nto them from 1993 through 2009 shows general categories of the alleged offenses for which the corporations were or would have been charged. The offense type categories are based on the nature of the offe nse rather than the actual offense charged, since for corporations that entered NPAs there were no formal charges filed. The three most common types of offenses, resulting in over half of the agreements during the period, were securities fraud, violations of the Foreign Anti Corrupt Practices Act, and fraud against the gov ernment or government programs. Table 7 4. Offenses committed by corporations that entered into deferred (DPA) or non prosecution (NPA) agreements from 1992 through 2009 Number of a gr eements Percent of total Offense NPA DPA Total Total 69 74 143 100 % Securities f raud 18 13 31 22 Violations of Foreign Anti Corrupt Practices Act 14 15 29 20 Fraud Against the government or government p rograms 6 13 19 13 Facil itating m oney laundering or internet g ambling 5 8 13 9 Violations of anti kickback s tatutes 7 5 12 8 Other offenses* 5 6 11 8 Labor and employment v iolations 5 3 8 6 Violations of the Food, Drug, and Cosmetics Act 4 2 6 4 Import an d export v iolations 1 4 5 3 Antitrust and anti competitive c onduct 4 0 4 3 Environmental v iolations 0 3 3 2 Health care fraud 0 2 2 1 *Includes offenses such as mail and wire fraud, obstruction of justice, aiding and abetting, and o ther frauds. While most of the offenses were fairly evenly divided in terms of whether the corporation was offered a deferred or non prosecution agreement there are some notable exceptions. Among corporations that committed fraud against the government the number that received a DPA was

PAGE 141

141 more than double the number that entered into an NPA. A greater number of corporations that facilitated money laundering or failed to report suspicious monetary transaction s enter ed into DPAs than NPAs. A ll four of the corporations that committed antitrust and entered into a prosecution agreement entered into n on prosecution agreements, while all three of the corporations that committed environmental violations entered into DPAs. In general, t he distributions may sugges t that it is not the type of offense but the severity of the violation within each offense category or the willingness of the corporation to cooperate and accept responsibility that determines whether the corporation is offered a DPA or an NPA. Hypothesis 1 Findings To analyze a legal procedure or policy from the therapeutic jurisprudence perspective, the first step is to conduct a secondary data analysis of existing sources of information pertaining to the topic and develop hypotheses regarding the expect ed therapeutic or anti therapeutic effects of the action based on this existing knowledge. The second step is then to empirically test these hypotheses. Two key hypotheses were derived from the non empirical analysis and the remainder of Chapter 6 is dedic ated to testing these hypotheses. The first hypothesis is centered on the role of separate components of the DPA on the therapeutic outcome o f the punishment. Hypothesis 1 predicts that deferred and non prosecution agreement s which contain clauses emphasiz ing cooperation without the waiver of employee protections, restitution to victims, business reforms, a strengthened compliance and ethics program, and the use of a compliance monitor will be most successful at rehabilitating corporations and preventing fu ture offending. Table 7 5 presents the descriptive statistics for the dependent and independent variables included in the testing of Hypothesis 1 The key outcome variable is whether the corporation had subsequent criminal sanctions during the period of o r following th e DPA. Given the binary

PAGE 142

142 nature of the dependent variable a logistic regression model is appropriate for testing Hypothesis 1. As seen in Table 7 5, 16% of corporations (23 corporations) that received DPAs had recidivated from the time of thei r agreement through August of 2012. Table 7 5. Descriptive statistics for variables included in the model testing hypothesis 1 Variables Mean Median Minimum Maximum Dependent v ariable Subsequent c harges 0.16 0 0 1 Independent v ariables Independent compliance m onitor 0.33 0 0 1 Reforms Prior reforms 0.38 1 0 1 Dismissal of employees or executives 0.04 0 0 1 Creation of new positions 0.15 0 0 1 Reorganization of corporation 0.22 0 0 1 Revision to or termination of spe cific business practice 0.43 0 0 1 Training requirements 0.4 0 1 0 1 Compliance and e thics 0.54 1 0 1 Fine Any monetary payment required 0.77 1 0 1 Restitution payment required 0.2 0 0 0 1 Cooperation Encourage individual cooperat ion 0.81 1 0 1 Disclosure of criminal activity 0.41 0 0 1 Waiver of attorney client protections 0.35 0 0 1 Waiver of work product protections 0.34 0 0 1 Time from offense (in months) 35 34 3 108 Controls Type of agreement (NPA=1; DP A=2) 1.52 2 1 2 Time from agreement to August, 2012 (months) 75 65 32 228 Prior federal criminal record* 0.05 0 0 1 Total monetary payment required ( thousands) $ 48,171 $2,500 $0 $780,000 Note: Unless otherwise specified variables are dichotomous an d coded such that 'Yes=1.' *Prior criminal record does not include any criminal records associated with parent or subsidiary companies.

PAGE 143

143 The independent variables inc luded in the model, reflect the components of DPAs that were examined in the non empiri cal analysis. In addition to these covariates, four controls are also introduced into the analysis : the type of agreement (DPA or NPA); the number of months from the time the agreement was signed to the present time, in order to control for the amount of t ime during which the corporation could have had a subsequent offense; whether the corporation had a prior criminal record; and the total monetary payment required by the terms of the agreement, which is treated as a proxy measure for the severity of the co Before running the logistic regression model to test Hypothesis 1 a basic crosstab analysis was used to examine the percentage of agreements that contained the DPA components of interest among the corporations with a subsequent crimin al charge. While the crosstabs are not informative about the strength or significance of the relationship between the outcome measure and the covariates, the analysis does provide a general sense of the direction of the relationship. For example, because t he findings from the non empirical analysis suggested that the presence of a compliance monitor would result in a therapeutic outcome, it would be expected that a low percentage of corporations that recidivated had a compliance monitor requirement in the D PA. In contrast, Table 7 6 shows that more than half (57%) of the corporations that recidivated had a compliance monitor in place throughout the term of agreement. Among the recidivating corporations, the majority of DPAs also included consideration of ref orms initiated prior to the agreement (52%), a monetary payment requirement (57%), a requirement that the corporation investigation (96%), and a requirement that any subs equent criminal activity be disclosed to the U.S. Attorney (96%). These findings suggest that some of the components of DPAs that were expected to be r elated to a therapeutic outcome may have had the opposite effect instead The

PAGE 144

144 inclusion of the covariate s in a logistic regression model will clarify the significance of the relationship between the components of DPAs and the therapeutic or anti therapeutic outcome of DPA s and allow for an assessment of the odds that a corporation will recidivate given that a particular requirement was included in the terms of the DPA. Table 7 6. Components included in the deferred (DPA) and non prosecution (NPA) agreements of corporations which had subsequent criminal sanctions following the agreement Agreement componen ts Percent of recidivating corporations Independent compliance monitor 57 % Business reforms Consideration of reforms already initiated 52 % Dismissal of employees or executives 4 Creation of new oversight positions 35 Reorganization of corporation 22 Revision to or termination of business practice 35 Training requirements 40 Compliance program reforms 48 % Fines Any monetary payment required 57 % Restitution payment required 13 Cooperation Encouragement of individual cooperation 96 % Disclosure of subsequent criminal activity 52 Waiver of attorney client protections 49 Waiver of work product protections 44 No formal criminal charges filed (NPA) 44 % Prior criminal record 9 % Again, b ecau se the dependent variable is a dichotomous measure of whether or not the corporation had subsequent criminal charges, a binary logistic regression model was selected for the test of Hypothesis 1 A logistic regression model is appropriate because the model allows for a combination of categorical and continuous independent variables, does not require that the

PAGE 145

145 dependent variable be normally distributed, and does not assume a linear relationship between the dependent and independent variables. To ensure that m ulticollin earity was not a problem for the model, the Variance Inflation Factor (VIF) and the tolerance value were examined. The VIF revealed that the variables measuring the waiver of attorney client protections and the waiver of work product protections were in fact highly correlated. This finding makes sense in light of the fact that a corporation is not likely to agree to waive some protections but not others, and thus, the variable measuring whether the corporation was expected to waive work product protections was removed from the model. The removal of the variable does not substantially impact the model, as the correlation between the two variables suggests that the measure of whether attorney client protections were waive d is reflective of whether protections in general were waived as a term of the DPA. Once the work product protection variable was removed, the tolerance values and VIF values for each of the variables w ere close to one, suggesting that multicollinearity would not be an issue in the model. Again, the first model regressed key components of DPAs on a measure of whether the corporation was rehabilitated, to assess the therapeutic or anti therapeutic nature of these components. Covariates were initially entered into the model in six bloc ks, with the last block including the four control variables. None of the control variables attained significance in the model and the likelihood ratio test suggested that the inclusion of the control variables did not result in a statistically significant improvement in model fit. Therefore, these four variables were left out of the final model. With the exclusion of the control variables the Hosmer Lemeshow goodness of fit test (p=0.945) suggests that the overall fit of the model is good.

