1 FEDERAL SENTENCING GUIDELINES AND DONALD BLACKS THEORY OF LAW: ANALYZING SENTENCE OUTCOMES AM ONG ORGANIZATIONAL DEFENDANTS By JASON DAVIS A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2008
2 2008 Jason Davis
3 To my sons, Royalton, Bryce and Dylan, my incr edible and beautiful wi fe Lisa, my supportive brother Quentin Davis, and my loving parents Cassandra and Alton Davis
4 ACKNOWLEDGMENTS I would like to express my deep appreciation to m y remarkable chair, Marian Borg, for her guidance throughout this long educational progre ssion. Her tremendous commitments will never be forgotten. I would also like to acknowledge my committee members, Ronald Akers, Lonn Lanza-Kaduce, Leonard Beeghley, and Michael Martinez for their patience, time, and critical assistance. On very short notice (particula rly Michael Martinez), th ey accepted tremendous responsibilities and commitments for which I am grateful. In addition, I cannot thank Nadine Gillis enough for her invaluable services and advice. I would like to pay a special tribute to my parents, Cassandra and Alton Davis, and brother, Quentin Davis. Because of their love and devotion I have never been afraid to seek or a ccept new challenges. Finally, I would like to recognize my lovely wife, Lisa Holland-Davis, for her unwavering support and encouragement. At every stage of this meandering journey, she wa s always there for me and believed in me. In the end, I could not have realized my dreams without the cherished strength and love of my wife and the Davis family.
5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ............................................................................................................... 4LIST OF TABLES ...........................................................................................................................8LIST OF FIGURES .........................................................................................................................9ABSTRACT ...................................................................................................................... .............10 CHAP TER 1 FEDERAL SENTENCING GUIDELINES, ORGANIZATIONAL DEFENDANTS, AND BLACKS THEORY OF LAW .................................................................................... 12Theoretical Orientation ....................................................................................................... ....13Significant Contributions ..................................................................................................... ...142 FEDERAL SENTENCING GUIDELINES FOR ORGANIZATIONS (FSGO) ...................20Federal District Courts ............................................................................................................20Historical Development of the Fede ral Sentencing Guidelines (FSG) ...................................21Historical Development of Federal Sentenci ng Guidelines for Organizations (FSGO) ......... 23Federal Sentencing Guidelines for Orga nizations (FSGO) Sentencing Process .................... 25Underlying Philosophy of the Federal Sent encing Guidelines for Organizations (FSGO): Good Corporate Citizen ship and Compliance Programs ....................................29Evaluations of the Federal Sentencing Guidelines for Organizations (FSGO) ...................... 303 DONALD BLACKS THEORY OF LAW ............................................................................ 33Overview of Blacks Theory of Law ......................................................................................34Stratification (Vertical Status or Rank) ........................................................................... 35Morphology (Integrational Status) ..................................................................................36Culture (Conventionality) ................................................................................................ 38Organization .................................................................................................................. ..40Social Control (Normative Status or Respectability) ...................................................... 41Empirical Tests of Blacks Theory of Law ............................................................................. 45Individual Level Evaluations of Blacks Theory of Law ................................................46Structural-Level Evaluations of Blacks Theory of Law ................................................ 50Multilevel Evaluations of Blacks Theory of Law ..........................................................51
6 4 BLACKS THEORY OF LAW AND CORPORATE CR IME SENTENCING .................... 57Stratification and Corporate Crime Sentencing ...................................................................... 58Corporate (Offender) Rank and Sentence Outcomes ...................................................... 58Federal District Rank and Sentence Outcomes ............................................................... 60Morphology (Integrational Status) an d Corporate Crime Sentencing .................................... 61Corporate (Offender) In tegrational Status a nd Sentence Outcomes ............................... 61Federal District (Integrational St atus) and Sentence Outcomes ......................................62Culture (Conventionality) and Co rporate Crime Sentencing ................................................. 63Corporate (Offender) Conventiona lity and Sentence Outcomes ..................................... 63Federal District (Cultural Dist ance) and Sentence Outcomes ......................................... 63Organization and Corporate Crime Sentencing ...................................................................... 70Corporate (Offender) Organizational Status and Sentence Outcomes ............................ 70Federal District Organizational Status and Sentence Outcomes .....................................71Social Control (Normative Status or Respect ability) and Corporate Crime Sentencing ........ 72Corporate (Offender) Respectabil ity and Sentence Outcomes ........................................ 72Federal District Respectability and Sentence Outcomes ................................................. 725 ANALYZING SENTENCE OUTCOMES UNDER T HE ORGANIZATIONAL GUIDELINES: DATA AND MEHTODS ............................................................................ 74Dependent Variable ................................................................................................................78Case Level (Legal) I ndependent Variables ............................................................................. 79Offense Seriousness .........................................................................................................79Culpability Score ............................................................................................................. 80Independent Variables: Ex tralegal Measures of Cor porate Offender Statuses ..................... 81Corporate Rank (Economic Status) ................................................................................. 81Corporate Integrational Status (Morphology) ................................................................. 82Corporate Cultural Status (Conventionality) ................................................................... 83Corporate Organizational Status ...................................................................................... 83Corporate Respectability (Social Control) ......................................................................83Independent Variables: St ructural Measures of Fe deral District Status ................................84Federal District Rank (Economic Status) ........................................................................ 84Federal District Integrational Status (Morphology) ........................................................84Federal District Cultural St atus (Conventionality) ..........................................................85Federal District Orga nizational Status .............................................................................86Federal District Respectability (Normative Status) ......................................................... 86Summary of Blacks Empirical Hypotheses Related to Corporate Offender and Federal District Statuses ...................................................................................................................87Analytic Strategy ....................................................................................................................91
7 6 PREDICTIVE POWER OF BLACKS THEO RY ON SE NTENCE OUTCOMES FOR ORGANIZATIONAL DEFENDANTS ................................................................................. 97Level I Random Coefficient Model: Fixe d and Random Effects Models of Legal Variables ..................................................................................................................... .97Full Random Coefficient Model: Fixed and Random Effects M odels of Blacks Theoretical Concepts ................................................................................................... 99Summary of the Empirical Findings ..............................................................................1057 SUMMARIZING THE IMPORTANCE OF BLACKS THEORY & FEDERAL SENTENCI NG GUIDELINES FOR ORGANIZATIONS .................................................. 107Significant Contributions and Implications ................................................................... 112Future Research Directions ...........................................................................................114APPENDIX A NUMBER OF CASES PER FEDERAL DIST RICT (2001-2004) ......................................121B GRAPHS OF SLOPE RELATIONSHIPS ........................................................................... 122LIST OF REFERENCES .............................................................................................................125BIOGRAPHICAL SKETCH .......................................................................................................135
8 LIST OF TABLES Table page 2-1 Partial list offens e level fine table ..................................................................................... .262-2 Minimum and maximum culpability multipliers. .............................................................. 282-3 Hypothetical fine ranges for various offense levels and culpability scores ....................... 285-1 Descriptive statistics for case and structural level variables .............................................. 885-2 Correlation matrix for case and district le vel variables related to total fine amount ......... 895-3 Summary of the full HLM model specified in equation format ........................................956-1 Unconditional model of log total fine amount ................................................................... 986-2 Random coefficient model of log total fine ..................................................................... 1006-3 Full random coefficient model of log total fine ............................................................... 102
9 LIST OF FIGURES Figure page 5-1 Conceptual model of Black s theory applied to corpor ate sen tencing: Effects of offender and district st atuses on fine amount .................................................................... 92B-1 Predicted slopes of ability to pay and logged total fine amount in a sample of districts 122B-2 Predicted slopes of ability to pay and logged total fine amount in all districts ............... 122B-3 Predicted slopes of number of counts a nd logged total fine amount in a sample of districts ..................................................................................................................... ........123B-4 Predicted slopes of number of counts and logged total fine amount in all districts ........123B-5 Predicted slopes of number of counts a nd logged total fine amount in a sample of districts ..................................................................................................................... ........124B-6 Predicted slopes of number of counts and logged total fine amount in all districts ........124
10 Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy FEDERAL SENTENCING GUIDELINES AND DONALD BLACKS THEORY OF LAW: ANALYZING SENTENCE OUTCOMES AM ONG ORGANIZATIONAL DEFENDANTS By Jason Davis May 2008 Chair: Marian J. Borg Major: Sociology This study used Donald Blacks theory of law as a framework to analyze sentence outcomes among organizations sanctioned under th e federal sentencing guidelines. More specifically, it examined the extent to which (1 ) offender and (2) location (federal district) characteristics affect fine amounts. The organizational guidelines were enacted to address corporate crimes and help eliminate sentence disparities related to extralegal offender characteristics including financia l status and size. While sent encing outcomes should be based on legal factors including offense seriousness an d offender culpability, post-guideline research indicates extralegal factors still exert influence on sentence outcomes. Blacks general theory of law was used because it focuses on the social structure of a case which includes characteristics of the actors involved in a dispute, th e relational ties between these actors, and characteristics of the social se tting. Moreover, he asserts the behavior of the law can be explained by five aspects of social life including economic, integrational, cultural, organizational and normative statuses. Black pro poses that across all setting offenders with lower statuses will be subject to greater law (higher fine amounts). At the same time, settings or locations with greater social statuses will us e more law (impose higher fines). Accordingly,
11 Blacks theoretical concepts helped to anticip ate the extent to which corporate offender and district statuses affect sentence outcomes. This study analyzed 813 cases from 2001 to 2004, using hierarchal linear modeling (HLM). Findings provided non-supportive evidence for Blacks theory. Despite the lack of theoretical confirmation, there are important conclusions. Firs t, there is significant variation across federal districts in terms of fine am ounts. Secondly, extralegal offender factors are significantly more important in de termining fines than legal factor s. The strongest predictor of fine amount is a corporations ability to pay (ext ralegal factor). Interestingly, culpability score (legal factor) is significantly but inversely related to fine amounts. Offenders with higher culpability scores are more likely to receive lower fines. Finally, districts differentially consider the importance of both legal and ex tralegal in deciding fines. This calls into question the formal rationality of the organizational guidelines and th e notion of one unified federal district court system.
12 CHAPTER 1 FEDERAL SENTENCING GUIDELINES OR GANIZATIONAL DEFENDANTS, AND BLACKS THEORY OF LAW This dissertation provides a theoretical analys is of the federal sentencing guidelines for organizations (FSGO) to determine the Effect of (1) offender characteristics and (2) location (federal districts) on sentence outcomes or fine amounts. In 1987, the US Sentencing Commission (hereafter Commission) created the federal sentencing guidelines (FSG) in an effort to eliminate unwarranted sentence dispari ties among individual offenders. By 1991 the Commission created organizationa l guidelines to more effectiv ely sanction offenses commonly referred to as corporate crime or crimes co mmitted by entities other than persons such as corporations, non-profit organizations and government entities (USSC, 2007)1. By implementing formalized instructions, the Comm ission sought to significantly curb judicial discretion and restrict sentence de cisions to legally relevant fact ors such as seriousness of the offense; prior criminal convictions; and the degr ee of harm caused by the crime. Accordingly, any sentence variations could be explained by legal variables. Overall, the intent of both the individual and organizational guidelines was to ensure consistent and uniform outcomes for similarly situated defendants or offenders who committed analogous crimes and who had comparable cr iminal histories (Tonry, 1993). Despite the Commissions stated goals, post-FSG research continues to discover that extralegal factors yield sentence disparities among individual (A lbonetti, 1997; Kautt, 2002; Kautt & Spohn, 2002; Steffensmeier & Demuth, 2000; Steffensmeier et al., 1998; Ulmer, 2005) and organizational offenders (Piquero & Davis, 2004; Beck & OBrien, 2000). 1 According to Chapter 8 of Federal Sentencing Guidelines Manual, organizations are defined as persons other than individuals and include corporations, partnerships, associations, unions, non-profit organizations, or government and political entities (USSC, 2007)
13 Since this dissertation examined the effect of offender and contextual characteristics on sentence outcomes, it provided a more systemic analysis of the federal sentencing guidelines than previous research. For the most part, FSG research has been lim ited to studying case-level effects, specifically, offende r characteristics on sentencing (Kautt, 2002). Recent studies, however, indicate that larger social or contextual factors such as economic political, and cultural norms influence sentence decisions and outco mes (Richardson & Vines, 1970; Heydebrand & Seron, 1990; Flemming et al, 1992; Ulmer, 2005; Kautt, 2002; Ulmer & Johnson, 2004; Britt, 2000; Alschuler, 2005; Weidner et al, 2005; Lambiras, 2003; Hofer et al, 1999). For instance, one study concludes regional disparities are the most troubling finding of the post-FSG era as locational variations among drug offenders actua lly increased or near ly tripled since the enactment of the guidelines (Hofer et al, 1999, p. 303). In this respect, political, economic, and cultural differences among criminal courts ap pear to be just as relevant as offender characteristics in predicting sentence dispar ities (Richardson & Vines, 1970; Heydebrand & Seron, 1990; Flemming et al, 1992; Ulmer & Johnson, 2004; Kramer & Ulmer, 1998). Accordingly, more definitive research is needed to ascertain how larger political and social contextual factors in conjuncti on with individual or intermed iate variables influence legal processes or outcomes (Ulmer, 2005). Theoretical Orientation To address multiple un its of analysis, an inclusive theoretical framework and methodical approach are needed. Donald Blacks (1976) gene ral theory of law is utilized because it focuses on the social structure of a case to explain legal processes and out comes. Social structure of a case includes (1) characteristics of all actors involved in a dispute including offenders, victims, and third parties (police officer s, prosecutors, defense attorneys, judges); (2) the relational distance among these actors (intimates, strangers, so cial equals); and (3) characteristics of the
14 location or setting (neighborhoods, cities, states) in which the case occurs (Black, 1995; Borg & Parker, 2001). Implicit in this th eoretical conceptualization is the idea that the application of law can be explained across different units of analys is. Blacks theory provi des a useful framework by which to examine variations in sentence ou tcomes among individual offenders and at the same time sentence variations across cities, st ates, or societies. His theory therefore simultaneously addresses how offender statuses a nd larger contextual ch aracteristics influence the behavior of law (Borg & Parker, 2001). Acco rding to Black (1976; 1995), the behavior of law can be explained by five aspect s of social life including: vert ical status (stratification or economic rank); integrational status (morphology); cultural status (conventionality); organizational status; and normative status (respectability). As appl ied to this dissertation then, Blacks theory provides a theoretical structure for examining how both th e characteristics of corporate offenders as well as characteristics of the geographical location (federal district) in which the case occurs impact sentence outcomes or fine amounts. In this respect, his theory is well suited for helping to explain which corporat e offenders are subject to more or less law (higher or lower fines) and which federal districts will impose more or less law. Significant Contributions Overall, this dissertation makes at least th ree important contributi ons to the field of criminology. First, the focus on organizationa l defendants provides empirical insight to offenders traditionally overlooked or margina lized in criminology. When Edwin Sutherland introduced the term white collar crime in 1939 he assailed his colleagues for their failure to seriously research the topic. Near ly seventy years later, scholars continue to be critical of the lack of attention afforded corporate crime. Snider (2000) wrote an obituary for corporate crime because of the lack of formal or academic research on the subject. In general, she argues that state regulation of corporate crime has esse ntially disappeared and critical discourse about
15 the subject has become obsolete. Thus, compared to the literature on st reet crime there is a relative dearth of corporate cr ime research. A comprehensive study by Lynch et al (2004) supports the notion of the death of corporate crime. In particul ar, they concluded that white collar crime topics including co rporate crime were significantly underrepresented throughout the criminology discipline in terms of textbook chap ters or pages, journal articles, and PhD curriculum. To be sure, the preponderance of FSG research has examined sentence disparities among individual offenders. Only a handful of studies have analy zed the organizational guidelines (Cohen, 1989, 1991, 1996; Pa rker & Atkins, 1999; Alex ander, Arlen, Cohen, 1999; Murphy, 2002; Piquero & Davis, 2004). While research on individual offenders has been informative, these studies reveal little about f actors that potentially affect sentence outcomes among organizational defendants. By examining cr iminal responses to organizational offenses, this dissertation shifts the balance of cr iminological research toward the topic of corporate crime. A second contribution of this dissertation is it s theoretical focus. Traditionally, critical perspectives have been used to explain the le nient sanctions or responses directed toward corporate offenders. In general, these perspectives assert corporat e status and financial resources negate meaningful or effective criminal le gislation, regulation, and sanctioning (Sutherland, 1949; Clinard & Yeager, 1980; Geis, 1967 & 2002; Quinney, 1980; 1974; Snider, 1990 & 2000; Calavita, Tillman, & Pontell, 1997; Calavita & Pontell, 1990; Chambliss & Seidman, 1982; Friedrichs, 2004; Gruner, 1992). Cr itical explanations have been fruitful and insightful however they have generally been accused of institu ting a dogmatic ideology and using untestable conceptual measures (Akers, 1999). Blacks (1976) theory of law helps overcome some of these criticisms since it offers a host of testable and empirical propositions. Further, his theory is advantageous because it is applicable across diffe rent units of analysis. While his theory has
16 received great attention and been evaluated in a variety of legal outcomes or settings, the author is unaware of any tests using orga nizations as the locus of atten tion. Instead, Blacks theory has been used to analyze individuals and comm unities (Gottfredson & Hindelang, 1979; Braithwaite & Biles, 1980; Myers, 1980; Kruttschnitt, 1980-81; Mooney, 1986; Smith, 1987; Lessan & Sheley, 1992; Borg & Parker, 2000). Additionally, most analyses have only examined the micro-level dimensions of Blacks theory (Lessan & Sheley, 1992, Borg & Parker, 2001). Fewer studies have investigated macrosoc iological dimensions or the eff ect of the social context on the behavior of law (Lessan & Sheley, 1992; Borg & Parker, 2001). Therefore, this dissertation provides a more comprehensive theoretical evalua tion of the organizationa l guidelines, and at the same time a more extensive examination of Blacks theory of law since it examines organizational offenders and incorpor ates aggregate measures of theo retically relevant concepts. A third contribution of this dissertation is it s focuses on the importance of contextual or structural factors on criminal sentencing. Acco rding to Kautt (2002), studies of the federal sentencing guidelines continue to neglect the influence of larger environmental factors on sentence outcomes. This failure to empirica lly recognize the link betw een contextual and caselevel variables on sentencing remains a major flaw in criminological research (Ulmer & Johnson, 2004; Kautt, 2002; Britt, 2000). To this end, this dissertation assesses th e effect of federal districts on sentence outcomes to ascertain if interdistrict variations exist. Additionally, this dissertation uses hierarchical linear modeling (HLM) statistical techniques in order to more accurately assess multilevel effects. Use of singl e-level regression models such as logistic regression or OLS are inappropriate primarily because they misestim ate the true effects of either contextual or offender factors on the depende nt variable (Bryk & Raudenbush, 1992; Ulmer & Johnson, 2004; Kautt, 2002; Weidner et al, 2005; Britt, 2000). Since HLM permits explanatory
17 factors to vary across and within different levels of analysis, it can more precisely estimate how district and organizational ch aracteristics affect sentence ou tcomes. In other words, HLM provides a more accurate picture of what fact ors impact sentence outcomes (Britt, 2000). Collectively, the focus on sentence outcomes directed toward corporate offenders, the reliance on a more inclusive theoretical framewor k, and the application of a multilevel analysis using appropriate statistical tech niques allows the dissertation to significantly contribute to the criminological literature in general and to corporate crime research more specifically. In order to accomplish these tasks, this dissertation proceeds as follows. Chapter 2 provides a general overview of the FSG with a specific focus on the organizational guidelines. Originally, corporations were subject to the same penalt ies as individual offenders, however by 1991 the Sentencing Commission created chap ter eight of the FSG to eff ectively deal with corporate offenders. In particular, the organizational gui delines sought to provide substantially stronger punishments commensurate to the harm caused, while at the same time promote good corporate citizenship by providing sentence incentives for organizations with effective compliance programs (Desio, 2004; Nagel & Swenson, 1993; Murphy, 2002; Ferrell et al, 1998; Izraeli & Schwartz, 1998). This chapter further discu sses the underlying basis for the organizational guidelines and focuses on the legal factors that should theoretically guide the se ntencing process. Chapter 3 provides an overview of Blacks (1976) theory of law. He defines law a governmental social control by state over its ci tizens (p. 2). His theory offers general propositions intended to predict how the law varies in respon se to stratification, morphology, culture, organization, and social c ontrol. Black asserts his theory of law is applicable across all stages of the legal process including investigation, arrest, prosecution, conviction, and sentencing. This chapter further explains each of these statuses at both the individual and
18 contextual units of analysis and concludes by summarizing empirical as sessments of Blacks theory. Chapter 4 focuses on existing corporate cr ime research that lends support for some of the propositions outlined by Blacks theory of law. Fo r the most part, such literature has focused on the impact of offender characteristics on sentence outcomes. Fewer studies have examined the effect of structural factors of sentence outcome s. Particular attention will be given to the potential impact of federa l district political climates (used as a measure of Blacks dimension of cultural status) on sentence outcomes. Finally, Ch apter 5 outlines the analytical framework for this dissertation. More specifically, this ch apter discusses the depe ndent and independent variables, describes the data s ources, explains that advantages of using hierarchical linear modeling (HLM) and outlines the hypotheses pr edicted by Blacks theo ry. The dependent variable measures total fine amounts imposed ag ainst guilty corporations sanctioned under the federal sentencing guidelines. The independent variables measures legal factors (offense seriousness and corporate culpability variables) as well as extralegal factors related to offender and federal district characteristic s. In particular, the extraleg al measures capture the social statuses predicted by Blacks theory. The financ ial status (vertical status or rank); ownership structure (integrational status); number of employees (organizationa l status); and criminal history (respectability) of corporations are measured to determine th eir influence on fine amounts corporate. At the federa l district level, measur es of economic characteristics within the district (vertical status or rank); integr ational status; political climate (cultural status); organizational status; and crime and homicide rates (normative stat us or respectability) ar e gathered to discern the extent to which these variables impact fines. Since this dissertation conducts a multilevel analysis, several data sources will be incorporated. Case level information (legal vari ables and fine amounts) as well as characteristics
19 of organizational defendants are obtained from data collected by US Sentencing Commission and disseminated by the Inter-University Consortium fo r Political and Social Research (ICPSR). Most of the contextual information is obtaine d from FEDSTATS which provides general census data for each federal district. Additional sour ces are used to better cap ture characteristics of each federal district.