PAGE 146

146 Table 7 7 prese nts the findings from the logistic regression pr edicting the likelihood that a corporation will recidivate given certain terms and requirements in the DPA. All of the covariates, with the exception of the variables measuring privilege waivers and the numbe r of months from offense to agreement, were expected to be negatively related to recidivism. However, o f the five components of DPAs significantly related to the corporation receiving subsequent criminal charges, only two of the terms were in the expected negat ive direction The requirements that a corporation retain an independent compliance monitor and that the corporation create a new leadership or oversight position were both positively and significantly related to corporate recidivism at the p<.05 leve l. T he odds that a corporation will reoffend increase by a factor of 6.5 or 550% when the DPA includes a monitor and by a factor of 8.1 or 710% when the DPA requires the creation of a new leadership position. Likewise, the DPA requirement that the corporat ion use its best efforts to encourage individual employee cooperation in the investigation was related to an increased likelihood of corporate recidivism at the p <.1 level. On the other hand, DPA requirements that the corporation institute or improve on an existing compliance and ethics program and that the corporation pay a monetary penalty were negatively related to r eoffending at the p<.1 level. Interestingly, although it was anticipated that restitution payments, above and beyond other types of moneta ry payments, like fines or forfeiture, would ha ve therapeutic effects, the requirement of a restitution payment did not obtain significance in the model. Taken together these findings may suggest that if monetary fines and an improved compliance program a re the only components of DPAs related to a therapeutic outcome, that DPAs may not, in fact, be desirable over filing formal criminal charges and forcing the

PAGE 147

147 corporation to accept a plea bargain or go to trial. The test of Hypothesis 2 will clarify whether DPAs are in fact more likely to result in a therapeutic outcome than formal criminal sanctions. Table 7 7. Results of logistic regression predicting likelihood that a corporation which received a deferred or nonprosecution agreement will recidivate given the terms of the agreement Variable Coefficient (S.E.) Odds ratio Monitor 1.866 (.87) ** 6.460 Business reforms Prior reforms 0.507 (.59) 1.660 Employees dismissed 1.858 (1.39) 0.156 New position created 2.092 (.95) ** 8.102 R eorganized corporation .680 (.957) 1.973 Change to business practice .817 (.863) .442 Training .563 (1.08) .570 Compliance and ethics program 1.535 (.91) .215 Monetary payment Fine 1.048 (.62) .350 Restitution .379 (.8 1) .685 Cooperation Individual cooperation 2.293 (1.20) 9.902 Disclosure .317 (.61) 1.372 Waiver of attorney client protections .228 (.673) 1.257 Months from offense to DPA .021 (.02) 1.021 Intercept 1.652 **p<.05, p<.1 Hypothesis 2 Findings In addition to the 143 corporate DPAs included in the analysis, data were also collected on a sample of 128 corporations that received formal criminal sanctions through a guilty verdict or plea agreement in U.S. Dist rict Cour ts from 1993 to 2009 The sample of sanctioned corporations was selected to match the offenses committed by DPA corporations The matching was done in order to narrow the diffe rences betw een the two sets of offenders for the test the second hypothesis that DPAs are a more therapeutic punishment option than criminal sanctions.

PAGE 148

148 Table 7 8 show s the offenses committed by the corporations with DPAs and the criminally sanctioned corporations by the year the agreement was entered or the case d isposed. For several reasons it was not possible to perfectly match the offenses within each year. Part of the difficulty with matching stemmed from a lack of available information about the nature of the offense for the corporatio ns sanctioned before 2005. While th e text of a DPA typically contains a s tatement of facts with intricate details about the offense, the closest such document for a criminally sanctioned corporation is a criminal information or indictment and these case documents were not consistently available elec tronically prior to 2005. When the criminal information or indictmen t could not be located, the primary source information about the offense was the actual criminal charges and these charges often fall und er broad statutes. Therefore, 38 of the sanctioned charged with generic violations such as mail and wire fraud, false claims and statements, or fraud. Difficulties in matching the corporations also stemmed from the types of of fenses disposed through DPAs rather than criminal s anctions. For example, of the 39 corporations identified as having violated the Foreign Anti Corrupt Practices Act (FCPA) from 2004 to 2009 29 entered into DPAs, while 10 were formally indicted and sancti oned. It is unknown whether the 10 corporations that received criminal sanctions were offered a DPA option but chose not to ac cept the terms of the agreement or whether they were not given the option of a deferment. Regardless, since only a quarter of corp orations with FCPA violations receiv ed formal criminal sanction s, it was not possible to match all DPA corporation with FCPA violations to corporations that were criminally sanctioned for the same offenses

PAGE 149

149 While there are some differences in the offenses committed by the corporations that entered DPAs and the sample of sanctioned corporations, in general the offenses were fairly evenly matched, particularly when the entire period is considered rather than each individual year. The next question, then, is how different or similar were the punishments administered through the two case outcomes. Table 7 8. Comparison of the offenses committed by corporations receiving deferred or non prosecution agreements and formal criminal sanctions, by year Prosecut ion agreement Criminal sanction Year Number Offense (number ) Number Offense 1993 2 Import and export violations (1); Fraud against the government (1) 2 Import and export violations (1); Other offenses fraud (1) 1994 2 Securities fraud (1); Ant i kickback violations (1) 2 Fraud against the government (1); Other mail fraud (1) 1995 1 Securities fraud (1) 1 Fraud against the government (1) 1996 1 Securities fraud (1) 1 Fraud against the government (1) 1998 1 Environmental violations (1) 1 Environmental violations (1) 1999 1 Securities fraud (1) 1 Other offenses mail fraud (1) 2000 1 Securities fraud (1) 1 Securities fraud (1) 2001 2 Securities fraud (1); Other offenses (1) 2 Securities fraud (1); Fraud against the gove rnment 2002 2 Labor violations (1); Fraud against the government (1) 2 Labor violations (1); Fraud against the government (1) 2003 5 Securities fraud (3); Money laundering (1); Fraud against the government (1) 5 Money laundering (1); Antitrust (1); Fraud against the government (1); Other offenses false claims and falsification of records (2) 2004 9 Securities fraud (5); Money laundering (1); FCPA violations (1); Other offenses (2) 9 Money laundering (1); Environmental violations (1); FCPA viol ations (1); Other offenses aiding and abetting, false claims, obstruction of justice (6)

PAGE 150

150 Table 7 8. Continued Prosecution agreement Criminal Sanction Year Number Offense (number) Number Offense (number) 2005 14 Securities fraud (5); Anti kickback violations (1); Money laundering (1); Fraud against the government (3); FCPA violations (2); Health care fraud (1); Other offenses (1) 14 Securities fraud (2); Money laundering (1); FDCA violations (1); Fraud against the government (4); FCPA violations (1); Health care fraud (1); Other offenses mail/wire fraud, aiding and abetting, false claims (4) 2006 22 Securities fraud (6); Labor violations (1); Money Laundering (1); Antitrust (3); Environmental violations (1); FDCA violations (2); Fraud against the government (5); FCPA violations (2); Other offenses (1) 22 Securities fraud (1); Labor violations (1); Antitrust (5); Environmental violations (1); FDCA violations (1); Fraud against the government (4); FCPA violations (1); Other offenses wire frau d, false statements (8) 2007* 40 Import and export violations (2); Securities fraud (4); Anti kickback violations (8); Money laundering (4); Antitrust (1); FDCA violations (3); Fraud against the government (5); FCPA violations (10); Other offenses (3) 2 5 Import and export violations (3); Money laundering (1); Antitrust (2); Environmental violations (1); FDCA violations (3); Fraud against the government (3); Health care fraud (3); Other offenses mail fraud, false statements (9) 2008 22 Securities fra ud (2); Labor violations (3); Anti kickback violations (1); Money laundering (3); Environmental violations (1); Fraud against the government (1); FCPA violations (9); Other offenses (2) 22 Securities fraud (1); Labor violations (2); Anti kickback violati ons (2); Money laundering (2); Environmental violations (3); Fraud against the government (4); FCPA violations (4); Other offenses mail fraud, false claims (4) 2009 18 Import and export violations (2); Labor violations (3); Anti kickback violations (1 ); Money laundering (2); FDCA violations (1); Fraud against the government (2); FCPA violations (5); Health care fraud (1); Other offenses (1) 18 Import and export violations (1); Labor violations (2); Money laundering (1); FDCA violations (3); Fraud a gainst the government (3); FCPA violations (3); Health care fraud (3); Other offenses fraud, false statements (2) *The number of corporations with DPAs versus the number with criminal sanctions differs in 2007 due to a cap on the number of corporations with formal criminal sanctions included in the file each year. The number of corporations with criminal sanctions was capped at 25 to avoid including corporations with drastically different charges from the corporations with DPAs.

PAGE 151

151 Despite the fact that t he nature of the two types of disposition is quite different, they do share some commonalities that can serve as a basis for comparison. For example, t he testing of Hypothesis 1 revealed that the majority of DPAs (77%) included a monetary payment requireme nt. C omparisons can be thus made between the percentage of sanctioned corporations and the percentage of DPA corporations that paid a monetary penalty or restitution, as well as the amount of the average monetary payment paid by each group Because propone nts of DPAs also argue that these agreements are beneficial for reducing the time and resources required to successfully prosecute a corporate case (see, for example, Corporate Crime Reporter, 2005), another area for comparison is the length of time from t he most recent date of offending to the disposit ion of the case through a DPA versus a guilty judgment, verdict, or plea agreement. DPA (Holder, 1999), it is also expected that a greater percentage of corporations with criminal sanctions will have a prior criminal record. Table 7 9 shows the comparison of the punishments and punishment outcomes among corporations that entered into DP As and corporations that were formally sanctioned. Not surprisingly, a greater percentage of the sample of criminally sanctioned corporations paid monetary fines and restitution than the corporations with DPAs. Surprisingly, though, the corporations that e ntered DPAs paid stiffer penalties on average than the criminally sanctioned corporations. This may be reflective of a disparity between the two groups in terms of the severity of the offense committed or in terms of the size of the corporation and the abi lity to pay. It also serves to demonstrate on e of the points made about DPAs, which is that, in practice, even extraordinary cooperation does not mean lenient or more lenient

PAGE 152

152 Table 7 9. Comparison of punishment, prior criminal record, length of time from offense until disposition, and recidivism among corporations that received deferred and non prosecution agreements and formal criminal sanctions Prosecution agreement Criminal sanctions Percent with monetary payment requirement 77 % 98 % Fine, forfeiture, special assessment 67 91 Restitution 20 30 Average monetary payment required (thousands) $48,171 $18,022 Percent on probation 100 % 62 % Percent with prior criminal record 10 % 5 % Average time from offense to disposition (months) 35 40 Percent with subsequent criminal sanctions 16 % 16 % *Information on the most recent offense date was not available for 9% of corporations. Since a DPA is a form of corporate probation, 100% of the corporations that entered DPAs were on probation as a term of the punishment. However, the majority of corporations that were criminally prosecuted were also put on probation for a period of time. Contrary to expectation, 9% of the corporat ions wi th a DPA had a prior criminal record, compared to 5% of the criminally sanctioned corporations. On the other hand, as expected, corporations that were formally prosecuted waited an average of 40 months from the offense to the disposition of the case, while the time from the offense to the signing of a DPA was 35 months on average. Finally, and most importantly in terms of the test of Hypothesis 2, there was no difference in the percentage of corporations with subsequent criminal charges among those that en tered a DPA and those that were formally indicted and prosecuted. Across the board, 16% of the corporations in the analysis reoffended at some point during the time from the case disposition through August of 2012. This finding does not provide initial sup port for Hypothesis 2 and the idea that DPAs are a more therapeutic punishment option than other criminal sanctions in terms

PAGE 153

153 of doing a better job of rehabilitating corporations. The final test, however, is the second logistic regression analysis. Table 7 10 presents the descriptive statistics for the dependent and independent variables included in the logistic regression analysis of Hypothesis 2 Again, Hypothesis 2 predicts that offending corporations that entered into DPAs rather than being criminally c harged and sanctioned will have a lower likelihood of committing a subsequent criminal offense due to the therapeutic and rehabilitative nature of DPAs. As in the first model, the dependent variable is a measure of whether the corporation had subsequent cr iminal charges following the case disposition. A s expected from Table 7 9 and the prior comparison of corporations with DPAs and criminal sanctions, despite the addition of the 128 sanctioned corporations to the sample, the mean of the dependent va riable w as unchanged from the first regression model The independent variable for the testing of Hypothesis 2 is a measure of case disposition; in other words, whether the corporation entered into a DPA or was criminally sanctioned. In addition to the independe nt variable the model also controls for the length of time from the offense to the disposition, the total monetary payment required by the corporation, which serves as a proxy measure of the severity of the offense, and whether the corporat ion had prior c riminal charges. T o reduce the number of small cells in the model, the variables measuring total monetary payment and the time from the offense to the case disposition were collapsed into categorical variables. Prior to running the binary logistic regress ion analysis to test Hypothesis 2, the variables were again checked for multicollinearity. The tolerance factor and VIF were both close to one, suggesting that multicollinearity should not be a problem for the model.

PAGE 154

154 Table 7 10. Descriptive statistics f or variables included in the model testing Hypothesis 2 Variables Mean Median Minimum Maximum Dependent variable Subsequent charges 0.16 0 0 1 Independent variable Disposition (0=sanctions; 1=DPA/NPA) 0.53 1 0 1 Controls Time from of fense to disposition (1=12 months or less; 2=13 through 36 months; 3=37 months or more) 2.37 2 1 3 Total monetary payment (0=$0; 1=$250,000 or less; 2=$250,001 $2.5 million; 3=more than $2.5 million) 1.93 2 0 3 Prior criminal record 0.07 0 0 1 Note: Unless otherwise specified variables are dichotomous and coded such that 'Yes=1.' The covariates were entered into the model in two blocks, with the independent variable entered first, followed by the three control variables. Because the logistic regres sion model does a listwise deletion of any variables with missing values, the model removed the 11 variables with missing data on the time from the offense to disposition, giving an N of 260. Unfortunately, the Hosmer Lemeshow test can only be used if ther e are three or more groups, so it was not an appropriate measure of goodness of fit for the second model. However, the chi square test of the second model resulted in a p value of .009, suggesting that the full model is at least statistically significant. As shown in Table 7 11, the independent variable measuring whether the corporation entered into a DPA or was formally sanctioned did not attain significance in the model and the null hypothesis could not be rejected. In other words, contrary to expectatio ns, the logistic regression analysis of Hypothesis 2 found no difference in the likelihood recidivism, regardless of the disposition of the case.

PAGE 155

155 Table 7 11. Results of logistic regression predicting the likelihood of corporate recidivism given the dispo sition of the case (deferred prosecution agreement versus criminal sanctions) Variable Coefficient (S.E.) Odds ratio Disposition .421 (.427) 0.657 Controls Total fine ($0) Total fine ($250,000 or less) 1.769 (.80) ** 0.171 Total fine ($250,001 $2.5 million) 1.263 (.60) ** 0.283 Total fine (more than $2.5 million) .416 (.47) 0.660 Time from offense to disposition (12 months or less) Time from offense to disposition (13 36 months) .886 (.56) 0.412 Time from offense to disposition (37 months or more) .404 (.51) 0.667 Prior criminal record 1.271 (.52) ** 3.563 Intercept 1.675 **p<.05 There were however, two covariates that emerged as significantly related to corporate recidivism. First, prior crimi nal record was positively and significantly related to recidivism. The odds that a corporation will reoffend increase by a factor of 3.6 or 260% if the corporation has a prior criminal record. The amount of the fine was also significantly related to corpor ate recidivism on two levels. Compared to no fine, fines of $250,000 or less and between $250,001 and $2.5 million were negatively related to recidivism. In other words, up to a certain amount fines are related to a reduced likelihood of corporat e recidivi sm. Fine amounts above $2.5 million, however, had no impact on whether the corporation had subsequent criminal sanctions. This finding appears to support the notion that fines have deterrent value only if they are set at a level at which the corporation ca n afford to pay, while still exceeding the monetary gains that could be generated through the criminal offense (Coffee, 1981). Presumable the higher fines were levied against corporations that could afford to pay higher amounts and also gained more from th e criminal offense.

PAGE 156

156 CHAPTER 8 CONCLUSIONS, LIMITAT IONS, AND FUTURE RES EARCH Discussion and Conclusion This study endeavored to accomplish two major objectives The first objective was to apply the therapeutic jurisprudence perspective to an area of crimina l justice policy where it had not been previously applied While the therapeutic jurisprudence perspective has continued to be largely associated with mental health law, Professors Wexler and Winick, who are credited with the development of the perspective have encouraged its broader application including to areas such as corporate crime la w (Wexler, 1995; Winick, 1997). Professor Wexler noted that if therapeutic consequences could be identified in areas such as corporate crime policy where they were lea st expected, these consequences would have major implications and relevance (Wexler, 1995). Thus, t he prosecution a greements for corporate offenders in the hopes of identifying a corpor ate crime policy with therapeutic consequences. corporate offenders, using the therapeutic jurisprudence perspective as the lens through which to examine the practice Ov er the past two decades the DOJ has been steadily increasing the use of DPAs to handle corporate offenders However, despite the fact that corporate offending and the potential collateral consequences of corporate sanctions can have detrimental effects on the economy and our society as a whole, this policy shift has largely escaped the attention of the criminological community. Limited research, primarily centered in the legal field, has considered specific issues pertaining to the administration of DPAs, s uch as the scope of prosecutorial discretion and the constitutionality of certain aspects of cooperation. However, no know n studies

PAGE 157

157 to date have considered the totality of DPAs and whether they are a favorable alternative to formal criminal indictment and prosecution. Th is lack of existing research as well as the underlying ideas behind DPAs made this practice an ideal c andidate for examination from the therapeutic jurisprudence perspective. First, the use of DPAs began increasing out of recognition that f ormal criminal sanctions against corporations could have serious collateral consequences for innocent employees, shareholders, and customers. Thus, DPAs developed as an approach to punishment designed to all eviate some of the negative or anti therapeutic c onsequences associated with traditional practices Moreover, since not much is known about the actual use of DPAs, the therapeutic jurisprudence perspective provided a useful framework, beginning with a non empirical analysis phase that allowed for the sys tematic formation of hypotheses on the therapeutic and anti therapeutic consequences of the policy. The non empirical analysis phase was then followed with a testing phase, in which the hypotheses were empirically tested. The non empirical analysis examin ed several major components of DPAs, including requirements pertaining to voluntary disclosure and cooperation, the retention of an independent compliance monitor, monetary fines and restitution, and business reforms and improved compliance and ethics prog rams. These components were examined in the context of existing research in related areas and what the research suggested about the expected therapeutic and anti therapeutic consequences of each component and DPAs as a whole. For example, the assessment of the use of monetary fines and restitution was grounded in existing research pertaining to the use of fines and restitution payments for other types of offenders, as well as research on rational choice and corporate offending.

PAGE 158

158 From the non empirical analys is, two hypotheses were develop ed about the therapeutic consequence of using DPAs for corporate offenders. The therapeutic consequence of interest for the empirical analysis was corporate rehabilitation, measured by whether the corporation had subsequent c riminal charges following the signing of the DPA. While there are other consequences of corporate punishments that are relevant for determining the therapeutic nature of the punishment, the case study analysis of the Bristol Myers Squibb DPA confirmed that recidivism can be more reliably measured than other collateral consequences of DPAs and highlighted that point that if the use of DPAs does not result in the same or lower rates of recidivism than formal criminal sanctions, the practice is fundamentally a nti therapeutic, regardless of other outcomes of the punishment. Hypothesis 1 then predicted that the most therapeutic DPAs were those that contained clauses requiring cooperation without a waiver of employee rights, restitution to victims, a change in co rporate structure, an improved compliance and ethics program, and the retention of an independent compliance monitor and that these components would be related to a reduced likelihood of reoffending. Hypothesis 2 predicted that corporate offenders which en tered DPAs were more likely to be rehabilitated than corporate offenders that were formally indicted and sanctioned and as such, would have a lower likelihood of recidivism. Unfortunately, neither of the two hypotheses was supported by the empirical anal yses. The results of the analysis of Hypothesis 1 suggested instead that the requirement that the corporation retain an in dependent compliance monitor, the business reform requirement that the corporation create new leadership positions and the requiremen t that the corporation encourage individual were related to a likelihood of reoffending. Since critics of DPAs have argued that the use of compliance monitors represents