20 CHAPTER 2 FEDERAL SENTENCING GUIDELINES FOR ORGANIZATIONS (FSGO) Federal District Courts By exam ining the federal sentencing guide lines for organizations (FSGO), this study focuses on decisions rendered in fe deral district courts. The federal judiciary is divided into four separate court systems: Magistrate Courts (low est federal court); Distri ct Courts; Courts of Appeal, and the United States Supreme Court (represents highest federal court). Currently, there are 94 separate federal districts occupied by approximately 678 judges. District boundaries are intentionally state-contained so that no district court crosses state bo rders. Within the fifty states, there are 89 federal districts. Each state has at least one district but no more than four districts. Most states (26) have only a single district, whereas twelve states are divided into two districts, nine states have three districts, and three states (California, Ne w York, and Texas) contain four separate federal districts. For example, Califor nia has a Southern, Centra l, Northern, and Eastern federal district. Beyond the US territorial district courts, there are five additional districts that represent Washington, DC, Puerto Rico, Guam, th e Virgin Islands, and the Northern Mariana Islands. Each district has two or more judges w ith the Southern District of New York having the largest number of judgeships with twenty-eight (Allian ce of Justice, 2007). Federal district courts are c onsidered the trial courts or the workhorse of the federal judiciary. Over the past five years, there have been well over 300,000 criminal and civil cases filed each year in district cour ts. In 2006, there were a tota l of 326,401 cases filed in federal district courts with judges hear ing an average of 102 criminal and 374 civil cases (Administrative Office of the United States Courts, 2007). Since they hear and dispose the majority of federal cases, district courts are often considered the gatekeepers of the federal judiciary (Lyles, 1997, Rowland & Todd, 1991). Judgments rendered in these courts are generally fi nal decisions as few
21 cases proceed to the higher courts. In this way, the federal district courts are the final arbiter in approximately 90 percent of all federal cases (Sch erer, 2005). Prior to th e enactment of the FSG, the federal district courts utili zed an indeterminate sentence stru cture that afforded judges wide discretion in determining the type and length of senten ces (Podgor & Israel, 1997). Consequently, considerable sentence disparitie s existed as sentence outcomes were generally more punitive against certain types of defenda nts namely the poor and ethnic minorities. Ideological biases of judges and perceptions of discrimination raised im portant questions about equality of the federal courts. Historical Development of the Federa l Sente ncing Guidelines (FSG) In 1984, Congress enacted the Sentencing Reform Act (SRA) for the purposes of (1) significantly reducing discriminatory senten cing discrepancies among similarly situated offenders -offenders who committed similar crimes and had equivalent criminal histories; (2) limiting unfettered judicial discretion; (3) increa sing sentence uniformity, certainty, and fairness; and (4) enhancing the severity of punishments in an effort to deter, incapacitate, and justly sanction offenders (USSC, 2007; Gruner, 1992; Nagel & Swenson, 1993; Lambiras, 2003). The SRA established a full-time, seven-member Unite d States Sentencing Commission to serve as an ongoing independent body of the judicial bran ch (USSC, 2007; Desio, 2004). By 1987, the Commission disseminated the fede ral sentencing guidelines which established a determinate and formalized sentencing structure for federal district courts. By and large, the FSG reflect a Weberian or formal legal model of criminal sentencing. According to this perspective, sentence outcomes should be dete rmined solely by legal factors such as the seriousness of the offense, prior crim inal history of the offender, and the amount of harm caused by the crime (Dixon, 1995). Precedent and established axioms should guide the decision-making process in order to ensure judi cial impartiality and objectivity (Dixon, 1995;
22 Segal & Spaeth, 1993; Yarnold, 1992). In this manner, criminal courts should operate like highly rational machines that generate predictable and consistent outcomes for all offenders irrespective of extralegal factors such as socioeconomic st atus, race, gender, or geographical location. Indeed, offense seriousness and prior criminal histor y are two primary factors that direct judicial discretion under the FSG. In short, the FSG use a point system to dete rmine sentence outcomes. Point totals are calculated using a basic three step pr ocess. First, judges determine a base offense score commensurate to the crime of conviction. For each offense, the Commission created base offense scores that serve as a starting poi nt in the sentencing process. Secondly, an offense seriousness score is calculated. In order to determine offense seriousness, judges consider aggravating and mitigating circumstances and ma ke upward adjustments (add points to the base offense score) or downward adjust ments (subtract points). Judges examine legal factors such as the type of weapon used during th e commission of a crime, the role of the offender, the severity or harm of the crime, and whether the defendant c ooperated with prosecutors or legal authorities. Once the offense seriousness score is calculated it is added or subtracted to the base offense score. This determination yields a final offense score Under the FSG, final offense scores range from 1 point (least serious type of offenses) to 43 points (most serious). The final step of the FSG sentence decision process i nvolves a computation of a criminal history score. Judges assess scores ranging from 1 point (first-time offenders) to 6 points (extensive history of offending). Using a two-dimensional sentenci ng table, judges measure the poi nt at which the final offense score and the criminal history level intersect. This point of intersec tion specifies a sentence range (in terms of months to be served in pris on). Judges are given di scretion to sentence an offender anywhere within the mandated sentence range (USSC, 2007). Under the FSG, any
23 departures outside the mandatory sentence range are subject to automatic appellate review. Judges must publicly disclose in court or through written opinion why they departed from the recommended sentencing range. Through thes e systematic sentencing instructions, the Commission attempted to ensure accountability for all federal sentencing decisions. While the Guidelines were originally drafted with individual offenders in mind, by 1991 the Commission created chapter eight of the FSG to addre ss organizational or co rporate offenders. Historical Development of Fede r al Sentencing Guidelines for Organizations (FSGO) Federal sentencing guidelines for or ganizations (FSGO) were developed to more adequately sanction offenses commonly referred to as corporate crimes. In general, corporate or organizational crime refers to illegal or socia lly injurious acts committed by corporate officials or employees primarily to promote or benef it corporate inte rests (Clinard & Quinney, 1973; Kramer et al, 2002). When considering the develo pment of the organizati onal guidelines, the US Sentencing Commission confronted two immediate problems rela ted to corporate sentencing. First, fines imposed against corporate criminals were historically weak a nd insignificant. Prior to 1984, guilty corporations or organizations were subject to the same penalties as individual offenders. In most instances, the maximum co rporate fines were miniscule and woefully inadequate (Gruner, 1992; Alexa nder et al., 1999). For instance, 60 percent of the fines imposed against criminal organizations totaled less than $10,000 (Alexander et al., 1999). Monetary penalties rarely equaled the losses or harm cause d by the crime (Gruner, 1992; Alexander et al., 1999; Nagel & Swenson, 1993). That is, the illega l financial gains easily exceeded the criminal fines imposed against corporate offenders. Th e underlying message of the federal sentencing structure was crime does pay and criminal fine s were just a minor business cost (Gruner, 1992). Thus, the Commission faced Congressional and public concern over the lenient sanctions imposed against corporate offenders. The prev ailing perception was that white collar crime
24 offenders were afforded preferential treatmen t and did not receive punishments as severe as those meted out against street criminals (Nagel & Swenson, 1993). A second flaw in corporate sentencing was noticeable sentence disparities rela ted to corporate size a nd/or financial status (Cohen, 1989). The Commission found many cases where organizational defendants committed nearly identical offenses yet one corporation would be fined over twice the loss of the crime while another organization would be fined only a fraction of the loss (Nagel & Swenson, 1993; Izraeli & Schwartz, 1998) Cumulative evidence demonstrat ed that corporate offenders were significantly underpunished and di d not receive penalties commens urate to the seriousness of their crimes. Thus, a primary objective of the Commission was to develop tougher more punitive sanctions in an effort to deter corporate cr iminality and increase perceptions of justice. At the same time, however, the Commission faced considerable opposition from businesses as well as political interests including the White House and Congress (Lofquist, 1993; Rodriquez & Barlow, 1999). A primary concern was that the overly aggr essive organizational guidelines would significantly hinde r corporate interests. The Reag an administration in the early 1980s ushered a political era that endorsed probusiness policies. These la issez faire economic principles supported deregulation and cooperative crime control polic ies. Strict regulation and sanctions were considered counterproductive to free market corporate activity (Snider, 2000, 1990; Calavita & Pontell, 1990). During the development phases of the organizational guidelines, business and political interests successfully reduced many of the tough fine provisions initially proposed by the Commissi on (Lofquist, 1993; Rodriquez & Barlow, 1999). Also, the Corporate Defense Attorney Worki ng Group was victorious in influencing the Commissions decision to institute generous incentives into the or ganizational guidelines (Nagel & Swenson, 1993). In many ways, the Commission was placed into a contradictory role of
25 needing to significantly increase corporate sanctions but at the same time provide opportunities or inducements to mitigate corporate criminal responsibility and offer sentencing leniency (Rodriquez & Barlow, 1999; Laufer 2002; Nagel & Swenson, 1993). After conducting extensive researc h, publishing a series of form al and informal drafts, and holding public hearings with key groups includ ing judges, practitioners, and academics, the Commission promulgated the Federal Sentencing Guidelines for Organizations (FSGO) on November 1, 1991. The Commission specified th at organizations incl uding corporations, partnerships, associations, government agenci es, and non-profit companies were subject to guideline provisions if they committed any fe lony or Class A misdemeanor offense (USSC, 2007). These included offenses such as co rporate fraud, embezzlement, antitrust, and environmental violations. Despite pressures from the business comm unity to make the organizational guidelines non-binding, the Commi ssion decided to make them mandatory like the individual guidelines (Nagel & Swenson, 1993) They reasoned that organizations should be subject to the same standards as individual offe nders and believed that corporations should not benefit from the traditional leniency afford ed by the courts (Nagel & Swenson, 1993). By making the FSGO mandatory, the Commission sought to dramatically narrow the scope of judicial discretion and ensure c onsistent uniform punishments absent the influence of extralegal organizational characteristics. Similar to the individual Guidelines, the organizational guidelines instructed judges to base sent ence decisions on two legally re levant factors: (1) offense seriousness and (2) corporate criminal culpability. Federal Sentencing Guidelines for Organiz ations (FSGO) Sentencing Process Under the F SGO, fines are the primary sanction imposed against corporate or organizational defendants. Fine amounts are de termined by multiplying (1) a base fine score with (2) a culpability score. Base fine sc ore corresponds to offense seriousness which is
26 determined by choosing the greatest of three f actors including (a) an offense score, (b) the pecuniary gains from the offense, or (c) the pecuniary losses caused by the crime (USSC, 2007). To determine an offense score, judges use proced ures similar to the indi vidual guidelines. For each crime, the Commission created an initial offense score that se rves as a starting point in the sentence decision process. Judges then make upward or downward adjustments to this score depending on aggravating or mitigating circumstances. Once adjustments are made and a final offense score is calculated, judges consult a FSGO fine table that specifi es fine amounts (see Table 2.1 for partial list of offe nse levels and corresponding fine amounts). Final offense scores range from 6 or fewer points (equals corporate fine amount of $5,000) to 38 or more points ($ 72.5 million fine). This fine amount is measured against the pecuniary gain and loss caused by the crime. Whichever amount is greatest determines the base fine score To illustrate, consider a hypothetical case involving a corporate offender who commits embezzlement worth $1,000,000 that results in a p ecuniary loss of $500,000 to victims. If this offender is assigned a final offense score of 20 points then a $650,000 fine will be imposed. Comparing the three dollar amounts, the pecuniar y gain ($1,000,000) is greater than the fine ($650,000) and pecuniary loss ($500,000). Accordingl y, the base fine score for this offender would be $1,000,000. This dollar amount is then multip lied by the offenders culpability score. Table 2-1. Partial list offe nse level fine table Offense level Fine amount 6 or less $5,000 7 $7,500 8 $10,000 10 $20,000 15 $125,000 20 $650,000 25 $2,800,000 30 $10,500,000 38 or more $72,500,000 Source: U.S. Sentencing Commission, Guidelines Manual 8C2.4, 2007
27 An offenders culpability score determines their level of responsibility for a criminal act. Each organization begins with a culpability score of five points. Aggrav ating factors give cause for upward adjustments to the culpability score (as high as 10 or more points), while mitigating factors justify downward adjustments (as low a 0 or fewer points). Sent encing judges weigh four aggravating factors including (1) organizational tolerance of or i nvolvement in criminal activity, (2) obstruction of justice, (3) prior history of offending for similar crimes, and (4) violation of any existing court orders or injunctions. The FS GO authorize judges to give special attention to the role of high ranking managers when considering these aggravating factors. That is, judges must determine whether officials participated in, condoned, tolerated or willfully ignored criminal conduct; or if they violated their position of trust (Ferrell et al., 1988). At the same, judges contemplate four mitigati ng factors including (1) the existence of an effective compliance or ethics program, (2) organi zational cooperation with investigators, (3) the presence of self-reporting to officials upon discov ery of criminal activit y, and (4) acceptance of responsibility. In essence, these mitigating f actors lessen corporate re sponsibility and help reduce fine amounts. Final culp ability score establishes minimu m and maximum multipliers. For instance, a culpability score of 0 yields a minimum multiplier of .05 and a maximum of .20. A final culpability score of 10 includes a mi nimum multiplier of 2.00 and maximum of 4.00 (See Table 2.2 for a list of the minimum and maximu m culpability multipliers). These culpability multipliers are combined with the base offense score to provide a recommended or mandated fine range. Sentencing judges can impose a fine anywhere within this range. However, similar to the individual guidelines any departures from the recommended fine range is automatically open to appellate review (Alexander et al., 1999).
28 As a hypothetical example, a corporate defenda nt with a base offense score of 6 (fine amount equals $5,000) and a culpability score of 0 points (a minimum and maximum multiplier of .05 and .20) would have a recommended fine range of $250 to $1,000. A guilty corporation with a final base score of 10 points ($20,000 fi ne) and a culpability score of 5 (1.00 to 2.00 multipliers) would produce a fine range from $20,000 to $40,000. The most egregious organizational offender with a base offense score of 38 Table 2-2. Minimum and maximu m culpability multipliers. Culpability score Minimum Multiplier Maximum multiplier 0 or less .05 .20 1 .20 .40 2 .40 .80 3 .60 1.20 4 .80 1.60 5 1.00 2.00 6 1.20 2.40 7 1.40 2.80 8 1.60 3.20 9 1.80 3.60 10 or more 2.00 4.00 Source: U.S. Sentencing Commissi on, Guidelines Manual 8C2.6, 2007 points ($ 72.5 million fine) and a culpability level of 10 (2.00 to 4.00 multip liers) would receive a fine between $145,000,000 and $290,000,000 (Table 2.3 gives an example of hypothetical fine ranges). Beyond fines, the organizational guidelines in clude other sanctions such as probation, orders to pay restitution, and orders to issue public notices and apologizes. In all cases, these other punishments can be imposed in additional to criminal fines in order Table 2-3. Hypothetical fine ranges for various offense leve ls and culpability scores Offense level Culpability score Minimum multiplier Maximum multiplier 6 pts = $5,000 0 ( .05) = $250 ( .20) = $1,000 10 pts = $20,000 5 (1.00) = $20,000 (2.00) = $40,000 38 pts = $72.5 Million 10 (2.00) = $14 5 Million (4.00) = $290 Million
29 to remedy the harm caused by such offenses. Also, it should be noted that the FSGO are applicable to corporations or organizations ope rating primarily for criminal purposes. Indeed, the FSGO stipulate that unlawful organizations should be fined sufficient enough to divest the criminal firm of all of its assets (USSC, 2007). Underlying Philosophy of the Federal Sentenci ng Guidelin es for Organizations (FSGO): Good Corporate Citizenshi p and Compliance Programs Unlike the individual guidelin es, punishment and incapacitatio n were not the immediate goals of the FSGO (Murphy, 2002). Instead, restit ution and deterrence were prominent goals. Accordingly, fine provisions and complianc e programs became the centerpieces of organizational guidelines (Nagel & Swenson, 1993; Murphy, 2002; Desio, 2004; Ferrell et al., 1998). Fine provisions were desi gned to remedy the harms caused by corporate offenses and offset any illegal gains. Emphasis on complian ce programs was intended to promote and reward good corporate citizenship (Desi o, 2004; Nagel & Swenson, 1993; Fe rrell et al, 1998). Indeed, the culpability component of the sentence instructions was intended to evaluate the effectiveness of compliance programs prior to criminal activity and help absolve or reduce corporate liability for violations committed by rouge employees (Nagel & Swenson, 1993; Podgor & Israel, 1997; Laufer, 2002). The Commission viewed compliance programs as an important means of institutionalizing ethical behavior and promoting a strong commitment to legal standards (Ferrell et al., 1998; Izraeli & Schwartz, 1998; Desio, 2004; Mu rphy, 2002; Nagel & Swenson, 1993; Laufer, 2002). Through the maintenance of compliance policies any criminal conduct could be considered an aberration rather than the co rporate norm. Beyond compliance programs, the FSGO also placed substantial responsibility on high level officials or managers to exercise due diligence in terms of establishing and maintain ing compliance programs (Desio, 2004; Ferrell et al, 1998). Specifically, the organizational guidelin es stipulate that managers should institute
30 proactive objectives intended to (1) prevent and detect cr iminal activity, (2) offer continuous training and communication at all levels, (3) provide constant ev aluation by high-level managers, and (4) outline clearly defined steps to res pond to any violations (Lofquist, 1993; Nagel & Swenson, 1993; Ferrell et al, 1998; Murphy, 2002; Desio, 2004). In many ways, the underlying organizational guidelines philosophy parallels a carrot and stick approach to sentencing (Izraeli & Schwartz, 1998; Nagel & Swenson, 1993). Fine provisions escalate as organizational culpability increases, however corporations are rewarded for taking steps to help prevent and deter mi sconduct. The Commission believed that fine penalties should not only be propor tional to the costs of the crim e but should also deter future criminality (Nagel & Swenson, 1993). As such, mitigating credits and sentencing reductions were offered during every stage of the organiza tional sentencing process (Laufer, 2002; Izraeli & Schwartz, 1998). The organizational guidelines essentially rewarded corporations for adopting and preserving effective crime control policies. In so doing, corporate offenders could expect fine reductions, up to 95 percent in some cases (Desio, 2004). Evaluations of the Federal Sentencing Guidelin es for Organizations (FSGO) Overall, the organizational guidelines appear to have had a positive and immediate impact on corporate crime sentencing. Over a ten ye ar period, nearly 1500 FSGO cases resulted in $2.3 billion in fines, approximately $279 million in restitution, and over 3000 years of probation (Murphy, 2002). Also, the average criminal fine significantly increased since the implementation of the organizational guidelines. During the pre-guideline era, the mean fine imposed was $1.9 million but increased to $19.1 af ter the institution of the organizational guidelines (Alexander, et al., 1999). The percen tage of fines that exceeded $1 million increased from 37 percent (pre-organizational guidelines) to 59 percent (Alexa nder, et al., 1999). Nevertheless, problems persist with the organizational guidelines.
31 First, corporate convictions remain minimal. During a five-year period (1996 to 2000), less than one percent of all federal sentences were imposed against organizational defendants (Murphy, 2002). One study concluded that monetary penalties did not signi ficantly change after controlling for the harm caused by corporate offe nses (Parker & Atkins, 1999). That is, fine amounts remained relatively stable during the pr eand post-guideline eras when comparing the amount of restitution or loss incurred by cor porate crimes. Another criticism of the organizational guidelines is the heavy emphasis on rewarding corporate compliance programs. Inadvertently, the FSGO created a d ark side as corporate managers have become very eager to flip on less culpable employees in order to escape punishme nts or fines (Laufer, 2002). Finally and perhaps more importantly, research indi cates that extralegal characteristics continue to influence sentence outcomes (Piquero & Davis, 2004; Beck & OBrien, 2000). Small and closely-held firms (corporations controlled by a single owner or small group of owners whose stocks are not publicly traded) continue to repr esent the majority of defendants sentenced (Beck & OBrien, 2000). Furthermore, financial stat us and ownership stru cture have yielded differential fine outcomes among corporate defendants (Piquero & Davis, 2004). Despite efforts to eliminate unw arranted sentence disparitie s, there appears to be an empirical link between extralegal characteristic s and sentence outcomes. Some question whether the federal guidelines can truly use rationalized sentencing policies to totally eliminate judicial discretion (Savelsberg, 1992; Ka utt, 2002; Ulmer, 2005). Subjective evaluations of offenders coupled with larger contextual influences ma y unduly sway sentencing processes and decisions (Ulmer, 2005). This dissertation uses Donald Blacks (1976) general theory of law because it offers propositions for predicting variations in sentence outcomes (fine amounts) in response to characteristics of corporate offenders and charac teristics of the federal districts in which the
32 offenses occur. In this way, his theory helps explain how sentence disparities may vary across different levels of analysis. In particular, Black asserts the quantity of law is affected by five social statuses including vertical status (stratifica tion or economic rank); integrational status (morphology); cultural status (conventionality); organizational status; and normative status (respectability). The next chapter provides an overview of Blacks theory of law.