PAGE 159

159 government intrusion into a cor poration by people with limited business experience (Spivack and Raman, 2008), it may be the case that instead of the monitor providing positive assistance with the carrying out of the terms of the agreement, those within the corporation resent the presenc e of the monitor and perceive the requirement to be so unfair that the legitimacy afforded the punishment is reduced, as is the ability of the punishment to be rehabilitative (Braithwaite, 1989; Makkai and B raithwaite, 1994; Tyler, 2006). The finding that new leadership positions were p ositively related to recidivism suggests that corporations are creating these new positions but filling them with managers who continue to reinforce the cultural norms that facilitated the offending in the first place. Since managers are largely responsible for the ethical climate within a corporation (Hollinger and Clark, 1982, 1983), the creation of new leadership positions appears to be strengthening rather than correcting a negative corporate culture. Likewise, the finding that encouraging individual cooperation was related to reoffending may suggest that the individual employees cooperate with the law when required by management, but when managerial requirements related to law abiding and cooperation are relaxed, employees return to the same deviant behaviors On the other hand, it may also be the case that cooperation opens up lines of communication that had not previously existed between the corporation and the DOJ, and that because of these lines of communication, corpor ate offenses that may not have otherwise been discovered are proactively brought to the attention of the DOJ. The test of Hypothesis 1 did reveal that two components of DPAs had a slightly significant impact on corporate offending in the expected direction The payment of monetary fines, though not specifically restitution, and the requirement that the corporation strengthen its compliance and ethics program were both found to be therapeutic components of DPAs related to a decreased likelihood of recidivism Neither of these components is particularly unique to DPAs,

PAGE 160

160 however. Fines are commonly used in most corporate sanctions and corporate plea agreements can also include requirements pertaining to the development or improvement of corporate compliance prog rams. Thus, the finding that these were the only components of DPAs related to a reduced likelihood of reoffending does not lend strong support towa rds the use of DPAs over traditional criminal sanctions. Though Hypothesis 2 predicted that DPAs are better able to rehabilitate corporations than other criminal sanction s the empirical analysis instead revealed that the same percentage of corporations reoffended regardless of whether the case was disposed through a guilty verdict or plea bargain or through a DPA. There are several key points to note in relation to this finding, though. For starters, the analysis was unable to take into account regulatory and civil enforcement actions, which make up a growing proportion of the actions against offending corporat ions (Simpson, 2002). It may be the case that a similar percentage of corporations on either side had subsequent criminal charges but that one group or the other ha d a greater percentage of regulatory or civil actions against them. Likewise, while DPAs do not appear to result in a more therapeutic outcome in terms of rehabilitating the corporation, it may be that DPAs are still more therapeutic than criminal sanctions in terms of other measures that could not be obtained for the analysis, such as the collat eral consequences of the punishment on innocent employees and shareholders. Clearly if the practice of entering into DPAs with corporate offenders cannot rehabilitate the corporation, the practice is not fully therapeutic to the degree that Wexler and Wini ck would want to see. However, the analysis cannot conclusively s ay that DPAs are not still more therapeutic than formal criminal sanctions. Combined with the findings from the test of Hypothesis 1, the most conclusive finding that emerged from the analys es is that corporate punishments that included fines were more effective

PAGE 161

161 at preventing recidivism than punishments that did not include a monetary payment requirement. Just as fines were negatively related to reoffending among corporations that entered DPA s in the first model, the second model also found that, to a certain point, fines were negatively related to recidivism among all corporate offenders, regardless of how the case was disposed. This finding does not necessarily refute the argument that fines alone cannot deter corporate offenders since, at least in the case of DPAs, the effects of the fine cannot be fully separated from the rest of the punishment. However, it does suggest that there is perhaps more merit to the use of fines for punishing corp orate offenders than some corporate crime scholars would seem to believe (Coffee, 1981; Cohen, 1989; Geis, 2002). Limitations and Suggestions for Future Research The deficiencies in the current research highlight the need for additional research on the us e of corporate deferred prosecution agreements, particularly their ability to serve as a more therapeutic punishment option than traditional criminal sanctioning approaches. Two key shortcomings in the current analysis limit the ability to rule out DPAs as a more therapeutic punishment for corporations than other sanctioning options. The first limitation in the current research pertains to an inability to examine the consequences of DPAs and formal criminal sanctions on others impacted by the punishment bey ond just the corporation. The therapeutic jurisprudence perspective emphasizes the i mportance of examining the thera peutic and anti therapeutic consequences of legal practices on both offenders and those tangentially impacted by the criminal justice action While the current research was able to assess whether DPAs are better able to rehabilitate offenders than formal criminal san ctions, it did not examine the collateral consequences of DPAs on innocent employees and shareholders. As demonstrated through t he BMS case study, c ollateral consequences are difficult to measure One could examine the size of the corporation and corporate profits prior to and after

PAGE 162

162 the DPA to get a sense of the health of the corporation and whether innocent employees lost their jo bs. However, information on corporation size and profits is difficult to obtain for private corporations, and both of these measures could be impacted by a number of factors other than the corporate punishment, such as market changes for the particular ind ustry and the economy overall. Additionally, if the offenses committed by the corporation were falsely inflating revenue and profits, an adjustment would need to be made to take into considering the degree of inflation. Likewise, for publicly traded corpor ations one could also examine changes in share prices before and after the DPA to get an indication of the impact of the punishment on persons invested in the corporation. Again, though, the stock market and corporate share prices could be impacted by any and without extensive knowledge of how the markets work and the historical changes in the markets, it would be difficult to attribute increased or decreased share prices to one par ticular factor. That said, despite the challenges, it would be worthwhile for future research to attempt to quantify the collateral consequences associated with both DPAs and other criminal sanctions. Based on anecdotal stories about corporations like Art hur Andersen that were restricted from engaging in key business practices because of criminal sanctions, it is assumed that criminal sanctions have anti therapeutic consequences for innocent employees and shareholders and that DPAs alleviate some of these negative consequences. However, these assumptions have not been empirically tested. DPAs contain requirements regarding changes to business practices, and it is foreseeable that these changes could work to rehabilitate the corporation at the expense of cor porate sal es and growth, for example. Additionally, while the U.S. Attor neys overseeing the terms of DPA s typically take into account the health of the corporation in fining decisions, many

PAGE 163

163 of the fines and restitution requirements are still substantial an d could result in corporate downsizing or increases in the prices of goods and services provided by the corporation. Furthermore, most DPAs are publically available, often announced through DOJ press releases and in some instances required to be prominentl decided to utilize a competitor with a clean record instead. All of these scenarios could potentially have anti t herapeutic consequences for innocent employees and shareholders. Future research should incorporate measures of corporate growth and stock value to attempt to gauge whether DPAs have therapeutic or anti therapeutic consequences for those tangentially impac ted by the punishment. The second key shortcoming of the current research was the sole focus on federal criminal sanctions in the analysis of corporate recidivism and rehabilitation. While a good portion of corporate transgressions are resolved through ci vil or regulatory interventions (Simpson, 2002), there are reasons why a focus on criminal sanctions is appropriate. R egulatory actions for example, are geared towards ensu r ing that business practices adhere to rules and standards, and as such, often addr ess behaviors beyond those that are socially injurious. Civil actions require a lower burden of proof than criminal sanctions and a corporation may determine that the cost of fighting the case is simply not worth it, and settle despite a lack of solid evid ence that malfeasance occurred. In other words, both civil and regulatory actions cast a broader net than criminal sanctions. That said, compared to other widely cited studies examining the full range of enforcement actions against corporations, such as t hose conducted by S utherland (1949) and Clinard and Yeager (1980), the 16% recidivism rate found in the current study is substantially lower. This is

PAGE 164

164 due to the focus on criminal sanctions alone. If it is assumed, then, that the percentage of corporations with subsequent violations would increase with the inclusion of regulatory and civil enforcement actions, the question remains whether the percentage would increase disproportionately among corporations with criminal sanctions versus those that entered DPAs. Thus, future research should consider the full range of enforcement actions against corporations in examining how DPAs measure up to other enforcement actions in terms of the therapeutic consequences and rehabilitation of corporations. A long these same lines, for the purpose of the current research, corporations that entered into DPAs were matched with corporations that received criminal sanctions for similar types of offenses. ted cases tend to not be firms too small to 47). In other words, offense type may not have been the most appropriate for selecting matched corporations. If corporate size is related to corporate offending and the corporations with criminal sanctions tend to be smaller than the corporations that entered into DPAs, this could have an impact on the fin dings. Future research should perhaps consider selecting a sample of corporations with similar characteristics in terms of size and revenue as the corporations that entered into DPAs to see if DPAs have more of a rehabilitative impact on larger corporation s that have been found to have higher rates of offending and recidivism in the first place (Dalton and Kesner, 1988). In addition to these key limitations that should be addressed by future research, there are a number of other areas pertaining to the valu e and utility of DPAs that could and should be further explored. For instance, future research should give consideration to the finding that the average time from offense to case disposition was shorter among corporations with DPAs than

PAGE 165

165 those that were for mally sanctioned. For the current study, the issue of time from offending to disposition was more about punishment celerity than about the prosecutorial resources that go into a case. However, beyond the therapeutic consequences of DPAs, if they can reduce the time and resources that go int o handling corporate cases, with out impacting recidivism rates, this alone may be justification for their continued or increased use. Future studies should assess the hours spent on a case from the start of the DOJ invest igation through case disposition for cases that are criminally sanctioned and those that enter DPAs, taking into account the nature and complexity of the alleged offense. A nother major topic pertaining to the use of DPAs that was outside of the scope of th e future research but should be considered by corporate criminologists is the impact of DPAs on individual offenders charged in conjunction with the case. Cooperation is a central fo cus of DPAs and one still unanswered question is to what extent cooperatio n helps federal pros ecutors build stronger cases against culpable indiv iduals. Future studies should examine whether more individuals are prosecuted for their roles in the offense and whether the charges against them are different or more severe when a cor poration enters into a DPA than when the corporation is prosecuted alongside the individuals. Along these same lines, corporate offenders often face both crim inal action and civil law suits stemming from the same offense While DPAs often contain statement s of the facts of the case, which are signed off on and accepted by the corporation, u n like formal criminal sanctions, DPAs are not a finding of corporate guilt. Future research therefore, should also examine whether DPAs have a differential impact on civ il cases against the corporation than do formal criminal sanctions.

PAGE 166

166 In sum, despite the fact that the current study did not find conclusive support for the notion of DPAs as a therapeutic punishment option for corporate offenders, it is important to note t hat DPAs do not result in higher levels of recidivism than formal criminal sanctions. Since both punishment options result in similar level of recidivism, the DO ripe topic for future research to determine whether there are other benefit s to the use of these agreement s that would encourage even wider use. There is still much that is unknown about the impact of the use of DPAs for corporate offenders. What is known, however, is that addressing corporate crime can have major implications i n terms of corporate recidivism, deterrence, and the collateral consequences of sanctioning for those associated with the corporation and society in general. The criminology community has been remiss in not giving the use of corporate DPAs the attention an d evaluation that is necessary for a criminal justice practice wit h such great potential impact. The current study should serve to open the door for this much needed future research.

PAGE 167

167 LIST OF REFERENCES Andenaes, Johannes. 1975. General prevention revisit ed: Research and policy implications. Criminology 66(3): 338 365. Appelbaum, Steven H., Kyle J. Deguire, and Mathieu Lay. 2005. The relationship of ethical climate to deviant workplace behavior. Corporate Governance 5(4): 43 55. Baysinger, Barry D. 1991. O rganization theory and the criminal liability of organizations. Boston University Law Review 74:341 376. Baucus, Melissa S. and Janet P. Near. 1991. Can illegal corporate behavior be predicted? An event history analysis. Academy of Management Journal 34(1) :9 36. Beccaria, Cesare B 1819. An Essay o n Crimes and Punishment Translated by Edward D. Ingraham. Philadelphia: Philip H. Nicklin. Benson, Michael L. and Francis T. Cullen. 1998. Combating Corporate Crime: Local Prosecutors at Work Boston: Northeaster n University Press. Birgden, Astrid. 2000. Maximising therapeutic effects in treating sexual offenders in an Australian correctional system. Behavioral Sciences & the Law 18 (4): 479 488. Birgden, Astrid. 2007. Serious Se x Offenders Monitoring Act 2005 : A therapeutic jurisprudence analysis. Psychiatry, Psychology, and Law 14 (1): 78 94. Blackstone, William. 1962. Commentaries on the Laws of En gland Vol. 4., A dapted by Robert Malcolm Kerr. Boston: Beacon Press. Blumstein, Alfred J. Cohen, and Daniel Nagin 1987. Estimating the Effect of Criminal Sanctions on Crime Rates Washington, DC : Nationa l Academy Press Bohrer, Barry A. and Barbara L. Trencher. 2007. Symposium: Corporate criminality: Legal, ethical, and managerial implication: The challenge of coopera tion: Prosecution deferred: Exploring the unintended consequences and future of corporate cooperation. American Criminal Law Review 44: 1481 1506 Bowman, Frank O., III. 2003. Sarbanes Oxley Act and what came after. Federal Sentencing Reporter 15(4): 231 23 3. Braithwaite, John. 1982. Enforced self regulation: A new strategy for corporate crime control. Michigan Law Review 80: 1466 1507. Braithwaite, John. 1984. Corporate Crime in the Pharmaceutical Industry London: Routledge and Kegan Paul. Braithwaite, Jo hn. 1989. Crime, Shame, and Reintegration Cambridge, UK: Cambridge University Press.

PAGE 168

168 Braithwaite, John and Toni Makkai. 1991. Testing an expected utility model of corporate deterrence. Law and Society Review 25:7 41. Bratton, William W. 2002. Enron and th e dark side of shareholder value. Tulane Law Review 76:1275 1362. Brickey, Kathleen. 2003. From Enron to WorldCom and beyond: Life after Sarbanes Oxley. Washington University Law Quarterly 81: 357 402. Bristol Myers Squibb. 2007 Annual Report Pursuant to Section 13 or 15(d) of the Securities Exchange Act of 1934. Retrieved June 1, 2012 from U.S. Securities and Exchange Commission EDGAR database. Bristol Myers Squibb. 2008 Annual Report Pursuant to Section 13 or 15(d) of the Securities Exchange Act of 1934 Retrieved June 1, 2012 from U.S. Securities and Exchange Commission EDGAR database. Bristol Myers Squibb. 2009 Annual Report Pursuant to Section 13 or 15(d) of the Securities Exchange Act of 1934. Retrieved June 1, 2012 from U.S. Securities and Exchange Commission EDGAR database. Bristol Myers Squibb. 2010 Annual Report Pursuant to Section 13 or 15(d) of the Securities Exchange Act of 1934. Retrieved June 1, 2012 from U.S. Securities and Exchange Commission EDGAR database. Bristol Myers Squibb. 2011 An nual Report Pursuant to Section 13 or 15(d) of the Securities Exchange Act of 1934. Retrieved June 1, 2012 from U.S. Securities and Exchange Commission EDGAR database. Bristol Myers Squibb. 2012 Annual Report Pursuant to Section 13 or 15(d) of the Securit ies Exchange Act of 1934. Retrieved June 1, 2012 from U.S. Securities and Exchange Commission EDGAR database. B rownlee, Ian D. 1995. Intensive probation with young adult offenders: A short reconviction study. British Journal of Criminology 35:590 612. Bucy Pamela H. 2007. Symposium: Corporate criminality: Legal, ethical, and managerial implication: Why punish?: Trends in corporate criminal prosecutions. American Criminal Law Review 44:1287 1305. Canfield, George F. 1914. Corporate responsibility for crime. Columbia Law Review 14(6): 469 81. Casey, Pamela and David B. Rottman. 2000. Therapeutic jurisprudence in the courts. Behavioral Sciences and the Law 18:445 457.

PAGE 169

169 Christie, Christopher J. and Robert M. Hanna. 2006. A Push Down the Road of Good Corporate Citizenship: The Deferred Prosecution Agreement Between the U.S. Attorney for the District of New Jersey and Bristol Myers Squibb Co.. American Criminal Law Review 43: 1043 1062 Clark, Michael D. 2001. Change focused drug courts: Examining the critical i ngredients of positive behavioral change I n National Drug Court Institute Review vol. 3., ed s. Judge Karen Freeman Wilson, Michael P. Wilk osz and Carson L. Fox, Washington, DC: National Drug Court Research Center. Clear, Todd and Anthony A. Braga. 1995. Community Corrections. In Crime eds. James Q. Wi lson and Joan Petersilia, San Francisco: Institute for Contemporary Studies. Clinard, Marshall B. 1983. Corporate Ethics and Crime The Role of Middle Management Thousand Oaks, CA: Sage Publications Inc. Clinard, Marshall B. and Peter B. Yeager. 2005. Corporate Crime Piscataway, New Jersey : Transaction Books. Coates, John C. IV. 2007. The goals and promise of the Sarbanes Oxley Act. The Journal of Economic Perspectives 21(1): 91 116. Coffee, John C., Jr. 1981. No soul to damn: No body to kick: An unscandalized inquiry into the problem of corporate punishment Michigan Law Review 79: 386 459. Cohen, Fred and Joel A. Dvoskin. 1993. Therapeutic jurisprudence and corrections: A glance New York Law School Jou rnal of Human Rights 10:777 804. Cohen, Deborah Vidaver. 1993. Creating and maintaining ethical work climates: Anomie in the workplace and implications for managing change. Business Ethics Quarterly 3(4): 343 358. Cohen, Irving. 1944. The integration of re stitution in the probation service. Journal of Criminal Law, Criminology and Police Science 34: 315 321. Cohen, Mark. 1989. Corporate crime and punishment: A study of social harm and sentencing practice in the federal courts, 1984 1987. American Criminal Law Review 26:605 660. Coleman, James S. 1974. Power and the Structure of Society New York: Norton. Coleman, James S. 2002. Organizational actors and the irrelevance of persons. In Corporate and Governmental Deviance eds. M. David Ermann and Richard J. L undman, New York: Oxford University Press. Coleman, James W. 1995. Motivation and opportunity: Understanding the causes of white collar crime. In White Collar Crime eds. Gilbert Geis, Robert F. Meier, and Lawrence M. Salinger, New York: The Free Press.

PAGE 170

170 Co oper, Richard M. 2005. Deferred prosecution: A primer. The National Law Journal July 25. Available at (last accessed May 7, 2010). Conklin, John E. 1977. Illegal but Not Criminal: Business Crime in America Englewood Cliffs, NJ: Prentice Hall. Corporate Crime Reporter. 2005a. Interview with Mary Jo White, Partner, Debevoise & Plimpton, December 12. Available at http://w interview010806.htm (last accessed May 7, 2010). Corporate Crime Reporter. 2005 b Professor Coffee: Anderson committed suicide, not murdered by prosecutors. Available at (last accessed May 7, 2010). (last accessed March 23, 2012). Couden, Carmen 2005. The Thompson Memorandum: A revised solution or just a problem? The Journal of Corporation Law 30:405 423. Cressey, Donald R. 1995. Po verty of theory in corporate crime research. In White Collar Crime eds. Gilbert Geis, Robert F. Meier, and Lawrence M. Salinger, New York: The Free Press. Crouch, Ben M. s for prison ov er probation. Justice Quarterly 10(1): 67 88. Cullen, Francis T., and P.J. Dubeck. 1985. Myth of corporate immunity to deterrence: Ideology and the creation of the invincible criminal. Federal Probation 44(3): 3 9. Cullen, Francis T., William J. Maakestad, and Gary Cavender. 1987. Corporate Crime Under Attack Cincinnati: Anderson, 342 353. Dallas, Lynne L. 2003. A preliminary inquiry into the responsibility of corporations and their mise. Rutgers Law Journal 35: 1 68. Dalton, Dan R. and Idalene F. Kesner. 1988. On the dynamics of corporate size and illegal activity: An empirical assessment. Journal of Business Ethics 7(11): 861 870. Desio, Paula. 1991. An Overview of the Organizationa l G uidelines. Washington, DC: United States Sentencing Commission. Available at (last accessed February 15, 2009).