33 CHAPTER 3 DONALD BLACKS THEORY OF LAW This chapter exam ines Donald Blacks (1976) general theory of law. In short, Black defines law as governmental social control by the state over its citi zens. He maintains that law is quantifiable and can be explaine d by variations in social stat uses or conditions (Black, 1976, 1995; Greenberg, 1983; Doyle & Lu ckenbill, 1991; Gottfredson and Hindelang, 1979; Lessen & Sheley, 1992). One of the benefits of his theory is that it focuses on the social structure of a case. According to Black (1976; 1995) the social structure is shaped by five statuses or aspects of social life including: verti cal status (stratification or rank); integrational status (morphology); cultural status (conventionality); organizational status; and norm ative status (respectability). Case includes (1) characteristics of the actors in volved in a conflict (i.e., offender, victim, 3rd party participants such as attorneys or judges) (2) the relational ties or distance among these actors (i.e., intimacy, strangers, or equals), and (3) characteristics of the geographical location or setting (neighborhoods, cities, or states) in which the conflic ts arise (Black, 1995; Borg & Parker, 2001). Taken together then, the social stru cture (five social status es) of the case (actor characteristics, relational ties, and larger contextual factors) helps predict the quantity of law. In this way, Blacks theory can be applied across diffe rent units of analysis. His theory can help explain legal variations among individual offenders as well as variations among neighborhoods, cities, or states. This dissertation utilizes Blac ks theory to determine whether the quantity of law or sentence outcomes (fine amounts) varies by the social statuses or characteristics of (1) corporate defendants and (2) federa l districts. More specifically, Blacks theory is used to predict the extent to which corporate offenders encounter more or less law (higher or lower fines) and districts impose more or less law (f ines) after controlling for the impact of legal factors. This chapter reviews his theory of la w and summarizes each of the five social statuses
34 that affect the behavior of law. It concludes by summarizing empirical st udies of Blacks theory of law conducted across different levels of analysis. Overview of Blacks Theory of Law Blacks (1976:x) theory of law provides a quantitative, predictive and general theory that can explain the conditions wh erein law will be used or app lied. Law, according to Black, reflects governmental social cont rol over its citizens in the form of legislation, litigation, and adjudication. Law includes the reporting of cr ime, arrests, prosecutions and sanctions. Black (1976:3) contends any initiation, invocation, or app lication of law increases its quantity. Thus, law is considered a quantitative variable that can be measured by the scope and rate of its usage. Any complaint, investigation, arrest, indictment, prosecution, or punishment constitutes more law compared to the absence of a complaint, in vestigation, arrest, and so forth. Black also asserts any increment or increase in severity re presents the presence of more law. Thus, an offender detained during pretrial procedures indicates the use of more law than an offender released on bond. Incarceration sign ifies more law than probation. Similarly, a ten year incarceration sentence reflects more law than a five year sentence. According to Black, every aspect of social life behaves including cities, governments, organizations, and families. Law is no different. It behaves or varies across time and place. It changes from one decade to another, one year to the next, or even day to day, hour to hour. Law behaves differently across societies, regions, co mmunities, or courts. Since law is a social phenomenon, Black argues it has no concep t of human nature and can be explained irrespective of individua l or psychological motiv ations (Black, 1976:7). In other words, his theory is not concerned with individual motivations such as rationality or goal-oriented pleasure seeking behaviors (Gottfredson & Hindelang, 1979). Rather, Black attempts to explain how the behavior of law is invariably tied to the larg er social or normative climate (Lessan & Sheley,
35 1992; Greenberg, 1983). Accordingly, he presents a series of pred ictive statements intended to anticipate the quantity and direction of law in any and every social setting or wherever it is possible to measure law and other aspects of soci al life (Black, 1976:6). Black contends that the behavior of law is dependent upon five social statuses or aspects of social life including stratification (rank); morphology (int egrational status); culture (c onventionality); organization; and social control (respectability ). In terms of the applicatio n of the law, it can be applied upward against offenders with greater social statuses or downward against those with fewer statuses. For Black (1976: 21), downward la w is more common acro ss all settings since people with lower statuses are subjected to more law or greater punishments. Lower status offenders have a greater likelihood of being arre sted, convicted, or hars hly punished. At the same time, Black argues that these five statuses help predict structural varia tions in the law. In other words, his theory helps explain why some locations use more law (i.e., drafting legislation, reporting crime, making arrests, and imposing harsh sentences) th an other areas. The following section further examines the relationship betw een law and each of these social statuses. Stratification (Vertical Status or Rank) According to Black (1976), strati fication represents the vertical aspect of social life or the uneven distribution of material goods. More succinctly, stratification symbolizes wealth inequality. In terms of the relationship between stratification and the law, Black (1976:17, italic in original) predicts law varies directly with rank All else being equal, lower ranks are more vulnerable to the application of the law. For instance, Black hypothesi zes that the poor, women, and nonwhites are more susceptible to the law th an the rich, men, and whites primarily because the former groups have less rank within the stratification system. In te rms of economic rank, the poor are more likely to have th eir behaviors defined as criminal. They are more likely to be subjected to criminal investigation, arrest, prosecution, and harsh sanctions. Thus, the poor have
36 traditionally encountered more law or been severely sanctioned throughout the criminal justice process. By comparison, the wealthy have traditiona lly been immune from th e applications of the law. The rich are less likely to have complaints filed against them, less likely to be arrested or convicted, and less likely to have punitive measur es imposed against them. In this way, higher ranks have less law imposed against them. For Black (1976), the proposition of rank holds constant for groups, communities, regions, or even entire societies. Locations have various degrees of wealth and therefore possess different ranks. In this respect, the law varies with the rank of a location. According to Black, poor societies have traditionally had less law (i.e., criminal laws, cour ts, and lawyers) than wealthier industrialized societies. For inst ance, he notes that the African kingdom of Dahomey was largely inhabited by poor groups and thus largely untouche d by the law. Similarly, he maintains that poor communities in the United States have pr oportionately used le ss law than wealthier suburban areas. Residents in wea lthier areas are more likely to re port crimes and to have their complaints result in arrests. Offenders who comm it crimes in wealthier areas are subject to more punitive sanctions compared to those who offend in poorer communities. As such, the more wealth within a geographical setting the more law typically used. Morphology (Integrational Status) Beyond stratification, Black (1976) argues morphology affects the application of the law. He refers to morphology as the horiz ontal aspect of social life or a societys division of labor, network of interactions, intimacy, and integration, (p.37). In terms of integration, Black argues that social life has center and periphery. Those that participate near th e center of social life increase their level of integration, while peop le that operate at the peripheral are weakly integrated or marginalized. Morphology can th erefore be measured by the extent to which people participate in social life or the scope, frequency, and lengt h of their interactions with
37 others. For Black, factors such as employment, marital status, and residential stability help determine levels of integration. The employed, married, and residentially stable have more integrational status than the unemployed, single, vagrants, and transients. Accordingly, Black (1976:51) asserts law varies i nversely with the integration of the offender. Marginalized offenders are subject to more law, while the integrated encounter le ss law or enjoy legal advantages. A crime by a marginalized offende r (i.e., unemployed, single, or transient) is considered more serious than an offense committed by an integrated defendant (i.e., employed, married, community ties, or residentially stable). Moreover, a marginalized offender is more likely to have the police called against him or her, to be arrested, prosecuted, or severely punished. Black (1976:55) concludes that at ever y stage of the legal process, a marginal person is more vulnerable to law a person without work, a family, or other involvements is more likely to get into troubl e with the law. At the structural level, Black (1976) predicts that law varies with differentiation. In particular, more law exists in locations with gr eater exchange networks, divisions of labor, and economic transactions. Conversely, there is le ss law in regions with limited exchanges of property or services, divisions of labor, or economic transactions. In relatively simple societies, people are more likely to participate more fully in each other lives. As a result, less law exists because people are unwilling to invoke the law ag ainst intimates such as family members, close friends, neighbors, or coworkers. In this respect, structural inti macy leads to immunity from the law. For example, Black points out simple nomadic societies have relatively few contract or property laws since their economic system is relatively rudimentary. However, with economic expansion more law is needed to address vari ous issues of ownership and rights. Contract, property, and criminal laws generally emerge in industrialized or capitalistic societies as labor
38 becomes differentiated and excha nges become more impersonal. Black (1976) points out that more law emerged in New York City during the early nineteenth century because of industrialization and population growth. Severa l new legal agencies and inspectors were developed to address a host of i ssues including tobacco, flour, fish, restaurants, and alcohol regulations. According to Black, social differentiation increases social distance meaning people essentially become strangers. Consequently, citizen s must resort to the la w or independent third parties to settle grievances. Thus, Black ( 1976: 40) asserts that as the division of labor increases, so does the severity of law. Differe ntiated societies use more law or impose harsher punishments, while simple societies appl y less law or less punitive sanctions. Culture (Conventionality) A third social status that a ffects the quantity or direction of the law is culture. Black (1976:61) defines culture as the symbolic asp ect of social life in cluding values, norms, languages, ideologies, folklores or conceptions of right and wrong. Culture also includes sciences, technologies, religions, pol itical parties, and fine arts (i.e., music, literature, painting, arts, or clothing). For Black, certain cultural stat uses are more conventional simply because they occur or appear more frequently. For example, in the United States Protestants or Catholics are more conventional than Jewish parishioners si mply because there are more Protestants and Catholics. Similarly, high school graduates ar e more conventional than dropouts; a bourgeois lifestyle more conventional th an a bohemian lifestyle; De mocrats and Republicans more conventional than Communists; and tobacco smoke rs are more conventional than cocaine users (Black, 1976: 68). With respect to the behavi or of law, Black (1976: 69, italic in original) concludes law is greater in a direction toward less conventionality than toward more conventionality In other words, law is generally applied downward against the unconventional or those with the least amount of culture. Crimes committed by unconventional offenders are
39 generally considered more serious particularly if they commit an offense against a conventional victim. The unconventional are more likely to be arrested, convicted, and receive a harsh sentence. Therefore, Black states political radi cals are subject to more law or tougher sanctions than Democrats or Republicans; ethnic minorities subject to more law than whites; the uneducated are more susceptible than the educ ated; and unconventiona l religious followers encounter more law than more co mmon religious worshippers. He concludes that at every stage of the criminal process [a]ny person who is unconventional in his dress, speech, manner, ideas, or anything else, is mo re vulnerable to law of ever y kind, (Black, 1976:71). All locations have distinct cultural or normative climates. Communities, cities, states, and regions generally have dominant political, religious, ideologica l, moral, linguistic, or ethnic norms. For instance, in the United States some st ates are considered red or Republican states because they have more Republican politicians, vot ers, and policies, while others are blue or Democratic states because Democr atic political ideologies are more frequent. With respect to the behavior of law, Black argues the law vari es directly with the cultural distance between offender and location. In particular, the greater the cultural distance betw een an offender and the larger social setting, the more law they enc ounter primarily because they are considered unconventional by cultural standards. For instance, Black (1976:71) points out than in colonial Tanzania the harshest sentences were imposed against the most unconventional tribes (Baraguyu) while members of the most convent ional tribe (Kaguru) encountered the least amount of law. In the United States, the ethn ic composition of a location is an important determinant of the application of the law. Black asserts that in setti ngs where everyone shares the same ethnicity, language, religion, or custom s, there is little or no law primarily because there is limited cultural diversity or conflict. In essence, the cultural distance between the
40 offender and cultural climate is minimal. Black (1976) states Italian police officers are less likely to arrest an Italian offender than an Af rican American. Similarl y, Italian judges are less likely to convict or severely punish an Italian defendant compared to other ethnic defendants. The same applies to other ethnic enclaves in the United States. In American Chinatowns, disputes between Chinese and non-Chinese individua ls result in more law or litigation than matters between fellow Chinese (Black, 1976). Howe ver, as the cultural distance increases so does the presence of law. Thus, in the United St ates ethnic offenders that commit crimes in predominantly white areas have generally been subjected to the most severe punishments because they are viewed as unconventional in te rms of ethnicity. In this respect, cultural uniformity or intimacy provides immunity from the law while cultural diversity or distance increases ones vulnerability to the law. Organization Organizational status represen ts the corporate aspect of so cial life or the capacity for collective action. It can be measured by the pres ence and number of administrative officers, the centralization of decision-making, and the abil ity for collective action. According to Black (1976:85), individuals, families, groups, firms, po litical parties, cities, and societies possess various levels of organization. Individuals who belong to several clubs or groups are viewed as more organized than a person with limited affilia tions or memberships. Any group is considered more organized than a single individual since groups by definition have more people. Among groups, those with more employees, functional dutie s, subunits, and admini strators are deemed more organized than smaller organizations with fewer employees, functio nal responsibilities, subunits, or officers. Accordingly, Black (1976:92) predicts that law is greater in a direction toward less organization than toward more organization, (italic in original). All else being equal, individuals are more vulne rable to the law than groups or corporations. For example,
41 Black notes that businesses are more likely to file complaints against indivi duals than vice versa. In such cases, the police are more likely to arre st an individual and the individual defendant is more likely to be prosecuted, convicted, and se verely punished. Among corporations, smaller firms are more susceptible to the law than larger firms since they have less organization. Black (1976:93) concludes that organiz ation provides an immunity from law, and the more organized the offender, the more of this immunity is enjoyed. Structurally, Black (1976:86, italic in original) asserts that law varies directly with organization The greater the organizational status of a social setting, the more law found. Simple tribal societies use little or no law because they have fewer public or private organizations (i.e., schools, factor ies, businesses, interest groups associations, or hospitals) and thus fewer legal conflicts. However, as soci eties become more modern they develop more organizations. Black (1976) maintains that the organizational status of a particular location increases during wartime, economic expansion, fore ign trade, or disasters. During such times, new legislation, powers, or rights are needed to centralize decision-making and guide behaviors. Black also notes that in most contemporary so cieties organizational law is so pervasive and voluminous that corporations generally hire la wyers on a permanent basis. Accordingly, as cities, states, or societ ies develop more organizations per capita, they become increasingly litigious and punitive. In other words, the more organizations in a particular location the more law they use. Social Control (Normative Status or Respectability) Finally, Black predicts that social control or normative status impacts the behavior of law. Normative status defines expectations of right and wrong, deviant behaviors, violations, obligations, and abnormalities. Black measures normative status by respectability, or the amount of social control a particular group or individual is subject ed. The more social control
42 one encounters, the less respectable they are gene rally considered. Thus, juveniles often have lower normative status (respectability) than adults because they are subject to greater parental, educational, and governmental controls. In addi tion, Black states that th ose subjected to legal social control (law) are typically viewed as less respectable than those subject to nonlegal or informal controls (family controls). Convicted criminals are viewed as less reputable than a person shunned by family or friends. Indeed, criminals have traditionally been branded, mutilated, or otherwise had their reputations tarnished (Black, 1976). Accordingly, Black (1976:114, italic in original) predicts law is greater in a direction toward less respectability than toward more respectability. He notes social deviants of all kind including criminals, prostitutes, homosexuals, drug addicts, the menta lly ill, or the homeless are more vulnerable to the application of the law at every stage of the criminal justice process compared to nondeviants. The disreputable are more likely to be arre sted, convicted, and severely punished. At the structural level, normative status can be measured by the respectability of the people in its environment (Black, 1976: 112). The more social contro l people are subject to within a particular location, the less respectable it is. Thus, locations with high crime rates, several ex-offenders, and social deviants includi ng prostitutes or drug users generally have less respect than areas with low crime rates and fe w offenders or social deviants. As such, disreputable social settings are more vulnerable to applications of the law. These geographical areas are more likely to be patrolled by police an d to have higher arrest rates. Offenses in disreputable locations are more likely to re sult in prosecution, c onviction, and severe punishments. All else being equal then, more law exists in locations with greater social control because these settings are cons idered less respectability.
43 To conclude, Black advances several empiri cal propositions concerni ng the effect of the social statuses of offender and th e social statuses of geographical location on the behavior of law. He asserts greater or more law will be applied in a downward di rection toward offenders with less (1) vertical status (rank), (2) morphology (int egrational status), (3) culture (conventionality), (4) organization, and (5) normative status (respec tability). Corporate defendants with lower rank, less integration, less conventionality, less organization, and lower respectability should encounter more law or higher fi nes. Simultaneously, Black argue s that federal districts with greater (1) stratification, (2 ) morphology (differentiation), (3) cultural diversity, (4) organizational status, and (5) less respectability will use more law. Before proceeding, it should be noted that Black suggests his th eory not only explains the quantity of law, it can also explai n the conditions in which different styles of law will be used. These styles of law consist of (1) penal, (2) compensatory, (3) therapeutic, and (4) conciliatory forms of social control. According to Black (1976), each has its own definition, language, and logic of deviant behavior as we ll as its own responses to such behaviors. The penal style of control deals primarily with criminal offenses in which formal groups (criminal justice system or government) enact laws to prohibit certain c onduct, decide guilt or innocence, and impose punishments. For the most part, Blacks disc ussion of the quantity of law as related to stratification, morphology, culture, organization, and respectability (socia l control) addresses penal styles of social control. That is, Blacks theory anticipate s the type of criminal offenders that are more or less likely to be subjected to penal forms of control incl uding arrest, prosecution, and/or imprisonment. Under compensatory control, the victim rather than formal groups initiates the proceedings against a defendant. Compensatory styles of control are mo re interested in civil remedies or payments to settle conflicts. Ther apeutic control is also initiated by the victim who
44 seeks support or intervention. The goal of therap eutic control is to offe r assistance to people in trouble. Victims seek therapy from an individual or organization in order to help or improve his or her condition. In essence, the victim desires a return to normalcy. Fina lly, conciliatory styles of control seek social harmony or social repair. Parties involved in a dispute generally search for an independent third party or me diator to help restore damaged relationships to their previous friendly conditions. The goal of conciliation is to work together to develop mutual acceptable resolutions. While these styles of control can be included in the same case (a penal or criminal case may include compensatory or therapeutic elements), Black (1976:5) asserts that in most cases it possible to identify the dominant style. Additionally, he notes that just like the quantity of law, the style of law varies across time and place, so cial setting, and characte ristics of the parties involved in a dispute. For instance, Black anticipates that penal control may be more prominent during one century or decade compared to another, more dominant in one community in relation to another. Some settings may use more punishment (penal), while other areas are more inclined to use compensation or therapy. Likewise, one offender may be more likely to be punished, while another might pay compensation or attend a mediation session (conciliatory). Overall, Black states both the quantity and style of law are aspects of th e behavior of law and it is possible to predict either given information about the social struct ure of a case. Interestingly, the federal sentencing guidel ines for organizations (FSGO) include both penal and compensatory styles of punishments. By and large, the FSGO impose penal sanctions including fines and probation that serve the interests of the g overnment (formal group) rather than a particular victim. However, one of the punishments included in the organizational guidelines is restitution paid to victims. In th is respect, restitution represents a compensatory
45 style of social control. This presents a potentia l empirical or analytical problem because a guilty organization may receive a small fine (less la w or penal) but pay a significant amount of restitution (more law or compensation). Thus, only examining variations in fine amounts to the exclusion of restitution can poten tially miscalculate the amount of law an organization is subjected to. A more complete or accurate a ssessment of the FSGO should therefore examine the impact of both fines and restitution. The remi nder of this chapter di scusses prior tests of Blacks theory conducted at both the indi vidual and structural levels of an alysis. Empirical Tests of Bl a cks Theory of Law Empirical tests of Blacks theory call for an examination of the relationship between the quantity of law and stratification, morphology, cultu re, organization, and social control. Since his theory addresses all aspects of the law and is applicable across all social settings or groups (i.e., individuals, communities, citie s, or entire societies), attempts to examine the entire breadth of his theory prove daunting. Accordi ngly, studies have tended to address one aspect of the criminal justice process and have typically used the individual as the unit of analysis. For instance, these studies have examined decisions to invoke the law or report crime (Gottfredson & Hindelang, 1979; Braithwaite & Biles, 1980; Mo oney, 1986); prosecutorial decisions (Myers, 1980); variations in sentence outcomes (Moone y, 1986; Kruttschnitt, 1980-81; 1982); police decisions to arrest (Smith, 1987); and community perceptions of deviance or responses to deviance (Doyle & Luckenbill, 1991; Hembroff, 1987) A few have applied Blacks theory to general criminal justice activities at the macro or structural leve l. For example, studies have examined variations in police expenditures, arrest or incarceration rates, and homicide clearance rates across cities (L essan & Sheley, 1992; Borg & Park er, 2001). The following briefly summarizes the major studies of Blacks theory performed at the individual, structural, and multi-levels of analysis.
46 Individual Level Evaluations of Blacks Theory of Law In a study that exam ined five legal outcomes (e.g., seriousness of prosecutors charges, case dismissal, guilty plea, trail verdict, and sentence imposed), Myers (1980) tested Blacks theory to ascertain whether defenda nts social statuses significan tly affected the application of the law at various stages. Some of her findi ngs confirmed Blacks theoretical pr edictions. Specifically, she found that prosecutors charge s were more likely to be serious in cases involving a white victim and Black offender (rank); cases where the victim was older than the offender (rank); and offenses that included multipl e victims (organization). However, offenders sex, race, or age (vertical status or rank); le vel of intimacy, employment, or marital status (morphology or integrational status); education (culture or conven tionality); and prior criminal history (normative stat us or respectability) did not significantly impact legal outcomes. Also, measures of social control includi ng crime setting (public or private)2 and time did not have a substantial impact on case outcomes. In some cas es, Myers findings were contrary to Blacks expectations. For instance, she discovered prosecution charges were likely to be less serious when the victim was married (greater integrati onal status) or never b een arrested (greater respectability). Black proposes victims with greater integrational status (married) and respectability (never been arrested) should enj oy more law and thus crim es against them should be considered more serious. Overall, Myers conclude d that there was inconsistent and insubstantial support for most of Black s theoretical propositions. Mooney (1986) tested whether coll ege students social statuses affected either the initiation or application of the law within a private, quasi-legal setting (colle ge university). In particular, 2 Black (1976) notes that time and place are important measures of normative status or social control. He maintains that nonlegal social control is weak er at nighttime and in public places. Subsequently, law increases or becomes more important. People are more likely to invoke the law at nighttime when they are as leep and nonlegal social controls are nonexistent. Likewise, Black predicts that pe ople are more likely to call the police in public disputes since private conflicts tend to have more nonlegal social control.