PAGE 171

171 Doyle, Charles. 2007. The T hompson Memoran Attorney Client Work Product P rotection. Washington, DC: Congressional Research Service Report for Congress Duggin, Sarah Helene. 2008. The McNulty Memorandum, the KPMG decision and corporate cooperation: Indiv idual rights and legal ethics. Georgetown Journal of Legal Ethics 21:341 409. Edward, First Baron Thurlow, Lord Chancellor of England, 1731 1806. 1844. In Literary Extracts from English and Other Works; Collected During Half a Century: Together with Some O riginal Matter ed. John Poynder Charleston, SC: Nabu Press. Elston, Michael. 2007. The challenge of cooperation: Cooperation with the government is good for companies, investors, and the economy. American Criminal Law Review 44: 1435 1440. Erickson, May nard L., Jack P. Gibbs, and Gary F. Jenson. 1977. The deterrence doctrine and the perceived certainty of legal punishments. American Sociological Review 42:305 317. Fagan, Jeffrey and Tom R. Tyler. 2004. Police, order maintenance and legitimacy. In Polici ng in Central and Eastern Europe: Dilemmas of Contemporary Criminal Justice eds. Gorazd Mesko, Milan Pagon and Bojan Dobovsek Ljubljana, Slovenia : Koda Press Ferrell, O.C., Debbie Thorne LeClair, and Linda Ferrell. 1998. The Federal Sentencing Guidelin es for organizations: A framework for ethical compliance Journal of Business Ethics 17(4): 353 364. Finder, Lawrence D. and Ryan D. McConnell. 2006. Devolution of Authority: The Department Saint Louis University La w Journal 51:1 52. Fisse, Brent and John Braithwaite. 1993. Corporations, Crime and Accountability Melbourne, Australia: Cambridge University Press. Flory, Christopher M., David C. May, Kevin I. Minor, and Peter B. Wood. 2006. A comparison of punishment e xchange rates between offenders under supervision and their supervising officers. Journal of Criminal Justice 34(1): 39 50. Ford, Christie and David Hess. 2009. Can corporate monitorships improve corporate compliance? Journal of Corporation Law 34: 679 737 Fortune Magazine. 2000 Fortune 500 Retrieved June 2, 2012 from Fortune Magazine. 2001 Fortune 500 Retrieved June 2, 2012 from Fortune Magazine. 2002. Fortune 500 Retrieved June 2, 2012 from

PAGE 172

172 Fortune Magazine. 2003. Fortune 500 Retrieved June 2, 2012 from Fortune Magazine. 2004. Fortune 500 Retrieved June 2, 2012 from Fortune Magazine. 2005. Fortune 500 Retrieved June 2, 2012 from tune/fortune500_archive/full/2005/ Fortune Magazine. 2007. Fortune 500 Retrieved June 2, 2012 from Fortun e Magazine. 2012. Fortune 500 Retrieved June 2, 2012 from Friedman, Lawrence. 2000. Essay in defense of cor porate criminal liability. Harvard Journal of Law and Public Policy 23:833 58. Garrett, Brandon L. 2007. Structural reform prosecution. Virginia Law Review 93: 853 952. Garrett, Bran don L. and Jon Ashley N.D.a Federal Organizational Plea Agreements. Univ e rsity of Vi rginia School of Law. Available at (last accessed August 31, 2012). Garrett, Bran don L. and Jon Ashley. N. D.b Federal Organizational Prosecution Agreements, University of Vi rginia School of Law. Available at hp (last accessed August 31, 2012). Geis, Gilbert. 1973. Deterring corporate crime. In Corporate Power In America eds. Ral ph Nader and Mark J. Green, New York: Grossman. Geis, Gilbert. 2002. The heavy electrical equipment antitrust cases: Price fixing t echniques and rationalizations. In Corporate and Governmental Deviance eds. M. David Ermann and Richard J. L undman. New York: Oxford University Press. Geis, Gilbert and Joseph F. C. DiMento. 2002. Empirical evidence and the legal doctrine of corporate cri minal liability. American Journal of Criminal Law 29 : 341 76 Gorden, Margaret A. and Daniel Glaser. 1991. The use and effects of financial penalties in municipal court. Criminology 29(4): 651 76. Grasmick, Herald G. and Byrjack, G. 1980. The deterrent eff ect of perceived severity of punishment. Social Forces 59:471 91. Green, Mark. 1973. The corporation and the community. In Corporate Power In America eds. Ral ph Nader and Mark J. Green, New York: Grossman.

PAGE 173

173 Greenberg, Li ane and Julian Barling. 1996. Emplo yee theft. In Trends in Organizational Behavior vol. 3., eds. Cary L. Cooper a nd Denise M. Rousseau, Hoboken, NJ: John Wiley and Sons Ltd. Harris, Gardiner. 2002. Bristol Myers's Finance Chief, Schiff, Resigns Amid Criticism The Wall Street Journal Onli ne April 17. Available at (last accessed February 17, 2012). Harrison, Jeffrey L. 1994. Class, personality, contract, and uncon sionability William and Mary Law Review 35:445 502. Heinz, J., B. Galaway, and J. Hudson. 1976. Restitution or parole: A follow up study of adult offenders. Social Service Review 50: 148 156. Hess, David. 2007. A business ethics perspective on Sarbanes Ox ley and the organizational sentencing guidelines. Michigan Law Review 105(8):1781 1816. Holder, Eric. 1999. Memorand um from Eric Holder, Deputy Attorney General, U.S. Department of Justice to Component Heads and U.S. Attorneys on Bringing Criminal Charges Against Corporations, June 16. Available at corps.pdf (last accessed August 28, 2012 ) Holling er, R ichard C. and John Clark 1982. Formal and informal social controls of employee deviance. The Sociological Quarterly 23 : 333 43. Hollinger, Richard C. and John Clark. 1983. Theft by Employees Lexington, MA: Lexington Books. Arthu r Andersen LLC. In Depth Records. Available at companyId= 618194096 (last accessed September 25, 2011). 2012 a In Depth Records. Retrieved June 2, 2012 from Lexis Nexis Academic database. 2012 b Bristol Myers Squibb Company Hoov In Depth Records. Retrieved June 2, 2012 from Lexis Nexis Academic database 2012 c Eli Lilly PLC In Depth Records. Retrieved June 2, 2012 from Lexis Nexis Academic database 2012 d Teva Pharmaceutical Industries Limited pany Records In Depth Records. Retrieved June 2, 2012 from Lexis Nexis Academic database.

PAGE 174

174 Hora, Peggy Fulton, The Honorable, The Honorable William G. Schma, and John T.A. Rosenthal. 1999. Therap eutic jurisprudence and the drug treatment court movement: America. Notre Dame Law Review 74:439 537. Horowitz, Michael E. and April Oliver. 2006. Forward: The state of feder al prosecution. American Criminal Law Review 43:1033 41. Hudson, J. and S. Chesney. 1977. Research on restitution: A review and assessment. In Offender Restitution in Theory and Action eds. B. Galaway and J. Hudson, Toronto: Lexington Books. Jeffries, Sam antha. 2005. therapeutic jurisprudence Current Issues in Criminal Justice 17: 245 68. Jiang, Yu. 1994. Punishment celerity and severity: testing a specific deterrence model on drunk driving recidivism. Journal of Criminal Justice 22(4):355 66. Kamp, John and Paul Brooks. 1991. Perceived organizational climate and employee counterproductivity. Journal of Business and Psychology 5(4): 447 58. Kennedy, Christopher. 1985. Criminal sentences for c orporations: Alternative fining mechanisms. California Law Review 73(2): 443 82. Khanna, Vikramaditya and Timothy L. Dickinson. 2007. The corporate monitor: The new corporate czar. Michigan Law Review 105: 1713 56 Kingsnorth, Rodney, Debra Cummings, John Lo pez, and Jennifer Wentworth. 1999. Criminal sentencing and the court probation officer: The myth of individualized justice revisited. The Justice System Journal 20(3):255 74. Lattman, Peter. 2006. Frederick Lacey: An independent monitor with teeth. The W all Street Journal September 16. Available at lacey an independent monitor with teeth/ (last accessed Marc h 2, 2012). Lee, Matthew T. and M. David Ermann. 2002. Pinto madness: Flaws in the generally acc epted landmark narrative. In Corporate and Governmental Deviance eds. M.David Ermann and Richard J. Lundman, New York: Oxford University Press. Lipsey, Mark W. 1999. Can rehabilitative programs reduce the recidivism of juvenile offenders? An inquiry into the effectiveness of practical programs. Virginia Journal of Social Policy and Law 6: 611 42. Litchblau, Eric. 2008. In justice shift, corporate deals replace trials. New York Times April 9. Available at (last accessed June 19, 2009).

PAGE 175

175 Makkai, Tony and John Braithwaite. 1994. R eintegrative shaming and compliance with regulatory standards. Criminology 32: 261 386. Martz, Stephanie. 2005. Trends in deferred prosecution agreements The Champion 29:45 46. McCallum, Robert D. 2005. Memorandum from Acting Deputy Attorney General Rober t D. McCallum, Jr. to Head of Department Components and U .S. Attorneys on Waiver of Corporate Attorney Clie nt and Work Product Protections, October 21. Available at http:/ / ( last accessed August 28, 2012). McKendall, Marie, Beverly DeMarr, and Catherine Jones Rikkers. 2002. Ethical compliance programs and corporate illegality: Testing the assumptions of the Cor porate Sentencing Guidelines. Journal of Business Ethics 37(4): 367 83. McLucas, William R., Howard M. Shapiro, and Julie J. Song. 2006. The decline of the attorney client privilege in the corporate setting. Journal of Criminal Law and Criminology 96(2):621 41. McNulty, Paul J. 2006. Memorandum from Deputy Attorney General Paul J. McNulty to Heads of Departmen t Components and U.S. Attorneys on principles of federal prosecution of business o rganizations December 12. Available at (last accessed August 28, 2012). McPhee, Joan. 2006. Deferred prosecution agreements: Ray of hope or guilty plea by another name. The Champion 30:12 14. Moffatt, S teve and Suzanne Po ynton. 2007. The deterrent effect of higher fines on recidivism: Driving offences Crime and Justice Bulletin 106:1 16. Mokhiber, Russell. 2005. Crime without conviction: The rise of deferred and non prosecution agreements Corporate Crime Reporter December 28. Available at (last accessed February 16, 2008). Moohr, Geraldine Szott. 2007. Of bad apples and bad trees: Considering fault based liability for the complicit corporation. American Criminal Law Review 44:1343 64 Morford, Craig S. United States. 2008. Memorandum from Acting Deputy Attorney General Craig S. Morford to Heads of Department C omponents and U.S. Attorneys on selection and use of monitors in deferred prosecution agreements and non prosecutio n agreements with corporations, March 7. Available at http://www.justice .gov/dag/readingroom/dag 030708.pdf (last accessed August 21, 2012). Morris, Norval and Michael Tonry. 1990. Between Prison and Probation New York: Oxford University Press.

PAGE 176

176 Murphy, Kristina. 2009. Public satisfaction with police: The importance of procedu ral justice and police performance in police citizen encounters. The Australian and New Zealand Journal of Criminology 42(2): 159 78 Nader, Ralph, Mark Green and Joel Seligman. 1976. Taming the Giant Corporation : How the Largest Corporations Control Our L ives New York: W.W. Norton and Company, Inc.. Nagel, Ilene H. and John L. Hagen. 1982. The sentencing of white collar criminals in federal courts: A socio legal exploration of disparity. Michigan Law Review 80: 1427 1465. Nagin, Daniel S and Greg Pogarsky. 2001. Integrating celerity, impulsivity, and extralegal sanction threats into a model of general deterrence: Theory and evid ence. Criminology 39(4): 865 92. Orland, Leonard. 2006. The transformation of corporate criminal law. Brooklyn J. Financial and Comm ercial Law 1:45 85. Outlaw, Maureen C. and R. Barry Ruback. 1999. Predictors and outcomes of victim restitution orders. Justice Quarterly 16(4): 847 69. Paine, Lynn Sharp. 1994. Managing for organizational integrity. Harvard Business Review 2:106 17. Parker Jerry and Harold G. Grasmick. 1979. Linking actual and perceived certainty of punishment. Criminology 17:366 79. Passas, Nikos. 1990. Anomie and corporate deviance. Crime, Law and Social Change 14(2): 157 78. Paternoster, Raymond. 1987. The deterrent ef fect of the perceived certainty and severity of punishment: A review of the evidence and issues. Justice Quarterly 4(2): 173 217. Paternoster, Raymond and Piquero, Alex 1995 Reconceptualiz ing deterrence: An empirical test of personal and vicarious experiences Journal of Research in Crime and Delinquency 32(3) : 251 86. Paternoster, Raymond and Sally S. Simpson. 1993. A rational choic e theory of corporate crime. In Routine Activity and Ratio nal Choice eds. Ronald V. Clarke and Marcus Felson, Piscataway, New Jersey; T ransaction Publisher Paulson, Erik. 2007. Imposing limits on prosecutorial discretion in corporate prosecution agreements. New York University Law Review 82: 1434 69. Perino, Mic aspects of the Sarbanes Oxley Act of 2002. 76(4): 671 698. Perlin, M.L. 1994. The Jurisprudence of the Insanity D efense Durham, NC: Carolina Academic Pres s.

PAGE 177

177 Petersilia, Joan. 1990. When probation becomes more dreaded than prison. Federal Probation M arch. 23 27. Petersilia, Joan. 1997. Probation in the United States. Crime and Justice 22: 149 200. Petersilia, Joan and Susan Turner. 1990. Intensive Supervisio n for High Risk Probationers: Findings from Three California E xperiments Santa Monica, CA: The RAND Corporation. Available at s/2007/R3936.pdf (last accessed May 28, 2011). PR Newswire. 2000. Bristol Myers Squibb a nnounces a 'Strateg y for Growth' company sharpens its focus on medicines b usine ss PR NewsWire September 28. Available at releases/bristol myers squibb announces a strategy for growth --company sharpens its focus on medicines business --73444657.html (last accessed February 17, 2012). Pratt, Travis C and Francis T. Cullen. 2005. Assessing macro level predictors and theories of crime: A meta analysis. In Crime and Justice ed. Michael Tonry Chicago : University of Chicago Press. Pratt, Travis C., Francis T. Cullen Kristie R. Blevins, Leah E. Daigle, and Tamara D. Madenson. 2006. The empirical status of deterrence theory: A meta analysis. In Taking Stock : The status of criminological theory eds. Francis T. Cullen, John Paul Wright, and Kristie R. Blevins Piscataway, NJ: Transaction Publishers. Ramirez, Mary Kreiner. 2005. The science fiction of corporate criminal liability: Containing the machine through corporate death penalty Arizona Law Review 47:933 1002. Robi nson, James K., Philip E. Urofsky, and Christopher R. Pantel. 2005. Deferred prosecutions and the independent monitor. International Journal of Disclosure and Governance 2(4):325 46. Roderick, Dennis and Susan T. Krumholz. 2006. Much ado about nothing: A c ritical examination of therapeutic jurisprudence. Southern New England Roundtable Symposium Law Journal 1:201 23. Romano, Roberta. 2005. The Sarbanes Oxley Act and the making of quack corporate governance. Yale Law Journal 114: 1521 1611. Romero, Simon and for bankruptcy; Largest U.S. case. The New York Times July 22. A vailable at s collapse the overview worldcom files for bankruptcy largest us case.html (last accessed August 27, 2010). Ronner, Amy D. 2000. Therapeutic jurisprudence on appeal. Court Review 37: 64.

PAGE 178

178 Ross, H. Laurence. 1987. Administrative license revocation in New Mexico: An evaluation. Law & Policy 9(1):5 16. Ruback, R. Barry and Mark H. Bergstrom. 2006. Economic sanctions in criminal justice: Purposes, effects, and implications. Criminal Justice an d Behavior 33:242 73. crime. Crime, Law & Social Change 32: 59 82. Sarbanes Oxley Act of 2002 2002. U.S. Code Title 15, secs. 7201 226. Senjo, Scott and Leslie A. Leip. 2001a. Testing and developing theory in drug court: A four part logit model to predict program completion Criminal Justice Policy Review 12:66 87. Senjo, Scott and Leslie A. Leip. 2001b. Testing therapeutic jurisprudence theory: An empirical assess ment of the drug court process. Western Criminology Review 3(1): [Online] Available at (last accessed May 24, 2010). Sherman, Lawrence W. 1993. Defiance, deterren ce, and irrelevance: A theory of criminal sanctions. Journal of Research in Crime and Delinquency 30:445 73. Shostak, Arthur B. 1974. Modern Social Reforms New York: Macmillian. Shuman, Daniel W. Shuman. 1993. Therapeutic jurisprudence and tort law: A lim ited subjective standard of care. Southern Methodist University Law Review 46: 409 32 Simon, Herbert A.. 1976. Administrative Behavior: A Study of Decision Making Processes in Administrative Organizations. 3 rd ed., New York: Free Press. Simon, Leonore M.J 1996. Therapeutic jurisprudence approach to the legal processing of domestic violence cases In Law in a T herapeutic K ey eds. David B. Wex ler and Bruce J. Winick. Durham, NC: Carolina Academic Press. Simpson, Sally S.. 2002. Corporate Crime, Law, and S ocial Control Cambridge: Cambridge University Press. Simpson, Sally S. and Christopher S. Koper .1992 Deterring corporate crime Criminology 30: 347 375. Sims, Barbara. 2000. Victim restitution: A review of the literature. The Justice Professional 13: 24 7 69. Slobogin, Christopher. 1995. Therapeutic jurisprudence: Five dilemmas to ponder. Psychology, Public Policy and Law 1(1): 193 219. Spiva c deferred prosecution agreements. American Criminal Law Review 45( 2):159 93.

PAGE 179

179 Staltzburg, Stephen A. 1991. The control of criminal conduct in organizations Boston University Law Review 71:421 38. Stolle, Dennis P. and David B. Wexler. 1997. Therapeutic jurisprudence and preventive law: A combined concentration to invigorate the everyday practice of law Arizona Law Review 39(25): 25 34. Stolle, Dennis P., David B. Wexler, Bruce J. Winick, and Edward A. Dauer. 1997. Integrating preventive law and therapeutic jurisprudence: A law and society based approach to lawyering California Western Law Review 34(16):15 52. Stone, Christopher D. 1975. Where the Law Ends New York: Harper and Row. Stotland, Eliza. 1982. The role of law enforcement in the fight against white collar crime. In White Collar C rime: An Agenda for Research eds. Herbert Ed elhertz and Thomas D. Overcast. Lexington, MA: Lexington Books. Sutherland, Edwin H. 1940. White collar criminality. American Sociological Review 5:1 12. American Sociological Review 10: 132 139. Swigert, Victoria Lynn and Ronald A. Farrell. 1980 81. Corporate homicide: Definitional processes in the creation of deviance. Law & Society Review 15(1):161 82. Tappan, Paul W. 1947. Who is the criminal? American Sociological Review 12:96 102. The Yale Law Journal. 1979. Structural crime and institutional rehabilitation: A new approach to corporate sentencing. The Yale Law Journal 89(2):353 75. Thompson, Larry D. 2003. Memorandum from Deputy Attorney General Larry Thompson to Heads of Departmen t Components and U.S. Attorneys on Principles of Federal Prosecution of Busines s Organizations, June 20. Available at porate_guidelines.htm (last accessed August 28, 2012). McCallum, Robert D. 2005. Memorandum from Acting Deputy Attorney General Robert D. McCallum, Jr. to Head of Department Components, United States Attorneys on Waiver of Corporate Attorney Clie nt and Wo rk Product Protections, October 21. Available at ( last accessed August 28, 2012). T onry, Michael an d Mary Lynch. 1996. Intermediate Sanctions. Crime and Justice : A Review of Research 20:99 144. Turner, Susan, Joan Petersilia, and Elizabeth Piper Deschenes. 1992. Intensive supervision probation/parole (ISP) for drug offenders. Crime and Delinquency 38:5 39 556.