47 she performed a partial an alysis of Blacks theory to determine how students rank, cultural, or organizational statuses affected the likelihood of seeking official universi ty decisions (e.g., file complaints, seek legal remedies, or contact police) or receiving a university sanction (e.g., warning, expulsion, fine/ticket, or legal repudiations). She found limited support for Blacks theory. In terms of the initiation of the law students with greater organizational status (e.g., employed, married, member of a campus or commun ity organization) were significantly more likely to use the law or initiate university le gal proceedings. While significant, students rank (e.g., sex, race, income, grade point average) was i nversely, rather than di rectly, related to the probability of initiating the law. In contrast to Blacks theoretical expectations, female students, nonwhites, students with low monthly incomes, a nd those with low grade point averages (lower rank) were more likely rather than less likely to invoke the law. As related to the application of the law stratification was dire ctly related to the likelihood of receiving university sanctions. This finding validates Blacks hypothesis that st udents with lower rank ar e subject to more law or more likely to receive a university sanction. Examining anticipated responses to various neighborhood problems (e.g., street crime, drugs, poor property care, landlord abuses, noise), Doyle and Luck enbill (1991) tested Blacks theory to determine which residents were more likely to contact any legal official or agency. Black proposes residents with greater rank, integr ational status, conventionality, organizational status, and respectability should be more likely to invoke the law and co ntact a legal official regarding neighborhood issues. Overall, Doyle and Luckenbill found a weak or insignificant relationship between the mobiliz ation of the law and stratification, morphology, culture, and organization. Residents sex, race, income (rank); employment, martial status, sociability (integrational status); education (cultural status); or organizatio nal status (numbe r of community
48 memberships) did not consistently predict the likelihood of contact ing officials. In terms of social control and law, Doyle and Luckenbill discovered a signifi cant, but positive, relationship between nonlegal or informal soci al control and the law. Whereas Blacks theory predicts an inverse relationship between nonleg al social control and the law, they found that those most likely to use alternative nonlegal so cial control (i.e., directly talk to person involved in dispute) were also more likely to mobilize the law and contact an official. In a study that examined sentence outcomes among female offenders, Kruttschnitt (198081) found consistent support for Blacks theory that the law varies by a defendants rank, integrational status, conventionalit y, and respectability. In particul ar, she examined the extent to which a womans race and income (vertical status or rank); employment status (integrational status); age (culture); and prior criminal hi story (respectability) impacted sentence severity across five different types of offenses (e.g., disturbing the peace, assault, forgery, drug violation, and petty theft). Kruttschnitt discovered that female offenders with lower economic status, integrational status, and respectability received the most severe sentences. In a similar study, Kruttschnitt (1982) ag ain found support for Blacks theory that womens so cial statuses significantly impact sentence severity. In an assessment of police responses to violence encounters among citizens, Smith (1987) conducted a partial examination of Blacks theory to determine if arrest decisions varied by offender or victim social statuses. Specificall y, Smith measured the e ffect of race and sex (vertical status or rank); offe nder and victim intimacy (morphology) ; and social control (dispute in private or public setting) on police respons es. He found supportive ev idence that victims rank influenced how police handled violent crimes. As predicted, the police were less likely to arrest in violent cases involving Black or female victims (lower rank). Smith also included a
49 structural measure of stratifica tion as he used household income rates to measure the economic status (stratification or vertical status) of neighborhoods. In support of Black, Smith discovered that the police were significantly more likely to arrest in low income neighborhoods than in upper income neighborhoods. He concluded that the economic distance between police officers and residents were greater in poor communities a nd thus the police were more likely to use the law against residents with lower rank than themselves. Hembroff (1987) tested Blacks theory agai nst the perceived seri ousness of two crimes (i.e., stabbing and theft). He created and manipulated 32 different crime scenarios to determine if respondents perceptions of se riousness varied according to the so cial statuses of offenders and victims involved in a conflict. For instance, respondents were aske d to judge the serious of an accountant killing a family-man banker in self-defense or a bartender killing a family-main bank executive in self-defense. In general, the re sults corroborated most of Blacks propositions. There was clear support that conventionality (cul ture) affected percepti ons of seriousness. Crimes by conventional offenders were judged as less serious than the same acts committed by less conventional offenders. Integrational stat us (morphology) also si gnificantly affected perceptions as offenses committed against margin alized victims were considered less serious than similar acts committed against socially in tegrated victims. Supportive evidence, although not statistically significant, showed that victim respectability (normative status) influenced perceptions. Acts committed ag ainst respectable victims were generally judged more serious than acts against disreputable victims. Hembroff found no relationship between organization and perceived seriousness. Unexpected ly, he discovered st ratification or vertical status results contradicted Blacks theory. Black anticipates that upward deviance (crime by low ranked offender against a high ranked victim) should be deemed the most serious of offenses, while
50 downward deviance (crime by high ranked offender against low ranked victim) should be judged the least serious of offenses. In other words, a poor offender who commits a crime against a rich victim is considered more serious than a rich offender committing a crime against a poor victim. Hembroffs results showed that respondents judg ed cases involving people within the same rank (poor offender, poor victim; rich offender, rich vi ctim) as less serious th an scenarios depicting upward deviance (poor offende r, rich victim). Structural-Level Evaluations of Blacks Theory of Law In the first aggregated exam ination of Blacks theory, Lessan and Sheley (1992) focused entirely on structural statuses and tested whethe r changes (increases or decreases) in (1) police expenditures, (2) yearly arrest rates, (3) incarc eration rates, and (4) co mmunity correction rates corresponded with changes in structural stratification, morphology, culture, organization, and social control. Specifically, they sought to ascer tain if labor versus co rporate income ratios (stratification or vert ical status); division of labor (morphology); relig ion, technology or number of patents within a location (cul ture); years at war, number of corporations per 100,000 people (organization); and homicide rates (social control) co rrelated with changes in the quantity of law. They discovered that only year of war (organization) was consistently and significantly related to changes in law enforcement and correctional ac tivity. Legal activities (police expenditures; arrest, incarceration, and commun ity correction rates) tended to increase during wartime and subside during peacetime. The relationship between stratification, culture, or social control and the law was highly inconsistent and they found no significant link betwee n division of labor (morphology) and law. In several instances, the relationship between stru ctural variables and law activity was in a direction opposite of that hypothesized by Black. In perhaps the strongest affirm ation of Blacks theory of law, Borg and Parkers (2001) macrolevel analysis examined the impact of ci ty characteristics on homicide clearance rates.
51 Their study included twelve aggregated urban community measures related to stratification, morphology, culture, organization, and social control. On the whole, their results substantiated that city characteristics significantly influenced the likelihood of clearing homicide cases. Specifically, cities marked by greater racial in equalities (stratification); higher levels of education (culture); and lower hom icide rates (social control) had the highest homicide clearance rates. While less support was found for morphology and organization, at least two community measures impacted homicide clearance rates. Cities with greater residential stability (integration) had more law or higher clearance rates, as did urban areas that spent more expenditure for educa tion (organization). Multilevel Evaluations of Blacks Theory of Law One of the earliest tes ts of Blacks theory focu sed on the initiation of law or variations in the likelihood of reporting crime. Gottfreds on and Hindelang (1979) used United States victimization data to conduct a multilevel analysis and determine which individuals and neighborhoods were most likely to invoke the law. In particular, they measured characteristics of people who more or less likel y to report crimes and characte ristics of the neighborhoods that were more or less likely to report crimes to th e police. They found mixed empirical support that a victims (1) family income level (rank); (2 ) employment or marital status (morphology or integrational status); (3) educati onal level (culture or conventiona lity); (4) organizational status (individual versus business) impacted the probab ility of reporting crimes to the police. As predicted by Black, married victims (morphology) were significantly more likely to report crime than single people; victims with more education (culture) were slightly more likely to report crime; and businesses (organization) were more lik ely than individuals to report crime. At the structural level, none of the neighborhood ch aracteristics including community poverty or household incomes (stratification or vertical status); community size, population density
52 (morphology); proportion of community with college degrees (cultu re); and (4) rural, suburban, or urban community (social control) had a significant influe nce on the likeli hood of reporting crime. Similar to previous studies, some of Gottf redson and Hindelangs (1979) findings were in a direction opposite of that anticipated by Blacks theoretical propositi ons. For example, Black (1976) predicts an inverse relationship between legal and nonlegal social control. Crimes occurring at home (greater nonlegal or informal social control) should be reported less often than crimes in public places (street) where nonlegal controls are nonexistent. Gottfredson and Hindelang (1979) discovered that that reporti ng was actually higher for victimization occurring within the home and lowest for victimization that occurred in public places. Among the structural variables, th ey found that the reporting of crime police actually decreased as the size of the city increased whereas Blacks theory expect s the quantity of law should increase as the size of community (morphology) increases. Gottfredson and Hindelang (1979) tested an alternative theory and concluded victim perceptions of seriousness (e.g., whether crime resulted in significant bodily harm or financ ial loss) was more important in determining victims decisions to report crime. Braithwaite and Biles (1980) replicated Go ttfredson and Hindelangs study and conducted a multilevel analysis of Blacks theory using Australian victimization data. As predicted, they found that intimates (morphology) were less likely to report crime to the police, while those employed (morphology) and victims with higher educational levels (culture) were more likely to report crimes. Also, reporting of crimes increased at night when nonlegal social control mechanisms were absent (social control). Si milar to Gottfredson and Hindelang (1979) however, none of their structural significantly affect ed the likelihood of reporting crime.
53 Overall, there has been inconsistent or te ntative confirmation of Blacks theory. Most assessments have found only moderate to weak s upport and in several instances findings have refuted or contradicted Blacks expectations. Consequently, so me researchers have concluded that there are serious questions concerning the validity, predictive power, and generality of his theory of law (Myers, 1980; Greenberg, 1983; Mooney, 1986; Lessan & Sheley, 1992). However, it should be noted these previous st udies conducted only a par tial test of Blacks theory. His theory is based on the idea that the social structure of a cas e should be the unit of analysis. As stated earlier, the social structure of the case includes statuses of the actors involved in a case (offender, victim, and third parties) the relational ties among these actors, and characteristics of the setting or location in which conflicts arise. Thus, a true test of Blacks theory would incorporate measures of all these so cial status factors. Accordingly, the lack of empirical support may be related to the absence of analysis based on the entire social structure of the case. Most empirical examinations have focused only on the microsociological dimensions of Blacks theory since they analyze characteri stics of persons invoking the law or against whom the law is applied (Lessan & Sheley, 1992; Borg & Parker, 2001). Other studies have focused exclusively on macrosociological or structural dimensions of Black s theory and investigated the effect of neighborhood or city characteristics on the behavior of law (Gottfredson & Hindelang, 1979; Braithwaite & Biles, 1980; Smith, 1987; Le ssan & Sheley, 1992; Borg & Parker, 2001). In this way, tests of Blacks theory have performe d a single-level analysis since they concentrate either on individual or macrolevel (city) statuses. In part, this c oncentration on the single unit of analysis is due to the lack of exhaustive data sources available (Borg & Parker, 2001). While this dissertation encounters similar data limita tions, it begins to address past limitations by including more aspects of the cas e. In particular, this research simultaneously examines the
54 affect of corporate offender characteristics and structural characteristics of the federal districts on sentence outcomes. Also, this dissertation diverg es from previous analys es of Blacks theory because it examines corporate rather than individu al offenders. Since his theory is applicable across all units of analysis it is possible to determ ine the effect of corporate social statuses on the behavior of law. The author is unaware of any examinations of Blacks theory using corporations as its unit of analysis. It should also be noted that despite the l ack of empirical confir mation, Blacks theory appears well suited for examining sentence outcomes under the federal sentencing guidelines. A measure of at least one of his social statuses is explicitly part of the guidelines. Blacks normative status or social control dimension predicts that law is greater in a direction toward less respectability. All else being equal, those with a criminal pa st are generally considered less reputable and therefore should be subject to more law or greater punishments. Under the guidelines, criminal history is a primary criteri on in the sentence decision process. It is an aggravating factor that justifies increases in sent ence severity. In this respect, sentence outcomes are directly linked to an offenders criminal pa st or, in Blacks term inology, their level of respectability. Beyond respectability (social control), Black identifies othe r, extralegal variables which also impact sentence outcomes. In particular, he asserts variations in legal outcomes can be explained by the rank (stratifica tion or vertical status), integr ational (morphology), cultural, and organizational statuses of the actors and locatio ns involved in a case. Indeed, post-guideline research continues to find that extralegal fact ors such as race, socio economic status and gender (rank) and education (integrational status) produce sentence disparit ies (Albonetti 1997; Steffensmeier et al., 1998; Steffensmeier, & Demuth, 2000; Ulmer, 2005; Ulmer & Johnson,
55 2004; Kautt, 2002; Kautt & Spohn 2002; Lambiras, 2003; Hofer et al, 199 9). Organizational status also appears to influen ce the behavior of the law. Black predicts law is greater in a direction toward less organizati on so that individuals are more vulnerable to the law than corporations. Since the implementation of the guidelines, individual offender cases are significantly more likely to appe ar in federal courts than co rporate defendant cases. For example, between 1996 and 2000 there were approximately 257,441 federal cases involving individual offenders compared to only 1,149 corporate crime cas es (Murphy, 2002). Beyond individual offenders, the statuses of corporations impact sentence outcomes. In particular, financial status (rank) and size (organization) significantly a ffect fine amounts as corporate offenders experiencing financial stress and smaller organizations receive the highest fines (Beck & OBrien, 2000; Piquero & Davis, 2004). Such findings undermine the very foundation of guidelines which were explicitly developed to eliminate the influence of nonlegal offender characteristics on sentence outcomes. Through th e use of formalized sentencing instructions, federal district courts are expected to perform like rational mach ines impervious to extralegal influences. However, Black asserts extralegal social factors will invariably impact the application of law and thus senten ce disparities are inevitable. Beyond the influence of characteristics of offenders, prior research al so suggests that the federal courts are inherently influenced by their immediate political, economic and cultural environments (Ulmer 2005; Stidham & Carp, 198 8; Benson & Walker, 1988; Kautt, 2002). Rather than being viewed as one rigid unified court system, federal district courts more accurately represent unique court communities th at supersede the formal rationality of the sentencing guidelines (Ulmer, 2005). Subsequentl y, Blacks focus on the social structure of the case is directly relevant here because his theory recognizes that ch aracteristics of the offender (in
56 this case corporations) as well as characteristics of the location in whic h law behaves (in this case federal districts) both impact legal outco mes. While not entirely exhaustive, this dissertation addresses prior shortcoming by ex amining as many dimensions of the social structure as possible including the affect of offender and locational st atuses on fine amounts. Chapter 4 reviews literature that indicates a link between Blacks theo ry and variations in sentence outcomes under the or ganizational guidelines.
57 CHAPTER 4 BLACKS THEORY OF LAW AND CORPORATE CRIME SENTE NCING This chapter discusses Blacks theory in re lation to corporate criminal sentencing. In particular, it examines the empirical relationshi p between offender and locational characteristics (as applied to stratification, mo rphology, culture, organization, and so cial control) and sentence outcomes. It should be noted however that there is a relative paucity of corporate crime research and data sources (Lynch, McGurrin, & Fenwick, 2004; Sutherland, 1949; Snider, 1993, 2000; Parker, 1989). Since the 1970s, just a few empi rical studies on corporate crime have been funded by major agencies in the United States (Snider, 2003). Consequently, there are only a handful of analyses on corporate crime senten cing. Concerning the orga nizational guidelines, studies have generally examined whether fi ne amounts substantially increased under the guidelines and the effect of organizational characteristics on sent ence outcomes (Cohen, 1989, 1991, 1996; Parker & Atkins, 1999; Alexander, Arlen, Cohen, 1999; Mu rphy, 2002; Piquero & Davis, 2004). To the authors knowledge, no study has directly examined the effect of structural conditions on sentencing under th e organizational guidelines. Th erefore, attempting to locate literature germane to Blacks concepts and corp orate sentencing is cha llenging. Black himself dedicates little time illustrating how his concepts relate to corporations with the exception of his discussions on organi zational status. Nonetheless, enough research exists to i ndicate a relationship be tween corporate and locational characteristics and the application of law. At the corporate offender level, the literature consistently finds an inverse re lationship between sentencing and corporate economic status and size (Goff, 1993). The most punitive sanc tions are meted out against economically marginalized and small corporations while the most lenient penalties are levied against economically resourceful and large corporati ons (Sutherland, 1949, Clinard & Yeager, 1980;
58 Shover, Clleland, & Lynxwiler, 1986; Yeager, 1991; Goff, 1993). At th e structural level, critical criminologists have traditionally as serted that corporati ons play an important role in terms of shaping the political-eco nomic structure. Through campaign contributions and lobbying efforts, corporations influence the draf ting of lenient corporate crime legislation and help establish probusiness policies or agendas (Sutherland, 1949; Quinney, 1974; 1980; Chambliss & Seidman, 1982; Clinard & Yeager, 1980; Goodman, 1979; Snider, 2000; Michalowski & Carlson, 2000; Pearce & Tombs, 2002; Kramer, et al, 2002; Mo khiber, 1988). Subsequently, nonenforcement or moderate sanctions have been the prefe rred method of punishing corporate criminality (Snider, 1993). Stratification and Corporate Crime Sentencing Corporate (Offender) Rank a nd Sentence Ou tcomes Black (1976) predicts that law varies directly with rank. A ll else being equal, offenders with less rank enjoy few legal advantages. Acco rdingly, corporate offende rs with less economic rank should be sentenced more harshly than or ganizations with greater rank. Corporate crime research has generally supported this hypothesis. In his groundbreaking book, White Collar Crime Edwin Sutherland (1949) examined advers e decisions against the seventy largest manufacturing corporations and formal legal res ponses to these violations. One of his major findings was that there was a differential implem entation of the law. Corporate offenders were sanctioned leniently partly because of their hi gh economic status. Criminal justice officials including legislators, prosecutors, regulators, and judges were afraid to impose aggressive sanctions against corporate managers because of a fear of losing potential campaign funds or support. In addition, he maintained that crimin al justice officials generally viewed corporate offenders as respectable businessmen who made technical violations th at did not warrant tough penalties. Instead, corporations were more likely to be subjected to alternatives to criminal law
59 such as civil or regulatory penalties that Sutherland believed were less punitive and stigmatizing. Indeed, a persistent theme of corporate crime is the lack of criminal charges or sanctions imposed against corporate offenders (Gof f, 1993). According to Snider (1990, 1993), cooperative, informal sanctions (warnings, fines, cease and desist orders) are generally preferred over confrontational criminalization penalties (s ubstantial fines, imprisonment, etc). Clinard and Yeager (1980) conducted a study si milar to Sutherlands and analyzed law violations and sentence outcomes among the 582 largest corporations in th e United States during 1975 and 1976. They concluded that the largest co rporations (annual sales $1 billion or more) received the most lenient penalties despite being th e most habitual or serious offenders. That is, corporate criminals with a significant history of offending and those that committed the most harmful crimes typically received the most in consequential punishments. Large corporations committed an average of 5.1 violations and accounted for over 70 percent of the serious violations and 60 percent of the moderately seri ous offenses. However, the average fine for large corporations was $1,000 compared to $1690 for medium sized corpor ations (annual sales of $500-$999 million) and $750 for small firms (a nnual sales of $300-$499 million). In addition, Clinard and Yeager found that 86 percent of all fines imposed against large businesses were $5,000 or less and just 6.5 percent were fined over $45,000. By comparis on, just 54 percent of all fines against smaller firms were $5,000 or less and 38.5 received fi nes $45,000 or greater. Studies of the federal sentencing of cor porate criminals have found similar results. Cohen (1989) studied pre-organiza tional guideline sentence outcome s and discovered that only a small percentage (approximately te n percent) of corporations sent enced in federal court had sales over $1 million (Cohen, 1989). A subsequent stu dy analyzing outcomes under the organizational guidelines found that just about 15 percent of the cor porations sentenced had sales exceeding $1
60 million (Cohen, 1991). Yet another study of 961 corporations sentenced in federal courts from 1984 to 1990 determined that only six percent of those convicted had sales in excess of $50 million. Conversely, corporations in financial stress or lower rank generally received tougher sanctions. A study found that org anizations deemed financially unable to pay the minimum fine amount were more likely to receive a higher fi ne amount, (Piquero & Davis, 2004: 651). Overall, these studies tend to confirm Blacks pred iction that the law is more likely to be applied downward against offenders with lower economic rank. Federal District Rank and Sentence Outcomes At the structural level, B lack (1976) maintains that the law va ries directly with locational rank. Federal districts with greater rank or ec onomic status will use more law or impose higher fines. Conversely, districts w ith less rank will impose less la w. While no study has directly tested this prediction, some research has ar gued that contextual economic conditions impact corporate sentencing. For instance, corporations tend to actively seek poor or economically depressed locations primarily because they offer few regulatory restrictions and lenient sanctions (Goodman, 1979; Snider, 2000; Michalowski & Ca rlson, 2000; Pearce & Tombs, 2002). Economically depressed areas are reluctant to im pose harsh sanctions against corporate offenders because of the potential economic conse quences (Benson & Cullen, 1998, Aulette & Michalowski, 1993). Communities with weak econom ies rely heavily on corporate investments for essential jobs and tax re venues. Imposition of tough enfo rcement may cause corporate relocation and disinvestment (Benson & Cullen, 1998). Similarly, punitive sanctions may hinder the ability to attract new corporations or other industries to the area (Aulette & Michalowski, 1993). In this regard, poor area s are generally more willing to tolerant corporat e crime (Benson & Cullen, 1998; Aulette & Michalowski, 1993). Subsequently, poor ar eas are considered probusiness climates because they provide a hos t of benefits or enticements including