PAGE 180

180 Tyler, Tom R. 1992. The psychological consequences of judicial procedures: Implications for civil commitment hearings Southern Methodist University Law Review 46:433 45. Tyler, Tom R. 2006. Why People Obey the L aw. Princeton, NJ: Princeton Univers ity Press. Tyler, T om R., and E.A. Lind. 1992 A relational model of authority in groups. In Advances in Experimental Social P sychology ed. Mark Zanna. New York: Academic Press. U.S. Attorney for the District of Massachusetts. 1993. Offer of settlement to the U.S. Attorney for the District of Massachusetts, the Attorney General for the Commonwealth of Massachusetts, and the Massachusetts Securities Division from Paul A. Ehrnhardt, President, Aetna Capital Management, August 19. Available at (last accessed August 29, 2012). U.S. Attorney for the District of Massachusetts, 2 006. United States of Amer ica v Prudential Equity Group, LLC, August 28. Available at ( last accessed August 29, 2012). U.S. Attorney for the District of New Jersey. 2005. United States of America v. Bristol Myers Squibb Company Deferred Prosecution Agreement, June 15. Available at meyers.pdf (last accessed May 22, 2012). U.S. Attorney for the District of New Jersey. 2007. United States of America vs. Zimmer, Inc Deferred Prose cution Agreement, September 27. Available at 3.htm (last accessed August 29, 2 012). U.S Attorney for the Eastern District of Virginia. 2004 United States of America v. America Online, Inc. Deferred Prosecution Agreement, December 15. Available at (last accessed August 29, 2012). U.S. Attorney for the Southern District of New York. 1994. Letter from Mary Jo White, U.S. Attorney to Scott W. Muller, Esq and Carey R. Dunne, Esq. RE: Prudential Securities Incorporated, October 27. Available at (last accessed Augu st 29, 2012). U.S. Attorney for the Southern District of New York. 2005. Letter from David N. Kelley, U.S. Attorney to Robert S. Bennett, Esq RE: KPMG Deferred Prosecution Agreement, August 29. Available at (last accessed August 29, 2012).

PAGE 181

181 U.S. Attorney for the Southern District of New York. 2006. Letter from Michael J. Garcia, U.S. Attorney to Law rence Byrne, Esq RE: Royal Ahold N.V. and U.S. Foodservice, Inc., September 25. Available at old.pdf (last accessed August 29, 2012). U. S. Attorney for the Southern District of New York. 2007. Letter from Michael J. Garcia, U.S. Attorney to Charles M. Carberry, Esq. RE: Chevron Corporation, November 8. Available at (last accessed August 29, 2012). U.S. Department of Justice. 2007 a Statement from U.S. Attorney Christopher J. Christie on Expiration of Deferred Prosecution Agreement with Bristol Myers Squibb. Washington, DC. Available at 14 (last accessed March 23, 2012). U.S. Department of Justice. 2007b. Bristol Myers Squibb pleads guilty to lying to the Federal Government about dea l involving blood thinning drug, May 7. Washington, D.C.: U.S. Department of Justice. Available at (last accessed June 1, 2012). U.S. Department of Justice, 2008. Justice D epartment Revises Charging Guidelines for Pr osecuting Corporate Fraud August 28. Washington, D.C.: U.S. Department of Justice. Available at odag 757.html (last accessed June 6, 2 010). U.S. Department of Justice, Corporate Fraud Task Force. 2003. First Year Report to the President. Washington, DC: U.S. Department of Justice. U.S. Department of Justice, Criminal Division, Fraud Section, 2003. United States of America vs. PNC ICLC C orp Deferred Prosecution Agreement, June 2. Available at (last accessed August 29, 2012). U.S. Department of Justice C riminal Division, Fraud Section. 2004. Letter from Joshua R. Hochberg, Chief, Fraud Section, Criminal Division to Brad D. Brian, Esq RE: General Electric Company and InVision Technologies, Inc., December 3. Available at (last accessed August 29, 2012). U.S. Department of Justice, Criminal Division, Fraud Section. 2005. United States of American v. Monsanto Company Deferred Prosecution Agreement, January 6. Available at s/pdf/monsanto.pdf (last accessed August 29, 2012).

PAGE 182

182 U.S. Department of Justice, Criminal Division, Fraud Section. 2008. United States of America v AGA Medical Corporation Deferred Prosecution Agreement, June 3. Available at (last accessed August 29, 2012). U.S. Department of Justice, Enron Task Force. 2003. Letter from Leslie R. Caldwel l, Director, Enron Task Force to Gary Naftalis, Esq. RE: Canadian Imperial Bank of Commerce, December 22. Available at t/prosecution_agreements/pdf/canadianimperialbank.pdf (last accessed August 29, 2012). U.S. Securities and Exchange Commission. 2001. Report of investigation pursuant to Section 21(a) of the Securities Exchange Act of 1934 and commission statement on the relationship of cooperation to agency enforcement decisions. Available at 44969.htm (last accessed March 26, 2009). U.S. Sentencing Comm ission. 2010 Federal Sentencing Gui delines Manual. Washington, DC.: U.S. Sentencing Commission. Available at 2_1.htm (last accessed February 18, 2011). United States v. Stein Southern District of New York 05 C.R. 888. Voreacos, David. 2010. Ex Bristol Myers Finance C hief Fred erick Schiff wins appeals court r uling Bloomberg April 7. Available at 04 07/ex bristol myers finance chief schiff wins appeal in securities fraud case.html (last acces sed March 23, 2012). Warin, F. Joseph and Andrew S. Boutros. 2007. Deferred prosecution agreements: A view from the trenches and a proposal for reform. Virginia Law Review 93: 121 134. Wexler, David B. 1990. Inducing therapeutic compliance through the crim inal law. Law and Psychology Review 14:43 58. Wexler, David B. 1991. Health care compliance principles and the insanity acquittee conditional release process Criminal Law Bulletin 27:18 41. Wexler, David B. 1995. Reflections on the scope of therapeutic ju risprudence Psychology, Public Policy and Law 1:220 36. Wexler, David B. 1996. Applying the law therapeutically Applied and Preventive Psychology 5:179 186. Wexler, David B. 2000. Therapeutic jurisprudence in the appellate arena. Seattle University Law R eview 24:217 21. Wexler, David B. 2008. Two decades of therapeutic jurisprudence Touro Law Review 24:17 30.

PAGE 183

183 Wexler, David B. and Bruce J. Winick. 1991. Therapeutic jurisprudence as a new approach to mental health law policy analysis and research Universi ty of Miami Law Review 45:979 1004. Wexler, David B. and Bruce J. Winick. 1996. Law in a Therapeutic Key: Developments in Therapeutic Jurisprudence Durham, NC: Carolina Academic Press. Weisburd, David, Elin Waring, and Ellen C ha yat. 1995. Specific deterre nce in a sample of offenders convicted of white collar crimes. Criminology 33 (4) :587 607. Weissmann, Andrew. 2007. Symposium: Corporate, criminality: Legal, ethical, and managerial implication: Why punish?: A new approach to corporate criminal liability. A merican Criminal Law Review 44:1319 1342. Welles, C. 1998. What led Beech Nut down the road to disgrace? Business Week 22:124 28. Wheeler, Stanton and Mitchell Lewis Rothman. 1982. The organization as weapon in white collar crime. Michigan Law Review 80(7) :1403 1426. Williams, K.R. and R. Hawkins. 1986. Perceptual research on general deterrence: A critical review. Law and Society Review 20:544 71. Winick, Bruce J. 1997. The jurisprudence of therapeutic jurisprudence Psychology, Public Policy, and Law 3(1): 184 206. Wood, Peter G. and Harold G. Grasmick. 1999. Toward the developmen t of punishment equivalencies: M ale and female inmates rate the severity of alternative sanctions compared to prison Justice Quarterly 16(1):19 50. Yeager, Peter C. 1991. The Limit s of the Law: Public Regulation of Private Policing Cambridge: Cambridge University Press. Yu, Jiang. 1994. Punishment celerity and severity: Testing a specific deterrence model on drunk driving recidivism. Journal of Criminal Justice 22(4): 355 66.

PAGE 184

184 BIOGRAPHICAL SKETCH Lynn Langton Austell received a Bachelor of Arts in a nthropology and s ociology from Centre College in Danville, Kentucky in 2002. In 2004, she w as awarded a Master of Arts in criminology, law and s ociety from the University of Florida. In 2006, Lynn accepted a position as a statistician at the Bureau of Justice Statistics (BJS), U.S Department of Justice, where she is currently employed working on the National Crime Victimization Survey. During her tenure at the BJS, she has p ublished more than a dozen federal statistical reports on topics ranging from civil trials, to law enforcement practices, to identity theft victimization. She has also published numerous articles in refereed journals focused on applying criminological theo ries to white collar crime and corporate offending. Lynn lives in Washington, D.C. with her husband Dan.