61 deregulated oversight agencies; no unions; low labor wages and benefits; nonexistent or relaxed safety obligations; corporat e tax breaks and subsidies; tort refo rms that limit damages or lawsuits filed against corporations; the ability to downsize without penalty; and few consequences for spatialization or capital flight (Kotz, 2003; Snider, 2000; Michal owski & Carlson, 2000; Calavita & Pontell, 1990; Pearce & Tombs, 2002; G oodman, 1979; Eisinger, 1988 ; Gordon et al, 1982; Grant II, & Martinez Jr., 1997; Grant II, & Wa llace, 1994; Kasarda, & Irwin, 1991). Overall, economically poor markets or locations have tended to minimize business expenditures and labor costs while at the same time providing various form s of corporate protectionism against criminal legislation, prosecution, and sentencing (Snider, 2000; Pear ce & Tombs, 2002; Simon, 2000; Michalowski & Carlson, 2000; Gordon et al, 1982) These patterns are all consistent with Blacks theoretical propositions that locations wi th lower rank tend to impose less law and tend to allow corporations to freely engage in que stionable, risky, and unlawful behaviors with impunity. Morphology (Integrational Status) and Corporate Crime Sentencing Corporate (Offender) Integrational St atus and Sentence Outcomes Black hypothesizes that law varies inversely with integrational status. Integrated offenders enjoy legal immunity, while marginalized offenders are subject to more law throughout the criminal justice process. Marginalized offenders are more likely to be arrested, prosecuted, or severely punished. Thus, marginalized corporat e offenders should be subject to harsher fines compared to more integrated defendants. One pot ential measure of corporat e integrational status may be their ownership structure. For instance, publicly owned businesses could be considered more socially integrated than closely-held co rporations. Publicly owned firms have individual stockholders from across the country and they ar e connected to the major financial institutions since their ownership shares ar e traded on open stock exchanges. By contrast, closely-held
62 corporations are controlled by a single individual or small gro up of owners and do not have nationwide investors. Thus, Blacks theory w ould predict that publicly traded corporations should be subject to less law sin ce they have greater integrati onal status. In support, Cohen (1996) discovered that only 8% of corporations sentenced in federal courts from 1984 to 1990 were publicly traded firms. In 1988, approxima tely 5% of corporate offenders sentenced in federal courts were publicly owned businesses (Cohen, 1991). On the other hand, closely-held organizations have been the primary corporations sanctioned at the federa l level. Of nearly 1,000 corporations convicted and sentenced in federal court from 1984 to 1990, approximately 70 percent of the firms were closely-held (Cohen, 1996). A study of organizations convicted under federal guidelines from 1991 to 2001 concluded that closely-held corporations received the highest fines (Piquero & Davis, 2004). Federal District (Integrational Status) and Sentence Outcomes According to Black (1976), law varies directly with structural integration. As m ore and more people participate in the major social ins titutions within a geographical area, law increases there as well. Conversely, in social settings where people are undiffere ntiated by functions or have little or no social exchanges there is less law. In this resp ect, federal districts with greater integrational status should use more law or impose higher fines compar ed to marginalized districts. Previous studies have examined city characteristics related to joblessness, unemployment, single-paren t households, and residential mobility to measure integrational status at the structural level. For instance, Borg and Parker (2001) found that communities with fewer married households, higher unemployment rates, a nd higher rates of short-term residency also tended to have less law or lower clearance rates. Applied to this research, Blacks propositions would suggest that federal districts located in more socially marginalized areas (i.e., higher unemployment, greater residential mobility, higher single family households or divorce rates,
63 etc) will also use less law or impose lower fi nes. Current literat ure on corporate crime sanctioning lacks information, anecd otal or otherwise, to eval uate this argument. This dissertation will include measures to evaluate the affect of integrational factors on sentence outcomes among corporate offenders. Culture (Conventionality) and Corporate Crime Sentencing Corporate (Offender) Convention ality and Sentence Outcomes The directio n of law, according to Black ( 1976) is generally applied downward against unconventional offenders, or those with a more ma rginalized culture. A ll things being equal, crimes committed by less conventional offenders are generally considered more serious and less conventional criminals are more likely to be ar rested, convicted, and severely punished. As related to this disserta tion, then less conventiona l organizational defendants should be subject to more law or higher fines. Black (1976) and most researchers us e education as an indication of cultural status or conventiona lity (Gottfredson and Hindelang, 1979; Braithwaite and Biles 1980; Myers, 1980; Mooney, 1986; Hembroff, 1987; Doyl e & Luckenbill, 1991; Borg & Parker, 2001). Educated offenders are more conventional and t hus encounter less law, while less educated defendants would be subject to gr eater law or tougher sanctions. Wh ile it is possible to measure the educational level of a corpor ation by measuring the aggregated educational credentials of its employees, the author is unaware of any st udy or sources that include such data. Federal District (Cultural Distance) and Sentence Outcomes While m easuring corporate cultural status is challenging, assessing the cu ltural status of a particular location is more practical Black (1976) states locations or social se ttings have distinct cultures including prevailing ethnic, educational, religious, philosophi cal, or political norms. He predicts the quantity of law increa ses with cultural diversity or distance: the greater the cultural distance between an offender and the larger soci al context, the more law applied against the
64 defendant. Thus, fine amounts s hould be higher in federal distri cts in which there is greater distance between a corporate offender and the larg er cultural climate. In districts with less cultural diversity or distance, fines should be lowe r. Most relevant for this dissertation is the political climate within federal districts. Inde ed, one of the most sali ent structural factors believed to impact corporate crime sentencing is political climate. For the most part, corporations have benefitted from Republican political climates th at endorse probusiness economic policies. Beginning with the Ronald Reagan pr esidency of the earl y 1980s, Republicans have generally championed neoliberal economic agendas. In short, neoliberalism supports selfregulation and free market activity with minimal government interven tion. Strict or aggressive governmental policies are considered economically detrimental because they hinder corporate ability to effectively compet e and maximize profit opportunities (Snider, 2000; Michalowski & Carlson, 2000; Calavita & Pont ell, 1990; Pearce & Tombs, 2002; Kotz, 2003). Corporate expenditures would be wasted on meeting tough regulatory standards and/or paying workers higher wages or more benefits. To advance its probusiness economic policies, the Reagan administration deregulated several important in dustries or agencies including the savings and loans industry, the Environmental Protection Agen cy, and the Occupational Safety and Health Administration (OSHA) (Calavita & Pontell, 1990; Calavita, Tillman, & Pontell, 1997; Calavita, 1983, Snider, 1993). In addition, Reagan frequently appointed former corporate lobbyists to top positions in regulatory agencies (Snider, 1993). Finally, the Reagan administration packed the federal courts with young conservative judges generally sympathetic to corporate interests (Schwartz, 1988; Goldman, 1997; Lyles, 1997; Scherer, 2005). Not surpri singly, soon after the Reagan administration took office several important cases agains t powerful corporations were
65 either dropped or significantly stalled (Calavita, 1983). Co llectively, some scholars have denounced Republican supported neoliberal econo mic policies because they offer corporate protectionism and criminal immunity (S nider, 1990, 2000; Chambliss & Seidman, 1982; Quinney, 1974, 1980; Pearce & Tombs, 2002; Calavi ta & Pontell, 1990; Calavita, Tillman, & Pontell, 1997; Sutherland, 1940; Friedrichs, 2004). The pioneer of white collar crime research, Edwin Sutherland (1949), asserted criminal justice and political leaders were primarily resp onsible for the preferen tial sentences afforded corporate offenders. He concl uded these officials shared a cu ltural homogeneity with corporate leaders because they had mutual social networ ks, attended the same universities, and shared ideologies that made them intimately familiar with each another. As such, prosecutors, judges, and politicians admired and respected businessmen and had difficulty conceiving of corporate offenders as real criminals. Sutherland noted this mutual respect was responsible for the differential or lenient treatmen t of corporate criminals. Du ring probusiness presidential administrations, corporations seldom encountered administrative regulations or sanctions. Laws originally drafted to criminalize corporate transg ressions were instead used against less powerful labor unions. For instance, Sutherland (1949) discovered that from 1890 to 1929, only 27 percent of criminal prosecutions under the Sherma n Act (a law originally drafted to criminalize corporate monopolistic practices) in volved corporate offenders compared to 71 percent of cases against trade unions. Since the 1930s, other studies have found a similarly strong association between political ties and corporate crime senten cing. Specifically, federal j udges appointed by Republican presidents have generally imposed lenient se ntences against corporate defendants (Carp & Rowland, 1983; Giles et al, 2001; Songer et al, 2003; Rowla nd & Todd, 1991; Stidham & Carp,
66 1988; Carp et al 2001). For in stance, Republican appointed judge s have been more supportive of corporate interests in cases pertaining to labor/ union-management conflicts and cases that have attempted to enforce government or environmen tal regulation over corporate practices (Carp & Rowland, 1983; Carp et al, 2001; Songer et al, 2003; Rowla nd & Todd, 1991). In corporate lawsuit cases, a study found that Reagan appoi nted judges were th e least supportive of underdog litigants (unions, employees, minorit y groups, aliens, and criminals) that sued uppperdog interests such as corp orations (Rowland & Todd, 1991). In environmental regulation cases, Republican ap pointed judges have been more inclined to issue pro-business, anti-environment decisi ons (Canes-Wrone, 2003; Austin et al, 2004; Sunstein et al, 2004; Center for American Progress, 2005). For instan ce, a study of National Environmental Policy Act (NEPA) di scovered Republican appointees issued pro-business decisions in nearly 58 percent of the cases as opposed to Democrat appointed judges who supported business interests just 14 percent of the time. Republican appointed judges made proenvironment decisions just over 28 percent of the tim e, whereas Democratic appointees did so nearly 60 percent of the time (A ustin et al, 2004). Thus, Republi can judges were more than four times more likely than Democratic appointee s to support corporate litigants in these environmental cases. Beyond individual voting patterns, other studies have asserted the political climate of the federal courts can affect other measures of law as conceptualized by Black. A study found the political composition of the fede ral courts affected the probab ility of the Army Corps of Engineers issuing business grants or permits for wetland development (Canes-Wrone, 2003). The Corps officials were more likely to grant permits in conservative Republican districts (districts in which the majority of judges were appointed by Republi can presidents) with
67 probusiness climates and fewer threats of environmental lawsuits. Conversely, the Corps was less likely to issue permits in Democratic di stricts that were more inclined to support environmental groups and challenge wetland developmen t. This suggests that districts are likely to develop partisan sentencing norms that ultimat ely influence how cases are adjudicated within particular social set tings or locations. There is also evidence that j udges are influenced by the politi cal voting behaviors of their immediate colleagues. In particular, studies of th e federal circuit courts assert judges are likely to be affected by the political party affiliation of other panel judges (Sunstein et al 2004; Revesz, 1997; Austin et al, 2004). Thes e panel effects can either amplify voting behaviors particularly when a judge sits on a panel with one or more judges appointed by the same political party or dampen decisions if sitting with one or more j udges appointed by a President from the opposing party (Sunstein et al, 2004; Reve sz, 1997). For instance, a three judge circuit panel that includes all Republican appointees would expectedly amplify voting behaviors and yield highly conservative decisions, while a pane l with all Democratic appointed judges would more inclined to issue liberal decisions. However, a lone Re publican appointee on a pa nel with two Democrats would likely dampen voting decisions and force the Republican to acquiesce and vote more liberally. Similarly, a single Democrat sitting on a panel with two Republicans would be more inclined to decide in a more conservative ma nner. As such, the political and ideological inclinations of a judges immedi ate colleagues are perhaps better predictors of voting behavior than a judges own personal ideology (Revesz, 1997; Austin et al, 2004). An analysis of Environmental Protection Ag ency (EPA) decisions issued in the D.C. Circuit from 1970 to 1994 confirmed that judicial voting behaviors were greatly affected by the
68 party affiliation of other judge s on the panel (Revesz, 1997). J udges were more apt to vote along with their ideological preferences when sitting on panels with at least one other judge of the same political party. A similar study examined nearly 5,000 published panel decisions covering thirteen different categories of cases ranging from abortion and civil liber ty cases to corporate lawsuits and environmental litigation (Sunstein et al, 2004). Overall, the partisan composition significantly affected voting patterns. For inst ance, panels composed of three Democratic appointed judges issued liberal decisions 61 percent of time, while circuit panels with all Republican appointees did so 34 percent of the tim e. Panels with two Democratic appointees and one Republican issued liberal decisions 50 percent of the time. They also found dampening effects as panels with one Democrat appoint ee and two Republicans voted liberally only 39 percent of the time. Of interest to this dissert ation, these studies have found that all Republican panels were the most likely to issue probusines s decisions including case s involving lawsuits filed against corporations; campaign finance laws; and environmental regulation. Beyond immediate panel effects, Sunstein et al ( 2004) also found evidence of larger contextual influences on voting behaviors. Specifically, fe deral circuits with a majority of Democratic appointed judges were more inclined to issue liberal decisions. Thus, the political contextual climate can potentially explain di fferential sentence outcomes. Overall, the existence of panel effects and id eological amplification or dampening can be attributed to collegial concu rrence (Sunstein et al, 2004). Judges are likely to succumb to conformity pressures because peer acceptance is c onsidered a valuable and powerful asset. Also, dissenting opinions may be considered burdensome, time-consuming, futile, and more importantly a potential source collegial te nsion (Revesz, 1997). Ultimately, collegial concurrence contends that judge s confronted with unanimous co lleague opinions will generally
69 show deference and adopt the prevailing attitudes. In situations where th e panel is unified along partisan lines (all Republican or all Democratic panels), group polar ization is likely to occur in which like-minded groups tend to adopt more extr eme or intense behavior tendencies (Sunstein et al, 2004). Thus, all Republican panels tend to be the most conservative, while all Democratic panels the most liberal. Building upon these studies of panel effects a nd collegial concurren ce, this dissertation hypothesizes that district court judges may be infl uenced by district effects or the prevailing political climate and norms. In districts in which a majority of the judges are appointed by Republicans Presidents, the judicial normative clim ate may be decidedly probusiness. In this way, Republican districts can either amplify the voting behaviors of othe r Republican appointed judges or dampen the sentencing decisions of Demo cratic appointees. While district court judges do not encounter intimate panels li ke circuit judges, they nonetheless work in relatively small workgroups. To be sure, most dist ricts have an average of 7 sitt ing judges with a median of 5 judges per district. This reas onably small number of judges pe r district means a judge is generally cognizant of the political and ideological preference s of his or her fellow colleagues. For example, (Cohen, 1991) found that district ju dges came under considerable peer pressure to issue opinions opposing the constitutionality of the US Sentencing Commission and the Federal Sentencing Guidelines. As a dist rict court judge in Kansas st ated, there was considerable institutional and peer pressure to overturn the guidelines, (C ohen, 1991:187). This suggests district courts judges encounter similar politi cal pressures that may ultimately influence sentencing decisions. Overall, the literature suggests a shared or mutual political relationship between Republicans and corporations. Since Republicans tend to endorse probusiness economic agendas
70 that consider criminal sanctions as imposi ng unfair advantages on corporations, there are legitimate concerns about Republicans commitm ent to effectively regulate and sanction corporate criminality (Snider, 2000; Michalowsk i & Carlson, 2000; Calavita & Pontell, 1990; Pearce & Tombs, 2002; Schwartz, 1988; Kotz, 2 003; Friedrichs, 2004; Gould, 2003). As related to Blacks theory then, corporate offenders in federal districts characterized as more Republican should encounter less la w or lower fines because ther e is greater cultural intimacy (versus distance) between these de fendants and the larger political c limate which is likely to be more probusiness. Stated differently, legal outc omes in Republican dominated federal districts should be more favorable toward corporate o ffenders (less law) since Republican economic policies tend to be probusiness. Organization and Corporate Crime Sentencing Corporate (Offender) Organizationa l Status and Sentence Outcomes According to Black, law varies directly with organizational status. The more organized an offender the more legal advantages they enj oy. Offenders with less organization are more susceptible to applications of the law. Traditionally, corporate crime research has concluded that the intensive investigations and harshest sanctions tend to be directed against the smallest or most peripheral corporations while larger companies more often enjoy criminal immunity (Snider, 1993; Goff, 1993). According to corpor ate crime researchers, smaller corporations are typically at a disadvantage because they do not possess the resources to effectively challenge criminal investigations or prosecution (Snider, 1 993; Goff, 1993). They also lack the structural complexity to easily conceal their crimes (Erm ann & Lundman, 2002). As such, it is easier for prosecutors and judges to procure convictions because they can mo re quickly investigate crimes by smaller corporations and pinpoint criminal liability or respons ibility.
71 A study of criminal prosecutions of corporat ions in federal court from 1984 to 1987 found that the majority of offenders punished were small businesses that employed fewer than 50 workers (Cohen, 1989). A ten year investigation of the organizati onal guidelines discovered that the median number of employees for guilty co rporations was 20 workers (Murphy, 2002). Thus, it appears that corporatio ns with a lower organizational status are subject to more law under the federal guidelines compared to those with greater organizational status. Federal District Organizational Status and Sentence Outcomes According to Black (1976), social settings or lo cations with greater organizational status use more law, while areas with lo wer organizational status use less law. With regard to locations such as cities or states, he measures organiza tional status by the capacity for collective action and the number of administrative agen cies within a specific location. This includes increases in the number of public or private establishments such as businesses, governmental agencies, schools, or public works agencies. Black asserts the numbe r of organizations within a setting tends to increase during wartime, economic expansion, foreign trade, or disasters because new legislation, powers, or rights are needed to cen tralize decision-making and guide behaviors. Accordingly, as locations develop more organiza tions per capita they use more law or become increasingly litigious and punitive. Similar to morphological status, the author is unaware of any corporate crime study that has analyzed the effect of organizations per capita on sentence outcomes. However, a couple tests of Blacks theory at the structural level have found evid ence that organizational status impacts the behavior of law. For example, Lessan and Sheley (1992) discovered year of war (organization) was significantly related to incr eases or decreases in law enforcement (police expenditures and arrest rates) and correctional activity (incarceration and community correction rates). These legal activities consistently increased (more law) during wartime (greater
72 organizational status) and decrease d (less law) during peacetime (l ower organizational status). Borg and Parker (2001) found that urban areas that spent more expenditure for education (organization) had higher homicide clearance rates. In respect to this research, Blacks theory predicts federal districts with more private or public establishments (taken as a measurement of organizational status) should use more law or impose higher fines. Social Control (Normative Status or Resp ecta bility) and Corporate Crime Sentencing Corporate (Offender) Respectability and Sentence Outcomes Black (1976 ) uses respectability as an indica tor of normative status. In particular, he asserts the law varies directly with respectability as those with lower levels of respect are subject to more law. For instance, Black predicts criminals are more likely to be severely sanctioned compared to noncriminals since criminals enjoy lower levels of respectability. In support of this prediction, Piquero and Davis (2004 ) found that organizations with a prior history of criminal misconduct were more likely to receive fines at the upper end of the recommended sentence range compared to offenders wit hout a prior criminal history. Federal District Respectability and Sentence Outcomes W ith regards to social locati on, Black (1976) proposes that ge ographical areas with greater social control are considered less respectable an d thus should encounter more law. Locations with higher crime rates, numerous felons and soci al deviants such as prostitutes or drug users typically have less respect compared to areas with lower crime rates and fewer offenders or social deviants. Accordingly, disreputable social settings use more law and impose harsher sanctions. Previous test s of Blacks theory have focused on homicide or crime rates to measure social control or levels of respectability (Lessan & Sheley, 1992; Borg & Parker, 2001). Again, the author is unaware of corporate crime research that has directly examin ed this relationship but studies of cities have generall y found that locations with highe r crime rates tend to be more
73 punitive (Borg & Parker, 2001; Weidner et al 2005; Smith, 2004; Chiricos & DeLone 1992; Jacobs & Helms, 2001, 1996). If Blacks proposition is correct, federal di stricts with greater homicide or crime rates (less respectability) s hould use more law or impose higher fines than districts with lower homicide or crime rates (greater respectability). In sum, literature on corporate crime holds qui te a bit of suggestive evidence to support many of Blacks theoretical proposi tions as applied to variations in legal punishments. Some of this evidence supports the notion that extralegal corporate offender characteristics impact legal outcomes, while other studies suggest structural c onditions or factors effect sentencing. At the same time, several of Blacks arguments have not been empirically a ddressed, directly or indirectly, by previous resear ch. Subsequently, the analysis that follows provides a more systematic examination of Blacks propositions The next chapter ou tlines the operational definitions and empirical hypotheses that guide this dissertation.
74 CHAPTER 5 ANALYZING SENTENCE OUTCOMES UNDER T HE ORGANIZATIONA L GUIDELINES: DATA AND MEHTODS This dissertation addresses two central ques tions pertaining to th e federal sentencing guidelines for organizations: (1) to what extent do corporate offender characteristics or social statuses impact sentencing and (2) to what extent do characteristics of the federal districts affect sentence outcomes? In other words, are potenti al sentence disparities under the organizational guidelines related to extralegal offender characteristics and interdistrict variations? Blacks theory of law is utilized since it provides a theoretical framework for analyzing the social structure of the entire case. In particular, his th eory anticipates the likelihood of more or less law given offender (in this case, cor porate) characteristics and struct ural conditions (here, federal district courts). While prior gui deline studies have concentrated almost exclusively on the effect of offender characteristics, more recent studies have asserted that structural variables are equally important in determining sentences (Richard son & Vines, 1970; Heydebrand & Seron, 1990; Flemming et al, 1992; Ulmer, 2005; Kautt, 2002; Alschuler, 2005; Lambiras, 2003; Hofer et al, 1999). Thus, research should examine how larger po litical and social contextual factors work in conjunction with case level variable s to influence legal processes (Ulmer, 2005). To this end, this dissertation provides a more systemic an alysis of the organizational guidelines. This dissertation examines sentence outcomes from 2001 to 2004 primarily because of its interest in political cl imate (measure of district cultural status). Including too many successive years may inaccurately measure political climate si nce the political landscape tends to shift or change every two to four years. Accordingly, it is imperative to capture a distinctive political atmosphere to ascertain its impact on sentence outcomes. Based on th e literature, the years included (2001 to 2004) are indi cative of Republican control ove r the presidency and other important branches of government including the fe deral judiciary. More importantly, these years
75 also witnessed the reemergence of distinctive probusiness pol itical policies (F riedrichs, 2004; Pearce & Tombs, 2002; Gould, 2003; Kotz, 2003). Also, it should be noted that only federal districts embedded within the United States political system are analyzed. The federal dist ricts covering US territories (Puerto Rica, Guam, Virgin Islands, and Northern Mariana Islands) are excluded primarily because states have additional privileges and author ities that territories do not a nd thus treating them alike may introduce nonrandom biases (Kautt, 2002). The Washington DC federal district is included however, because of its historical link to the US political system. Thus, a total of 90 federal districts will be examined. Overall, there were 813 cases or corporati ons sentenced under the FSGO from 2001 to 2004. However, six of the fe deral districts (West Virginia Northern; Georgia Southern; Mississippi Northern; Hawa ii; Oklahoma Eastern; and Wyoming) did not have a single case during this period and were thus eliminated from the analysis (see Appendix for frequency tabulation). Consequently, the analysis is based on information gathered from 84 federal district courts. Duri ng the years of the study, an aver age of 9.68 cases per district occurred, with 28 districts having 10 or more cases. Two Florida districts including Florida Southern (44 cases) and Florida Middle (42 case s) had the most number of cases during this period, while the California Central district had 41 cases. Two dist ricts had at least 30 or more cases (New York Southern and California Northern district) while seven di stricts had 20 or more cases. A total of 16 district courts had betw een 10 and 17 cases. At the other end of the spectrum, 17 districts had two or fewer cases from 2001 to 2004. Specifi cally, seven districts only had one case, while 10 districts had ju st two cases during the years of study. A limited number of cases per level II units (district courts) repr esent a common problem in multilevel modeling. According to Clarke and Wheaton (2007), data sparseness can produce
76 robust or inflated estimations of significance a nd thus render multilevel models inappropriate. Recent trends to address this problem have used cluster techniques in whic h units with few cases are matched with groups that have a greater number of cases. One of the more common techniques is the use of census tracts to match groups or areas together along socioeconomic or demographics characteristics. However, Clarke and Wheaton suggest that the elimination of singleton groups (units with only one case) or similarly small groups can introduce bias because these purged units may be systematically diffe rent than groups with a greater number of observations. More importantly, they argue th e use of cluster techni ques introduces withingroup heterogeneity which artificially combines observations that are largely disparate or incompatible but share enough group-levels values to be matched. In order to test the threshold at which data sparseness does not adversely a ffect the reliability or usefulness of multilevel modeling, Clarke and Wheaton (2007) conducted a series of simula ted models that included both singleton and cluster groups. While accurate esti mates could be attained with units or groups that had an average of three or more cases, they maintain units with an average of six or more cases generally produce unbiased estimates. At th e same time, cluster te chniques did appear to have meaningful, albeit, moderate effects on model performances. In the end, they state that in multilevel analyses the number of groups is gene rally more important than the number of cases per group. As such, Clarke and Wheaton (2007, 344 & 345) conclude res earchers are probably safer with more data sparseness than they thi nk (according to the simula tions) and researchers should aim to preserve the in tegrity of the between-group diff erences. Since there is an average of nearly 10 cases per district, this dissertation meets the acceptable threshold for number of cases per group and therefore dist ricts with two or fewer cases will remain unclustered.
77 Most of the information used for this disse rtation is obtained from organizational data collected by the US Sentencing Commission and disseminated by th e Inter-University Consortium for Political and Social Research (ICPSR). Each year, the Commission compiles annual information of organizational defenda nts sentenced under the Federal Sentencing Guidelines for Organizations (FSGO). Among othe r things, this data set includes information related to (1) sentences outcomes directed agains t corporate offenders in cluding fines, probation, or restitution, (2) legal or case leve l variables, and (3) extralegal corporate characteristics. In this way, the ICPSR dataset includes measures of the dependent variable (fine amount), the legal factors that should theore tically guide the sentence decision pr ocess including factors related to offense seriousness and organizational culpability, and the extralegal corporate social statuses indicative of Blacks theoretical concepts. A second dataset, FEDSTATS, is used because it contains relevant structural measures of the US federal district courts. This data source provides demographic information including economic trends, population characteristics, and crime statistics within each district. Thus, FEDSTATS statistics provide pertinent meas ures of structural stratification, morphology, organization, and normative status. An additio nal dataset, the Alliance of Justice (2007)3, was used to gather information on all active federa l judges and, more importantly, details about the political party (Presidents) responsible for appointing judges within each district. This 3 It should be noted that the Alliance contains information on all judges actively sitting on the federal courts. It does not specifically identify which judges were sitting on the bench from 2001 to 2004. Thus, the data used in this analysis contains information on district court judges appointed after 2004. However, a significant majority of judges confirmed after 2004 were appointe d by Presidents of the same political pa rty as those they replaced. That is, Republican appointees who retired, receive d promotions, or died were essentially replaced by judges also appointed by a Republican President. To be sure, over 80 percent of the judges appointed by George W. Bush were appointed by Republican presidents (Goldman et al, 2005). Thus, the data represents a close approximation of the partisan affiliation of judges sitting on the bench from 2000 to 2003.
78 information is used to calculate the percenta ge of judges within each district appointed by Republican presidents. Dependent Variable This research uses economic fine as a measure of law. Black proposes that any quantitative increment in punishment increases the presence of law. Thus, it becomes possible to ascertain which corporate criminals are subject to more law (higher fine amounts) or less law (lower fine amounts) and which locations use more or less law after contro lling for legal factors. Accordingly, the dependent variable measures total fine or dollar amount meted out against organizations criminally convicted under the fe deral sentencing guidelines. This measure was selected because fines are the most salient puni shment under the organiza tional guidelines. Due to the highly skewed nature of total fine amount, a log of the total fine was created to help normalize the distribution. In chapter 3 a brief discussion was given about different styles of law outlined by Blacks theory including penal, compensatory, therapeutic, and conciliatory forms of law. Penal styles of law include formal criminal punishments, compensatory law involves civil punishments including restitution, therapeutic styles of law seek therapy or intervention, and conciliatory styles of law are best characterized by media tion. Under the FSGO, organizational defendants are subject to both penal (fine or probation) as well as compensatory (restitution) styles of law. Thus, it is plausible that a cor poration may receive no fine (less law) but have a large restitution payment (more law) imposed against them. If this is the case, a focus just on fine amount may overlook the importance of restitution and thus undermine Blacks predictions concerning the behavior of law. Preliminary tests were conducted to discern the im pact of restitution on sentence outcomes. A frequency ta bulation revealed that the major ity of offenders (61 percent or 499 out of 813 cases) were not subj ect to a restitution payment ($0) Furthermore, a bivariate
79 correlation indicated no significant re lationship between the log of to tal fine and restitution (see Table 5-3). Finally, a dummy variable for the total fine amount was created to represent organizations that received no fi ne (232 cases received a fine of $0) versus corporations that received a fine (580 cases). Averag e restitution among organizations that did not receive a fine was $2,031,747, while the mean restitution paid by organizations that did receive a fine was $17,810,113. Collectively, these analyses suggest that corporations that did not receive a fine were not more or less likely to have a restitution payment imposed against them. As such, the log of total fine remains the only dependent variable used for this dissertation. Case Level (Legal) Independent Variables Federal sentencing guidelines for organizations were drafted with the intent of basing sentence decisions on legally rele vant factors such as seriousne ss of offense and the level of corporate culpability for criminal actions. Offenders who committed similar crimes and had similar levels of culpability should receive equivalent punishments irre spective of extralegal offender attributes or district characteristics. In this way, any sentence variations could theoretically be explained by le gal variables. As developed by the US Sentencing Commission, final offense score and culpability score are the two primary means of determining fine amounts. Offense Seriousness Under the Guidelines, offense seriousness is m easured by calculating a base fine which is the greatest of three factors including (a) the offense level that corresponds to the individual guidelines, (b) the pecuniary gains from the offense, or (c) the pecuniary losses caused by the crime (USSC, 2007). The higher offense score, the higher the fine imposed. Unfortunately, there are a limited number of cas es that contain information on th is variable within the ICPSR dataset. In particular, 43 per cent of the cases (346 out of 813 cases) are missing a final offense score. The inclusion of this variable in the analysis would introduce se rious bias problems and
80 risk misestimations of its importance on sentence outcomes. Accordingly, final offense score is not included in the analysis. Howe ver, in an effort to indirectly measure offense serious the total number of counts for which a corporation was convi cted will be used. In theory, the more counts an organization is convicted fo r the more law (higher fines) they should encounter. Since a majority of corporations were convicted of only one count (68 percent), this variable was dichotomized so that 0 equaled corporations that had only one count of conviction and 1 if an organization was convicted on more than one count. Culpability Score Under the or ganizational guidelines, guilty co rporations are assessed a culpability score which is determined by considering aggravating and mitigating circumstances. In particular, judges consider four aggravating factors including an (1) organizations level of tolerance of the crime; (2) the prior criminal history of an organi zation; (3) violation of an existing order; and (4) whether the organization obstructed the criminal investigation. Judges also examine two mitigating factors: (1) whether there was evidence of an effective compliance program in place during the time of the crime; and (2) whether the organization reported the crime. Similar to the final offense score variable, there is significant missi ng data for each of these factors. Five of the variables are missing information in 48 percent (390 cases) of the cases, while one variable (prior criminal history) is missing information on 50 per cent (408 cases) of the cases. The introduction of these variables in the statistical models woul d significantly reduce the nu mber of cases or data that could be sufficiently or effectively used for computation. Conseque ntly, this dissertation examines the calculated culpability score rather than focusing on each of the aggravating and mitigating circumstances separately. Only elev en percent (89 cases) of the cases are missing data on culpability score. The culpability scores range from 0 (less culpability) to 10 (greater culpability).
81 Independent Variables: Extralegal Mea sures of Corporate Offender S tatuses In addition to the legal factors, extraleg al offender characteristics are examined to measure Blacks theoretical propos itions. In terms of the dire ction of the law, Black (1976) asserts that downward law is more common across all setting. Offe nders with lower statuses are more likely to be harshly punished in comparis on to offenders with greater statuses. To operationalize offender statuses information from the ICPSR organizational dataset is used. Specifically, measures of corporate rank (financi al status); integrati onal status (ownership structure); and organizati onal status (number of employees) are gathered. Corporate Rank (Economic Status) Black (1976) predicts that the law varies directly with ec onom ic rank as offenders with greater rank are subject to less la w. At the same time, offenders with less rank should encounter more law or higher fines. While the ICPSR datase t does not contain a direct measure of financial assets or worth, three potential measures include financial status of an organization on the offense date; financial status of an organization on the sentence date; and ability to pay the fine imposed. Due to missing data problems, financial status on offense date (50 percent of cases are missing data) and financial status on sentence da te (34 percent missing da ta) the inclusion of these variables again present statistical proble ms because of the significant reduction in the number of cases that can be potentially analyze d. Thus, a corporations financial status will be measured by their ability to pay a fine. Within the ICPSR dataset, ability to pay is coded 1 for unable to pay entire fine, 2 for unable to pay a por tion of the fine, and 3 for ability to pay fine. This factor was recoded into a dummy variable in which 0 repres ents unable to pay a portion of or the entire fine and 1 indicates an ability to pay the fine. Accordingly, Blacks theory would hypothesize:
82 H1: Corporations unable to pay a portion of or the entire fine (lower economic rank) should be subject to higher fines compared to corporations able to pay the fine imposed (greater economic rank). Corporate Integrational Status (Morphology) In term s of morphology, Black (1976) states that the law varies direct ly with integration. Offenders more socially integr ated enjoy legal advantages th roughout the legal process, while marginalized or less integrated offenders should be subject to greater law or higher fines. To gauge an organizations integrational status, ownership structure is measured. The ICPSR dataset indicates whether an organization is an open or publicly-traded corporation, governmental, nonprofit, closely-held organization, partnership, sole proprietorship, or association. For the purpose of this dissertation, ownership is recoded so that 1 represents socially integrated corporations (publicly-traded, governme nt, nonprofit organizations or association), while 0 indicates marginalized bus inesses (closely-held organizations, partnership, sole proprietorship). Since publicly-traded, gove rnment, and nonprofit agencies are more likely to interact with more social institutions (financial markets, distributors, an d so forth) on a larger scale they are considered more socially integrated. In contrast, smaller closely-held organizations are considered to be more localize d and less likely to inte ract on a larger national scale. Frequency tabulations s how that this variable is highl y skewed as 94 percent of the organizations are coded 0 or closely-held (575 cases); sole proprie tor (23 cases); or partnership (13 cases). Just six percent of the corporations convicted from 2001 to 2004 were coded 1 or publicly-traded (29 cases); government (3 cases); non-profit (3 cases ) or association (1 case). Due to the uneven distribution this measure, it is not included in the subsequent empirical analysis because it is essentially performs as a constant variable.
83 Corporate Cultural Status (Conventionality) Because the ICPSR dataset does no t include e ducational variables it is difficult to adequately or accurately captures an offenders cultural status as commonly measured by Black and other researchers. As such, this dissertation does not include a measure for cultural status or conventionality. Corporate Organizational Status According to Black (1976), law varies with orga nization so that offenders with the capacity for greater collective action have higher organizational status a nd ultim ately greater immunity from the law. Corporations with less organiza tional status are more vul nerable to the law and should theoretically receive higher fines. To ope rationalize organizational status, the number of employees within a particular corporation is measured. Corporations with more employees will be considered to have greater organizationa l status than those with fewer employees. Consequently, an empirical hypothesis anticipates: H2: Corporations with fewer employees (lower organizational status) should be subject to higher fines compared to corporations with more employees (greater organizational status). Corporate Respectabilit y (Soc ial Control) Black (1976) proposes that law varies directly with respectab ility. Offenders with greater levels of respectability are subject to less law, while those with less re spectability are more vulnerable to law. Thus, defendants with a crimin al past (lower respectability) should receive harsher sanctions than offenders w ithout a criminal past (greater respectability). However, the organizational guidelines already include prior criminal history as legal factor that should directly impact sentence decisions and outcomes. Consistent with Blacks arguments, criminal history should increase the fine am ount imposed on corporations.
84 Independent Variables: Structural Mea sures of Federal District Statu s At the structural level, Black contends that th e quantity of law varies by locational statuses. Social settings with greater rank, integrational status (morphology) cultural diversity or distance between an offender and the larger normative cl imate, organizational status, and less normative status or respectability should impose more law or higher fines. The following are measures of federal district characteristics re lated to each social status. Federal District Rank (Economic Status) Black (1976) proposes that locations m arked by greater economic stat us or rank use more law. Thus, federal districts with greater econom ic status will use more law or impose higher fines, while districts with lower economic status will use less law. Acco rdingly, the total dollar amount of wholesale trade sales within each district is used to indicate the economic status of a district4. Federal districts with grea ter wholesale trade sales are considered to have greater rank or economic status than districts with lower earnings, retail sales, and wholesale trade sales. In order to normalize the distribution of this variab le, a log of wholesale trades was created. The resulting empirical hypothesis asserts: H3: Federal districts with great er wholesale trade sales (gre ater economic rank) will use more law or impose higher fines compared to districts with lower wholesale trade sales (lower economic rank). Federal District Integratio nal Status (Morphology) One aspect of m orphology is inte gration or the extent to whic h people participate in major social institutions, including employment and education. Black (1976) hypothesizes greater levels of integration lead to more law. As such, morphology is operationalized as net change in 4 It is important to note that FEDSTATS includes other economic indicators such as the total economic earnings within a district, the total dollar amount of retail sales within each district, and total dollar amount of manufacturing shipments in a district. However, due to highly significant correlations among these variables there are risks of multicolinearity. Since the log of total dollar amount of wholesale trade had the strongest correlation with the log of total fine it was selected as the measur e of structural economic status.
85 housing units within a district. This indicator is used to gauge resident ial mobility rates within each federal district. Black asse rts residential stability is an indication of greater integration compared to locations with resi dential instability. Thus, distri cts with fewer changes in housing units are deemed more integrated than districts with greater cha nges. As a result, the following hypothesis predicts: H4: Federal districts with fewer changes in hous ing units (greater integrational status) will use more law or impose higher fines compared to districts with more changes in housing units (lower integrat ional status). Federal District Cultural Status (C onventionality) According to Black (1976), the greater the cu ltural distance between an offender and the larger cultural environment, the more law will be used. In this way, cultural diversity increases applications of the law while cultural intimacy decr eases applications of law. This dissertation uses political climate as a measure of culture and asserts that the cultural distance between corporations and the larger federal district will be sma ller in Republican dominated environments. Since Republican economic polic ies generally support fewer regulations or sanctions over corporate activity, there is greater cultural intimacy or ties between corporations and the larger climate. In other words, cor porations are considered more conventional in probusiness climates and thus they should encounter less law. To cap ture the political climate of each district the percentage of judges in each dist rict appointed by Republican presidents is used. Also, the racial composition of each district is included as a me asure of structural culture. Black asserts that increa ses in the ethnic population will lead to cultural conflict and thus the use of more law within certain locat ions. In order to capture the ethnic population, the total number of ethnic residents (FEDSTATS includes five majo r ethnic groups including Black; Hispanic or Latino; American Indian or Alaska Native; Asian; and Hawaii an or Pacific Islander) within each
86 district is divided by the total district populat ion and multiplied by 100. The following indicates the empirical hypotheses: H5a: Federal districts with a greater percentage of Republican appointed judges (greater cultural intimacy between offender and larger political climate) will use less law or impose lower fines compared to districts with a lo wer percentage of Republican appointed judges (lower cultural intimacy between offe nder and larger political climate). H5b: Federal districts with a greater percentage of ethnic minorities (greater cultural conflict) will use more law or impose higher fines compared to districts with a lower percentage of ethnic minoritie s (lower cultural conflict). Federal District Organizational Status At the structural level, orga nizational status represents th e presence of administrative offices, associations, and centralization of govern ment. More law is used in locations where there are more organizations per capita. In districts with lower organizational status or fewer organizations per capita, there s hould be less law or lower fi ne amounts. To operationalize district organizational status the total number of private establishments within each district is measured. This measure helps capture the number of organizations within each district. Federal districts with more private establishments are d eemed to have greater organizational status than districts with fewer private esta blishments and building permits. Subsequently, Blacks theory would hypothesize: H6: Federal districts with more private establ ishments (greater organizational status) will use more law or impose higher fines compared to districts with fewer private establishments (lower or ganizational status). Federal District Respectabi lity (Normativ e Status) Black (1976) predicts social settings or locations with greater social control generally have lower respectability and thus are subject to more law. In other words, more law exists in locations considered less respectable. Previous studies have used homicide rates as an indication of the loss or lack of nonlegal control over its residents (Lessa n & Sheley, 1992; Borg & Parker, 2001). Accordingly, homicide rate within each district is used to operationalize normative status
87 as the district level. Federal districts with higher homicide rates are viewed as having lower respectability and thus should have more law or higher fine amounts. Districts with lower homicide rates are considered to have greater respectability and therefore should use less law or impose lower fine amounts: H7: Federal districts with higher homicide rate s (lower respectability) will use more law or impose higher fines compared to districts with lower homicide rates (higher respectability). In sum, this dissertation examines the effect of two legal variables, two extralegal offender statuses, and six district statuses on the log total fine amount. Sp ecifically, the case level (level I) include culpability score, counts of conviction, ability to pay fine, number of employees. At the district level (level II), the log of wholesale trades, cha nges in housing un its, percent of judges in a district that were appointed by Republican judges, per cent of the district population ethnic minority, total number of pr ivate business establishments in a district, and homicide rates in the district. Table 5-1 presents descriptive statistics for each of the legal and extralegal (case or level I) and the district (str uctural or level II) va riables. Furthermore, Table 5-2 provides a correlation matrix among the variables. Summary of Blacks Empirical Hypotheses Related to Corporate Offender and Federal District Statuses Based on Blacks theoretical pr opositions, the subse quent hypotheses restate or summ arize the expected relationship between corporate o ffender and federal distri ct statuses and the quantity of law or fine amounts. Corporate economic status: H1: Corporations unable to pay a portion of or the entire fine (lower economic rank) should be subject to higher fines compared to corporations able to pay the fine imposed (greater economic rank).
88 Table 5-1. Descriptive statistics for case and structural level variables Mean Std. Dev. Minimum Maximum Dependent variable Log total fine amount 8.122 5.54 0 19.3 Level 1 (legal & extralegal variables) Calculated culpability score 4.11 1.98 0 10 Counts of conviction 0.317 0.465 0 1 Ability to pay 0.63 0.484 0 1 Number of employees 1,249.03 11,014.57 0 156,992 Level 2 (structural/district variables) Log of wholesale merchandise sold 23.91 1.24 19.72 26.72 Change in housing units 153,019.95146,876.262930 637,522 % of Republican appointed judges 0.5 0.25 0 1 Race (% population ethnic minority) 0.27 0.16 0.03 0.66 Private businesses within district 91,321.55 79,236.08 16,444 453,897 Homicide rates 5.88 5.21 0.5 41.8 Corporate organizational status: H2: Corporations with fewer employees (lower organizational status) should be subject to higher fines compared to corporations with more employees (greater orga nizational status). Federal district economic status: H3: Federal districts with great er wholesale trade sales (gre ater economic rank) will use more law or impose higher fines compared to districts with lower wholesale trade sales (lower economic rank). Federal district integrat ional status (morphology): H4: Federal districts with fewer changes in hous ing units (greater integrational status) will use more law or impose higher fines compared to districts with more changes in housing units (lower integrat ional status). Federal district cultural status: H5a: Federal districts with a greater percentage of Republican appointed judges (greater cultural intimacy between offender and larger political climate) will use less law or impose lower fines compared to districts with a lo wer percentage of Republican appointed judges (lower cultural intimacy between offe nder and larger political climate).
89 Table 5-2. Correlation matrix for case and district level variable s related to total fine amount (1) (2) (3) (4) (5) (6) (7) (8) (9) (1) Log total fine amount 1.00 (2) Culpability score -.163** 1.00 (3) Final offense level(a) -.160** .098* 1.00 (4) Counts of conviction .008 .013 .178** 1.00 (5) Ability to pay .838** -.110 -.310** -.080* 1.00 (6) Financial status at sentence date (a) .438** -.002 -.199** -.039 .466** 1.00 (7) Financial status at offense date (a) .174** .062 -.082 .051 .194** .546** 1.00 (8) Ownership structure (b) .161** -.197** .020 -.008 .091* .123** .065 1.00 (9) Number of employees .134** -.070 .053 .005 .077 .050 .035 .327** 1.00 (10) Log of wholesale trade sales .094** -.065 .004 .043 .086* -.091* -.095 .037 -.010 (11) Log of total earnings (c) .051 -.063 -.021 .064 .021 -.140** -.096 .033 -.040 (12) Log of retail earnings (c) .039 -.058 .000 .051 .026 -.149** -.126* .036 -.030 (13) Log of manufacturing shipments (c) .086* -.013 -.037 .006 .087* -.085* -.083 -.003 .004 (14) Changes in housing units -.054 -.005 -.028 .044 -.033 -.144** -.042 .044 -.031 (15) Percent judges Republican appted .042 -.003 -.057 -.048 .053 .006 .016 .091* .081 (16) Percent ethnic population -.005 -.065 .050 .047 -.015 -.035 -.023 .085* -.062 (17) Private business establishments -.017 -.058 .016 .085* -.006 -.095* -.099* .019 -.056 (18) Homicide rates -.038 .018 -.019 .048 -.008 .066 .049 -.011 -.018 (19) Restitution .036 .020 -.003 -.023 .029 .037 .016 -.010 -.005 **p<.01 (2-tailed) p<.05 (2-tailed) (a) = Due to a significant number of missing data, these variable were excluded from final HLM models (b) = Due to highly skewed or uneven distribution, this variable is excluded from final HLM models (c)= Due to highly significant correlations with log of wholesal e trades, these variables were excluded from final HLM models b ecause of risks of multicollinearity
90 Table 5-2 Continued (10) (11) (12) (13) (14) (15) (16) (17) (18) (10) Log of wholesale trade sales 1.00 (11) Log of total earnings (c) .840** 1.00 (12) Log of retail earnings (c) .800** .909** 1.00 (13) Log of manufacturing shipments (c) .701** .605** .768** 1.00 (14) Changes in housing units .218** .511** .620** .317** 1.00 (15) Percent judges Republican appted -.143** -.127** -.054 -.075* .097** 1.00 (16) Percent ethnic population .456** .542** .368** .032 .236** -.093** 1.00 (17) Private business establishments .671** .827** .839** .546** .610** -.133** .505** 1.00 (18) Homicide rates -.121** -.041 -.290** -.381** -.092** -.069* .433** -.073* 1.00 (19) Restitution .028 .019 .023 .041 -.001 -.016 -.018 .005 .011 **p<.01 (2-tailed) p<.05 (2-tailed) (a) = Due to a significant number of missing data, these variable were excluded from final HLM models (b) = Due to highly skewed or uneven distribution, this variable is excluded from final HLM models (c)= Due to highly significant correlations with log of wholesal e trades, these variables were excluded from final HLM models b ecause of risks of multicollinearity
91 H5b: Federal districts with a greater percentage of ethnic minorities (greater cultural conflict) will use more law or impose higher fines compared to districts with a lower percentage of ethnic minoritie s (lower cultural conflict). Federal district or ganizational status: H6: Federal districts with more private establ ishments (greater organizational status) will use more law or impose higher fines compared to districts with fewer private establishments (lower or ganizational status). Federal district normative status (respectability): H7: Federal districts with higher homicide rate s (lower respectability) will use more law or impose higher fines compared to districts with lower homicide rates (higher respectability). Figure 5-1 provides a conceptual model that indicates the anticipated impact of both corporate and district status es on total fine amounts. Analytic Strategy Since this dissertation includes a two level stru ctural analysis hierarchical linear modeling (HLM) is used to more accurately examine the exte nt to which offender and federal district level variables affect sentence outcomes. Most social sc ience research involves hierarchical structures in which cases are nested within higher levels of analysis (Bryk & Raudenbush, 1992). For this dissertation, corporate crime cases ar e nested within federal district s. Corporations are likely to be influenced by characteristics of the districts in which they are embedde d or located. Thus, corporations tried and convicted wi thin the same district are likely to encounter similarities that help predict or explain sentence outcomes. In addition, districts are likely to reflect the economic, social, political, and cultural norms of the immediate environments (Ulmer, 2005). Subsequently, statistical techniques are needed to address how cont extual characteristics interact with or influence case level variables. The uses of single-level models such as ordinary leastsquares (OLS) are inappropriate for analyzing multilevel da ta (Bryk & Raudenbush, 1992; Hox, 2002; Kautt, 2002; Britt, 2000; Ulmer & Johnson, 2004: Weidner et al, 20 05). One of the
92 Corporate Offender Statuses Total Fine Amount Corporate Rank (Economic Status) Corporate Integrational Status (Morphology) Corporate Organizational Status Corporate Respectability Ability to pay fine Publicly traded, govt, nonprofit vs closelyheld, sole proprietorship, partnership # of employees Not included Federal District Statuses Total Fine Amount Federal District Rank (Economic Status) Federal District Integrational Status (Morphology) Federal District Organizational Status Federal District Repsectability Log of wholesale trade sales Residential mobility (Net change in housing units) # of private businesses Homicide rates Federal District Cutural Status % Republican appointed judges % Population Ethnic minority *Due to a highly skewed or uneven distribution this variable is not included in final HLM models Figure 5-1. Conceptual model of Blacks theory applied to corporate sentencing: Effects of offender and district statuses on fine amount
93 underlying assumptions of OLS is error terms at one level are unrelated to or independent of higher level variables. As suc h, single-level statis tical techniques generally produce type II errors in which inaccurate p-values provide mi sleading indications of significance. In other words, OLS misestimates the true effects of case le vel factors as well as structural variables on the dependent variable (Bryk & Raudenbush, 1992; Kautt, 2002; Britt, 2000; Ulmer & Johnson, 2004: Weidner et al, 2005). Thus, one of the primary advantages of HLM is the fact that it can simultaneously and more precisely test for the eff ect of offender and structural characteristics on dependent variables (Britt, 2000). Hierarchical linear modeling creates a se ries of successive formal submodels for each level in that analytical st ructure (Bryk & Raudenbush, 1992; Hox, 2002). These submodels produce regression coefficients that estimate the a ffect of higher level influences on case level factors. In addition, the submodels allow the slope and intercept to vary across different levels of analysis. For this dissertati on then, HLM can estimate how (1) legal or extralegal corporate offender statuses (case or level I variables) a ffect sentence outcomes and (2) federal district statuses (level II variables) impact fine amount s. The following formula illustrates the simplest two-level HLM model for fine amounts w ithin and between fe deral districts: Level 1 (Case Level): Y i j = 0 j + e i j where Y i j = log odds of total fine amount for offender i in district j; 0 j = average fine amount in district j; and e i j = the deviation of total fine from the district mean Level 2 (District Level): 0 j = 00 + r0 j where 00 = mean fine amount in district j; and r0 j = the deviation of district mean Combined model: Y i j = 00 + r0 j + e i j
94 As the combined formula indicates, intercepts at level one are deri ved from equations at level two. Thus, HLM creates an average fine amount for each of the federal districts (in this case 84 federal districts) in the an alysis. This allows for a dete rmination of which corporations received fines above or below the average fi ne within the district. Simultaneously, HLM estimates which districts imposed fines above or below the average across or between all federal districts. As applied to this dissertation then, th e case level (level I) pred ictors include two legal variables (culpability score and coun ts of conviction) as well as two extralegal variables related to Blacks social statuses incl uding ability to pay (economic rank) and number of employees (organizational status). At the di strict level (level II), a total of eight measures are incorporated in the model. These structural level variables incl ude the log of wholesale tr ade sales in a district (district economic rank); the change in housing un its in a district (int egrational status); the percentage of judges in a district appointed by Republican presidents (cultural status); the percentage of the district populat ion identified as an ethnic minority (cultural status); the total number of private business establishments in a dist rict (organizational stat us); and homicide rates in a district (respectability). Following Ulmer and Johnson (2004), grand mean centering is utilized because of the interest in assessing sentence variance across district s. Traditionally, inter cepts are interpreted as a certain value when a given independent variable equals zero. Since most of the variables in this dissertation do not have a meaningful zero poin t it is appropriate to change the location of these variables so that a signifi cant interpretation of the intercept can be made. Use of group mean centering, according to Ulmer and Johnson, artifici ally restricts differences at level II. It is therefore more beneficial to use grand mean centering since it more accurately assess level II differences or in this case variations across or between federal districts. Figure 5-2 outlines the
95 final HLM model that includes both case or level I (legal and extraleg al variables) an d district or level II variables. Table 5-3. Summary of the full HLM model specified in equation format Level I Model = B0 + B1*(Culpability score) + B2*(Counts of conviction) + B3* (Ability to pay) + B4* (Number of employees) + R Level II Model B0 = G00 + G01* (Log of wholesale sales) + G02* (Change in housing units) + G03* (% judges Republican appted) + G04* (% Ethnic minority) + G05* (Private businesses) + G06* (Homicide rates) + U0 BI = G10 + U1 B2 = G20 + U2 B3 = G30 + U3 B4 = G40 Despite its usefulness with multilevel data, HLM does have some limitations. First, HLM models are highly sensitive and cannot handle too many independent va riables at level I particularly when cases at level II are sparse. Since the intercept and slopes at the case level (level I) become outcome variable s at level II, it is also importa nt to use theoretically concise models to produce accurate findings (Bryk & Ra udenbush, 1992). In this case, the mean log total fine amount (intercept) and sl opes (culpability score; counts of conviction; ability to pay; and number of employees) become outcome variab les in the level II model (see Figure 5-2). Further complicating the capabilit ies of HLM models, level II in tercept and slopes are permitted to vary for each case at that level. As Kautt (2002: 646, italic in origin al) points out, a two-level HLM model involving federal dist ricts could potentially have 94 different intercepts and 94 different slopes for each case-level variable included in the model. As such, the inclusion of numerous predictors significan tly challenges the statistical re liability of HLM findings. In addition, the introduction of variables with missing data complicates HLM modeling.
96 Finally, HLM models require strict specifica tion of standard error terms in order to perform statistical analyses. Unde r HLM, each case level (level I) predictor or variable can have a random, nonrandom varying, or fixed effect across each level II equa tion (Bryk & Raudenbush, 1992). As applied to this dissertation then, the e ffect of culpability, coun ts of conviction, ability to pay, and number of employees can have a random (allowed to vary across districts), nonrandom varying (allowed to vary under certai n conditions), or fixed (predictors held constant) effect at the district level (level II ). Traditionally, slopes or predictors are fixed because it is assumed they are normally distributed across units of analysis. However, since this dissertation is interested in determining if cas e level factors are diffe rentially applied across districts, each of these variables were allowed to vary randomly with the assumption that they are not normally distributed and their importance differs from case to case. If any of the variables are misspecified at level II, HLM will enc ounter conversion problems and will not conduct subsequent tests. Thus, variab les theoretically expected to va ry across different levels of analysis may be rejected by HLM procedures. Accordingly, several preliminary models are necessary in order to determine which specificati on of error terms is appropriate for analysis (Kautt, 2002). In this case, th e number of employees predictor is fixed while the remaining case level variables are permitted to vary randomly because the HLM model was unable to converge when total number of employees was allowed to vary randomly.
97 CHAPTER 6 PREDICTIVE POWER OF BLACKS THEO RY ON SE NTENCE OUTCOMES FOR ORGANIZATIONAL DEFENDANTS Under HLM, an initial unconditional model or one-way ANOVA test with random effects is necessary because it provides important preliminary information about the amount of variation within and between different levels of analysis This model is considered fully unconditional because there are no predictors specified at either level I or II (B ryk & Raudenbush, 1992). In this case, the preliminary ANOVA test indicates the amount of variance in log total fine amounts that occurs at each level of analysis including within federal districts (case or level 1) and between federal districts (structural or level II). Results of this initial test are illustrated in Table 6-1. The findings indicate a significant difference in average log fi ne amounts between districts. The proportion of variance explained within districts is .9125 which indicates that 91 percent of the variation in mean total fine amount occurs at the case level. C onversely, the amount of variation accounted for between different districts is .08796. Thus, about 9 percent of the total variance in the mean fine amount is accounted for by district level factors. Despite the small amount of variance between district s (level II), it is still theo retically important to further examine the potential predictive power of struct ural characteristics on sentence outcomes (Ulmer and Johnson, 2004). In this case, it is important to determine if Blacks theoretical propositions help explain the variance in fine amounts between different federa l district courts. Level I Random Coefficient Model: Fixed a nd Random Effects Models of Legal Variables Findings in Table 6-2 show the level 1 random coefficient model for the log total fine. In HLM, the fixed effect model is e quivalent to a single-level OLS regression since it indicates 5 The formula used to calculate the amount of variance within districts is = 00 / ( 00 + 2). See Bryk and Raudenbush (1992). 6 The formula used to calculate the amount of variance between districts is = 2/ ( 2+ 00). See Bryk and Raudenbush (1992).
98 Table 6-1. Unconditional model of log total fine amount Fixed effects Coefficien tS.E. t-ratio Df p-value Log total fine 8.25 0.28 29.08 83 0.000 Random effects Variance S.D. df X2 p-value Within districts (Level I) 27.77 5.27 Between districts (Level II) 2.68 1.64 83 164.79 0.000 Proportion of variance explained Level I 91.2 Level II 8.8 how an increase or decrease in particular independe nt variables or coefficients affect changes in the dependent variable. Fixed eff ect model results indicate that ca lculated culpability score has a significant7 effect on total fine amount, albeit, in a direction opposite of that expected. The negative coefficient suggests that the higher a co rporate defendants culpability score, the lower their fine amount. Under the organizational gui delines, increases in culpability score should directly correspond to higher fine amounts. That is, as an offenders culpability score increases so to should the fine amount imposed against them. In HLM, the random effect models estimates whether level I regression coefficients or predictors significantly vary across level II un its (Bryk & Raudenbush, 1992). As applied to this dissertation then, the random eff ect model indicates whether the application of culpability score and counts of conviction significan tly differ or vary across districts. The model intercept indicates significant variation in the mean total fine amount across district even after controlling for the two legal variables (culpability and counts of conviction). In other words, the mean fine amount varies significantly across districts despite organizations that have similar culpability scores or that were convicted on a similar number of counts. In addition, the variance 7 It should be noted this and subsequent empirical findings are based on two-sided tests and thus report significance at the .1 level.
99 component for the counts of convi ction is significant wh ich indicates that districts differentially assess or determine the number of counts impose d. Finally, Table 6-2 in dicates the amount of variance that culpability and c ounts of conviction factors account for at the case (level I) and district (level II). At the cas e level, the inclusion of culpabil ity score and counts of conviction only explain about .058 or 6 percent of the total vari ance for the log of total fine. In other words, these two legal variables explain only a small per centage of the variation in total fine amount in cases that occur within districts. In some ways, su ch a finding appears to undermine the inherent intent or goals of th e organizational guidelines. In theory, culpability score should account for a high percentage of the total varian ce on fine amounts since guideline instructions mandate that a corporations level of culpability be one of the principle factors that directs the judicial sentencing process. However, the lack of legal variables in the model including final offense score and aggravating and mitigating factors undoubtedly influe nces the lack of explained variance at the case level. In terms of the impact of calculated culpabi lity and counts of conviction at the district level, the results show that approximately 15 pe rcent (.147) of the expl ained variance in mean total fine amount between districts can be account ed for by these two legal variables. This suggests that culpability and count s of conviction is differentially applied from one district to another. In some districts, cu lpability may be more or less impor tant in determining total fine amount, while the importance of the counts for which a corporation is convicted tends to similarly vary across districts. Full Random Coefficient Model: Fixed and Ra ndom Effects Models of Blacks Theoretical Concepts In Table 6-3, Blacks social statuses at both the offender and structural level are introduced into the full model. In essence, this mode l provides empirical findi ngs concerning Blacks
100 Table 6-2. Random coefficien t model of log total fine Fixed Effects Coefficient S.E. Intercept 8.13*** 0.283 Calculated culpability score -0.387** 0.118 Counts of conviction 0.168 0.541 Random Effects Variance df X2 Intercept 1.51 35 97.372*** Calculated culpability score 0.403 35 45.478 Counts of conviction 2.261 35 60.919** Level I (Variance within districts) 26.145 Proportion of Variance Explained Level I (Within districts) 0.058 Level II (Between districts) 0.147 ** p<.01 *** p<.001 theoretical hypotheses or e xpectations about the application of th e law directed against corporate defendants sentenced under the organizational guidel ines. At the case level, his concepts of offender statuses including economic status (abi lity to pay fine) and organizational status (number of employees) are entered into the model in order to test hypotheses 1 and 2. At the structural level a total of six variables are incl uded in the model to gauge the explanatory power of hypotheses 3 through 7. These district level variables include measures of economic status (the log of wholesale trade sales) ; integrational status (change in housing units); cultural status (percentage of judges in district appointed by Republican presid ents and percent of district population ethnic minority); organizational status (number of private businesses in district); and respectability (homicide rates). At the case level, the fixed effect model rev eals that both ability to pay and the number of employees have a positive and highly significant effect on total fine amount. The findings indicate that corporations unable to pay a portion or the entire fine amount have lower fines levied against while corporations able to pay the fine are significantly more likely to receive higher fines. In terms of the number of employ ees, corporate defendants with fewer workers are
101 substantially more likely to receive lower fines and offenders with more employees more likely to have higher fines imposed against them. While these two extralegal f actors are statistically significant, they do not support Blacks theoretical propositions. Rather corporate economic and organizational statuses have an affect in a direction opposite of that expected by Black. For instance, Black asserts law varies directly with rank and that all else being equal, lower ranks including the poor are more vulnera ble to applications of the law at every stage of the criminal justice process including th e likelihood of arrest, pros ecution, conviction, and severe punishments. Contrary to this prediction, the results imply that corporate offenders with less economic rank (unable to pay) enco unter less law or lower fines. Conversely, corporations with greater economic status (sufficient financial resour ces to pay fine) have more law or higher fines imposed against them. Similarly, Black anticipates law is greater in a direction toward less organization. Among businesses, he contends that those with more employees, functional duties, subunits, and administrators have greater orga nizational status than smaller co rporations with few workers, functions, or subunits. In other words, greater or ganizational status provides immunity from law. However, the findings indicate otherwise as corp orate defendants with a greater organizational status (higher number of employees) encounter more law or higher fines, while defendants will less organizational status (fewer number of em ployees) have less law (lower fine amounts) imposed against them. Thus, while ability to pay and number of employees are significant predictors of fine amounts such findings do not support hypotheses 1 or 2. Beyond ability to pay and number of employees, the fixed effects mode l illustrates that the counts of conviction
102 Table 6-3. Full random coefficient model of log total fine Fixed effects CoefficientS.E. Intercept 8.04*** 0.168 Legal variables Calculated culpability score -0.044 0.114 Counts of conviction 0.674* 0.300 Extralegal variables (Corporate statuses) Ability to pay (Economic status) 8.987*** 0.441 Number of employees (Organizational status)8 0.027*** 0.008 Structural variables (District statuses) Log of wholesale trades (Economic status) 0.174 0.142 Housing unit changes (Integrational status)4 0.000 0.001 % Republican appted judges (Cultural status) 0.823 0.511 % Population ethnic minority (Cultural status) 0.012 0.011 Private businesses (Organizational status)4 -0.003 0.003 Homicide rates (Normative status) -0.028 0.018 Random effects Variance Df X2 Intercept 0.740 17 34.580** Calculated culpability score 0.161 23 33.412 Counts of conviction 0.625 23 32.645 Ability to pay (Economic status) 6.398 23 55.009*** Level I (Var iance within districts) 7.247 Proportion of Variance Explained Level I (Within districts) 0.722 Level II (Between districts) 0.676 p .10 (two-tailed test) p<.05 ** p<.01 *** p<.001 significantly affects total fine am ounts. In particular, corporati ons convicted of more than one offense had higher total fine amounts. This findi ng is interesting because it was not significant when included in the legal variable model (see Table 6-2). In addition, culpability score which was significant in the legal vari able model now becomes insignifi cant in the full model. This reversal indicates that ability to pay and number of employees cont rol for the effect of culpability and in some ways interacts with counts of conviction to produce significan t differences in total 8 Due to the scale of these original variab les their coefficients were very small. In order to interpret the coefficients, they were all multiplied by 1000 which indicates the expected change in the log total fine amount when the variables increase by 1000.
103 fine amounts imposed against corporate offenders. Since both the extraleg al factors are highly significant, it is plausible that the culpability score becomes meaningless if the defendant is unable to pay a fine or if they have a few employ ees. At level II, none of the structural factors, as predicted by Black, have a si gnificant affect on total fine amounts. As such, hypotheses 3 through 7 are not supported either. More specifically, the economic, integrational, cultural, organizational, or normative statuses of a particular federal district have no significant effect on the application of the law directed toward co rporations sentenced under the federal sentencing guidelines from 2001 to 2004. Stated differently, there is no statistical evid ence that the district statuses or characteristics specified in th e full model impact sentence outcomes. With regards to the random effects model, the results s how significant variation across districts. As previously stat ed, the random effect models in HLM estimate the extent to which level I regression coefficients vary across level II units. In the full, the random effect model indicates whether culpab ility score, counts of conviction, a nd ability to pay significantly vary across districts9. The significant model intercept suggests there is statis tically meaningful variation in the mean total fine amount across district even after controlling for the importance of legal and extralegal variables. In particular, the mean fine amount continues to significantly vary across districts even when corporate culpability scores, counts of conviction, and ability to pay are similar. The variance components for both of the legal variables show they are slightly significant across districts. The meaning or im portance of culpability and counts of conviction differs significantly from one district court to another implying that the organizational guidelines are not performing uniformly across districts (see Appendix B for graphs illustrating the slope of 9 In an initial test, the number of employees was allowed to vary at level II or across districts but HLM was unable to perform this test. This indicates that the full model was misspecified with number of employees allowed to vary randomly. As such, the number of employees was held constant at level II.
104 the relationships between the independent variables and the log total fine amount for each district). Similar to Ulmers (2005) argument, the districts appear to be developing their own unique interpretations of the guidelines which l ead to variations or differential sentence outcomes. Likewise, the extralegal factor (ability to pay) significantly and strongly varies across districts. This indicates that districts differentially asse ss the importance of a corporate defendants ability to pay. Sp ecifically, the ability to pay se ems to be a more important consideration in those distri cts that have low average fine amounts (see Appendix B for graphs)10. Lastly and perhaps more importantly, the fu ll model identifies the amount of explained variance accounted for by including the theoretical predictors in the model. The inclusion of ability to pay and number of employees at the ca se level (level I) dramatically increased the amount of explained variance within districts. Specifically, these two extralegal variables increased the amount of variance in mean fine amount over 65 per cent. In the legal model (see Table 6-2) culpability score and counts of conviction accounted for a mere 6 percent of the variance, however, with the addition of ability to pay and number of employees, 72 percent (.722) of the variation can be accounted for (see Ta ble 6-3). Thus, it appears these two extralegal variables are considerably more important in predicting fine amounts than legal factors. Collectively, the results indicate that much of the 91 percent variation in mean total fine amount that occurs within districts can be explained by culpability score, counts of convic tion, ability to pay, and total number of employees. Concerning the amount of explained variance be tween districts, the inclusion of Blacks concepts as related to offender and district statuses explain approximately 68 percent (.676) total 10 An examination of the Tau as correlation coefficient in dicates a strong and inverse (-.988) relationship between the slope of the relationship between ability to pay and log total fine across districts.
105 variance in average total fine amount (see Table 6-3). However, two comments are warranted with regards to these findings. First, none of the corporate st atus measures operate in the direction predicted by his theory and none of the structural characterist ics are significant. Secondly, it appears all or a subs tantial amount of the variation across districts is due to the ability to pay and total number of employees. Th e positive coefficients su ggest that districts or judges are sympathetic towards corp orate defendants unable to pay a portion of or the entire fine amount and are thus reluctant to impose more law or higher fines. Simila rly, corporate offenders with a small number of employees seem to encount er less law or lower fine s. Overall, then, the ability to pay and number of employees are strong predictors of mean total fine amount within and between districts. At both levels of analysis it appears these two ex tralegal factors help explain at least 50 percent of the variance in fine amounts. Summary of the Empirical Findings The initial A NOVA test showed that most of the variance (91 percent) in total fine amount existed at the case level, while a small but statistically significant amount of variation (9 percent) occurs between or across districts. Findings reveal most of the variation at the case level can be explained by two extralegal vari ables including a corporate offenders ability to pay a fine and its number of employees. It appears the inclusion of these two variables acc ounts for most of the 68 percent explained variance in total fine amount. The importance of these extralegal variables in the sentencing process undermines the formal rationality of the federal guidelines which were established to limit the influence of legally non-r elevant factors. Interestingly, the two legal variables including culpability and counts of conviction seem to have minimal impact in predicting total fine amounts, and no ne of the structural indicators have a direct effect on total fine amount.
106 Another important finding is th e fact that the case level variables, including culpability score, counts of conviction, and ability to pay, va ry significantly across federal districts. This calls into question the uniformity of the organizational guidelines and the goal of the 94 district courts performing like one rationalized machine. It appears that th e district courts are differentially interpreting and applying the organizational guidelines which may lead to noticeable sentencing disparities among similarly situ ated offenders. In particular, it seems that district courts differentially cons ider the importance or value of these case level variables so that in some districts culpability, c ounts of conviction and ab ility to pay are a determining factor in deciding total fine amount while in other districts these variables wield little influence in fine amount decisions. While this dissertation was un able to meaningfully account for the 9 percent of variance that occurs between or across districts, there are underlying structural factors yet to be identified that significantly affect the di fferential interpretation of the guidelines and application of total fine amounts.
107 CHAPTER 7 SUMMARIZING THE IMPORTANCE OF BLAC KS THEORY & FEDERAL SENTENCING GUIDELINES FOR OR GANIZATIONS This dissertation examined sentencing out comes under the federal sentencing guidelines for organizations (FSGO). By and large, the gu idelines reflect a Weberian model of criminal justice in which sentencing decisions and outcomes should be directly re lated to legal factors such as the seriousness of the offense, prior crim inal history of the offender, and the amount of harm caused by the crime (Dixon, 1995). One of the primary objectives of US Sentencing Commission in disseminating the guidelines was the elimination of sentence disparities related to irrelevant extralegal factors in cluding offender characteristics. Post-guideline research, however, continues to find that extral egal factors impact sentence outcomes among organizational offenders (Piquero & Davis, 2004; Beck & O Brien, 2000). To this end, this dissertation analyzed the extent to which corporate offender a nd federal district charac teristics affect fine amounts. Moreover, it used Blacks theory of law as a guiding theoretic al framework since it focuses on the social structure of a case. Social structure of a case includes (1) characte ristics of the actors involved in a conflict (i.e., offender, victim, 3rd party participants such as attorneys or judges), (2) the relational ties or distances among these actors (i.e., intimacy, leve ls of cultural similarity, etc), and (3) characteristics of the geographical location or se tting (neighborhoods, cities or states) in which conflicts arise (Black, 1995; Borg & Parker, 2001) Black (1976) also asse rts the application of the law is influenced or shaped by five statuses or aspects of social lif e including: stratification (economic rank); integrational status (m orphology); cultural status (conventionality); organizational status; and normative status (respectability). In ge neral, he proposes that across all settings offenders with lower statuses are s ubjected to more law including a greater likelihood of being arrested, prosecuted, convicted, and severely punished. In this case, corporate criminals
108 with lower statuses should en counter more law or higher fine amounts compared to defendants with higher statuses. At the same time, Black (1976) anticipates so cial settings or locations with greater social statuses will use more law or be more punitive toward criminal offending. Thus, federal districts with higher st atuses should use more law or impose higher fines while districts with lower statuses should use less law or i ssue smaller fine amounts. Accordingly, Blacks theory is ideally suited to understanding how quantit y of law or fine amounts varies by the social statuses or characteristics of (1) corporat e defendants and (2) fe deral districts. Use of multilevel data necessitated the use of hierarchical linear modeling (HLM) because of its ability to more accurately assess how indepe ndent predictors at different units of analysis impact outcome variables. HLM was also bene ficial because it indica tes how much of the explained variance exists at different levels of analysis includi ng the case (level I) and district level (level II). Relying on organizational data collected from 2001 to 2004, two legal factors (culpability and counts of convictio n) and a total of eight social statuses (two corporate offender statuses and six district statuses ) derived from Blacks theory were included in the final model to determine their effects on mean total fine am ounts. Overall, the findings offer non-supportive evidence for Blacks theoretical ex pectations. While measures of corporate statuses related to economic rank (ability to pay) and organizationa l status (number of employees) significantly influence fine amounts, both do so in a direction contrary to Blacks hypotheses. For instance, Black contends law varies directly with economic rank so that poor defendants are more vulnerable to the law compared to the rich. Empirical evidence contra dicts this proposition. Corporations with lower economic rank (unable to pay a portion of or the entire fine) are significantly more likely to receiv e less law or lower fines while offenders with greater economic rank (ability to pay a fine) encounter more law or higher fines. Likewise Blacks theory asserts
109 an inverse relationship between organizational status and the a pplication of the law. He maintains offenders with lower organizational st atus should encounter more law and defendants with greater organizational status should have less law or lower fines applied against them. However, the results illustrate corporate offenders with less organizational status (smaller number of employees) receive le ss law or lower fines and cor porate criminals with greater organizational status (higher number of employees ) come upon more law or higher fines. In general, the results of this study are consistent with several other evaluations which discover significant findings in a direction opposite of that expected predicted by Blacks theory (Gottfredson & Hindelang, 1979; Braithwait e & Biles, 1980; Myer s, 1980; Mooney, 1986; Hembroff, 1987; and Lessen & Sheley, 1992). C onsequently, these remerging contradictory findings raise important questions about the pr edictive power and utility of his theoretical propositions. In addition, none of the structural statuses had a significa nt impact on total fine amount. Districts with greater economic (gre ater wholesale trade sales); integrational (residential stability or fewer changes in hous ing units); cultural (mor e Republican appointed judges and fewer ethnic minorities); organizationa l (more private business establishments); or normative (lower homicide rates) statuses were not statistically more likely to impose more law or higher fines in comparison to districts with lower statuses. In spite of the lack of theo retical confirmation, the dissertation findings generate some important conclusions. First, the initial unconditional ANOVA random effects model (see table 6-1) reveals that most of the explained variance in mean total fine amount occurs at the case level (level I). More specif ically, 91 percent of the varian ce can be explained by case level predictors while 9 percent of the total variance in average total fine amount takes place across or between districts. This indicat es that there is something about the case (legal or extralegal
110 offender factors) that accounts for most of the va riance in mean total fine amount. Despite the small amount of variance, the fact that district level factors exert some statistically meaningful influence on sentence outcomes is important becaus e it illustrates that the federal district courts are not performing like one unified rational syst em. Instead, contextual factors have some influence on sentencing decision processes. With regards to the case level, this dissert ation accounted for subs tantial amount of the explained variance in total fine amount. In particular, the ability to pay and number of employee dramatically increased the variance to 72 percent when added to the case level model (see Table 6-3). In the legal model, culpability and count s of conviction by themselves only accounted for six percent of the explained variance in average fine amount (see Table 6-2). As such, extralegal rather than legal factors are surprisingly more important in determining total fine amount. Judges are more likely to base th eir sentencing decisions on a cor porate offenders ability to pay a portion of or the entire fine and/or the number of employees they have rather than their level of culpability or the number of counts or crimes they were convicted of. It should also be noted that the legal model which only included culpab ility and counts of conviction revealed an unusual finding. Culpability score was significantly but inversely related to fine amounts. Under the organizational guidelines, fine amount should be commensurate to culpability score. In essence, culpability scores measure the level of responsibility corporatio ns and their officials have in the commission and reporting of the crime. The greater their resp onsibility the greater their punishment or fine should be. The fact that offenders with higher culpability scores were significantly more likely to receive lower fines is peculiar. In the full model, however, the inclusion of ability to pay and number employees ne gates the significance of culpability score. It therefore seems that culpability scores become meaningless in the decision making process if a
111 corporate defendant is unable to pay their fine or if they have a small number of employees. At the district level, it appears ab ility to pay and number of employees are equally effective in terms of explaining the amount of variance in mean fine amount across or between districts. While the two legal variables accounted for approximately 15 percent of the variation between districts (see Table 6-2), the inclusion of Black s social statuses (two corporate offender statuses and six district statuses) helped explai n approximately 68 percent of the differences in fine amounts (see Table 6-3). Since none of the stru ctural measures were significant, it is reasonable that ability to pay and number of employees are respons ible for most of the variation. Another important finding is th e fact that the case level variables including culpability score, counts of conviction, and ability to pay significantly vary signifi cantly across or between federal districts. More specificall y, the inability to pay a fine is a stronger predictor of total fine in districts that impose lower mean fine amounts. Likewise, culpability score and counts of conviction have a weaker but statis tically significant effect on tota l fines in districts that issue lower average fines. Thus, it appears that the di strict courts are differentially interpreting and applying the organizational guidelines In particular, it seems that district courts differentially consider the importance or value of these case level variables so that in so me districts culpability, counts of conviction and ability to pay are a dete rmining factor in deciding total fine amount while in other districts these variables wield little influence in fine amount decisions. This calls into question the formal rationality of the or ganizational guidelines and the idea that the 94 district courts are one unified system. According to Ulmer (2005), the federal sentencing guidelines are more realistically embedded into localized district court communities that are invariably shaped by their provincial social, cu ltural, political, and ideological norms. Each district develops their own processual orders that shape interpretations and definitions of the
112 federal guidelines and ultimately the use or applic ation of the guidelines. Major district court actors including judges use localized discretion and adjust the federa l guidelines to coincide with the immediate contextual or normative climate. In support of these assertions, Ulmer (2005) found the some districts develop re latively broad or liberal definiti ons of the guidelines and thus are more likely to restructure mandated sentenci ng so they are less punitive. Conversely, he found other districts have strict or rigid interpretations and are less likely to depart from the recommended sentences. Ulmer therefore concludes that the effects of legal (offense seriousness, criminal history, trial, and plea) and extraleg al (race and gender) factors on the likelihood of imprisonment and sentence length vary significan tly between districts. In other words, the differential use and interpretation of key guide line provisions ultimately leads to sentence disparities across districts. Th is dissertation lends supporting evidence since the importance or uses of both legal and extralegal variables vary significantly across districts. Significant Contributions and Implications Overall, this dissertation exposes two majo r shortcomings of the federal sentencing guidelines for organizations. First, the volum e of missing data, particularly on key legal variables, is extremely troubli ng and disconcerting. Congress established the US Sentencing Commission with the expectations that the Co mmission would critically review and analyze sentencing data in order to revise refine, or amend policies and prac tices in order to ensure fair and consistent federal sentencing system (N agel & Swenson, 1993). Furthermore, Congress projected that the Commissions efforts a nd assessments would advance knowledge about criminal justice procedures (N agel & Swenson, 1993; Parker & At kins, 1999). However, Parker and Atkins (1999) point out that the Sentencing Commission has c onsistently been unable or unwilling to compile and release reliable organiza tional data. They conclude the organizational guidelines operate as if they are unimportant or as if they do not take themselves too seriously.
113 Parker and Atkins further comment that this colo ssal failure to provide meaningful data has made it extremely difficult to empirically evaluate th e effectiveness of the organizational guidelines and perhaps more importantly advance knowledge about sentencing practices directed against corporate offenders. As applied to this disserta tion, several of the impor tant legal factors were missing information. For example, 43 percent of the 813 cases during th e years of study (2001 to 2004) were missing data on final offense scor e, which in theory, should be a principal determinate of total fine amount. Likewise, cruc ial aggravating and mitigating factors including involvement/tolerance; violation of an existing order; obstruction of ju stice; existence of a compliance program; self-reporting, and prior history of offending were missing data in at least 48 percent of the cases. Further co mplicating the task of conducting e ffective analyses is the fact that the ICPSR data does not include a substan tial number of cases, part icularly cases that involve $1 million or more fines (Alexander, Ar len, & Cohen, 1999). Consequently, there are likely to be significant misestima tions concerning the average pe nalties, including fine amounts, imposed against corporate offenders. Notwithstanding these data problems, the ava ilable organizational information reveals a second troubling aspect of the organizational guidelines, namely the importance of extralegal variables on sentence outcomes. The FSGO were expressly disseminated so that sentencing decisions and outcomes would be directly or e xplicitly related to legally relevant factors including offense seriousness and culpability. Th e use of an objective point system was intended to ensure that similarly situated offenders th at committed comparable crimes and had analogous levels of culpability would be punished uniform ly across different cases and districts. Any sentence disparities could thus be explained by legal differences. In this way, the organizational guidelines sought to diminish the importance of corporate offender characteristics related to
114 financial status or size (Nagel & Swenson, 1993). However, the empirical findings paint a grim picture concerning the importance of legal factors on total fine amounts. The fact that culpability score and counts of conviction expl ained so little of the varian ce in the mean log total fine amount should be alarming. These two variable s accounted for only 6 percent of the explained variance. In theory, culpability scores should be a major determinate of fine amounts. Furthermore, the results indicate that higher culpability scores are significantly related to the imposition of lower fine amounts. Even at the district le vel, extralegal factors again appear to exert more influence in determining fine amounts than legal variables. In contrast, the two extralegal variables exert tremendous influence on fine decisions. In this dissertation, the ability to pay a portion of or the entire fine is the str ongest predictor of total fine amount. Organizations deemed unable to pay were significantly more likely to receive small fine or no fine. Similarly, corporate offenders with fewer employees are subs tantially more likely to receive lower fines. Overall then, this dissertation provides empiri cal evidence that the purported goal of the organizational guidelines utilizing strict and rational instructions to determine sentence outcomes is more of a faade than a rea lity. Instead, it appears the federal district courts haphazardly consider the importance of legal variables a nd more generally base sentence decisions on extralegal factors. In this case, the financia l status of offenders and their overall size as measured by total number of employees play a very important role in determining fine amounts. Future Research Directions In Blacks defense, this disserta tion did not offer a true or comp lete test of his theory. At best, this dissertation offered only a partial analysis of im portant offender statuses. Indeed, just two offender statuses were included in the fi nal model including measures of economic and organizational statuses. While th e indicator of organizational st atus (number of employees) is consistent with Blacks overview of orga nization, the operationalization of economic rank
115 (ability to pay fine) could understandably be viewed as problematic because it fails to directly measure an offenders financial status. Unfort unately, the ICPSR organizational dataset does not include unequivocal measures of economic assets or net worth for each corporate offender. Beyond ability to pay, two other potential indicators assessed the financial status of an organization on offense and sentencing date. These variables determined whether a corporation was financially stable (solvent and operating) or unstable (def unct or out of business; bankrupt; reorganization; financial stress). However, due to significant mi ssing data problems there were a limited number of cases that could be effectively analyzed. Approximately 50 percent of the 813 cases were missing information on financial status at offense date, while 34 percent were missing data on financial status at sentence date. Consequently ne ither could be used in the final analyses. Among the structural predictors, most measures used were consistent with own Blacks writings and previous assessments of his theory. For instance, similar to other studies this dissertation used residen tial mobility (integrational status), ethnic population (cultural status), number of busine sses (organizational stat us), and homicide rates (normative status or respectability) as measures of structural stat uses (Lessan & Sheley, 1992; Borg & Parker, 2001; Gottfredson & Hindelang, 1979; Bra ithwaite & Biles, 1980). However, the use of wholesale trade sales (measure of district economic status) and political climate (culture) could conceivably be questioned as crude measur es of Blacks concepts. It is also possible that the focus on total fine amount may be an inappropriate dependent variable or unit of analysis. Black proposes th ere are different styles of law including penal, compensatory, therapeutic, and conciliatory forms of social control. While this study focused on penal social control (criminal sanc tions or fines), research suggests the threat and/or use of civil litigation is the primary means of controlling and punishing co rporate criminals (Sutherland,
116 1949; Clinard & Yeager, 1980; Snider, 2000; Pear ce & Tombs, 2002). Victims generally seek punitive or monetary damages for injuries or loss es suffered as a result of corporate crimes. Civil litigation, according to Black, is indicative of compensatory styles of law or social control because victims demand payments for unfulfille d obligations. Accordingly, a more germane dependent variable may be civil or compensatory outcomes. More specifically, Blacks theory may be better suited for predicting how the social statuses of offenders and social locations impact compensatory damages levied against corporate offenders. In addition, this dissertation faile d to analyze the complete social structure of a case. Black (1995) contends that a true examination of the behavior of law should focus on (1) actors (2) relational distances, and (3) social setting. In general, this st udy conducted a micro-level (focus on offender characteristics) and macro-level (foc us on district statuses) analysis of Blacks theory. While it was beyond the scope of this dissertation, a more complete test would examine the importance of meso-level f actors such as the relational ti es between offender and victims (intimates, strangers, acquainta nces) and/or the relationship between offender and courtroom actors including judges and their effects on sentence outcomes. According to Black (1976), the greater the distance between offender and other part icipants the more law expected. Results of this dissertation may indicate that the relationa l distance between some corporate offenders and their victims or the even the j udges was smaller and as a result such defendants encountered less law or lower fines. In terms of the micro-level analysis, no facts were gathered on the statuses of the victims or courtroom actors. For instan ce, having knowledge or information about the characteristics of sentencing judges could poten tially account for the significant relationship between the inability to pay fine amounts and r eceiving lower fines. In some cases, offenders and judges could have shared similar statuses a nd been closely related a nd thus less law would
117 be expected. As Black (1976:16) states that, a judge with lowly origins might also be more sympathetic [towards a poor defenda nt] than a judge from the higher levels of society. As such, judges may have been more compassionate or un derstanding of corporate offenders unable to pay fine or smaller defendants (companies with fewer employees) and thus were more likely to impose less law or lower fines. Finally, while the results do not support Blacks theory, it may be that his theory is more applicable at a different stage of the criminal jus tice process, specifically th e prosecutorial stage. Some of the findings suggest th at corporations are differen tially prosecuted based on their statuses or characteristics. For example, a relatively small number of corporate crime cases occurred during the years of study. There were a total of 813 cases whic h translates to an average of only 203 per year among the 94 federal di strict courts. The fact that twenty three districts had two or fewer cases during this time span, including six districts without a single case, raises at least two importa nt questions. Are corporations not committing offenses or are prosecutors reluctant to pursue corporate crime cas es? If Blacks theory holds true, a plausible explanation may be that prosecutors are reluctan t to try offenders with greater organizational status (corporations) and more in terested in pursuing offenders with less organizational status (individuals). In assessing the characteristics of the offenders senten ced, a glaring pattern emerges. The vast majority of corporate offende rs fined are small busine sses. Nearly 95 percent of the organizations sentenced were small closely-held corporations and 80 percent employed less than 100 workers. This is consistent with corporate crime research that asserts the most punitive sanctions are meted out against the smal lest, most peripheral offenders (Sutherland, 1949, Clinard & Yeager, 1980; Shover, Cllela nd, & Lynxwiler, 1986; Yeager, 1991; Goff, 1993). It therefore appears that the only corporate criminals are small organizations and larger
118 businesses are model citizens that refrain from criminal behavior. Using Blacks theoretical framework, however, a more realistic explanation or possibility is that corporate offenders with less integrational and organizationa l status are more vulnerable to the law in terms of likelihood of prosecution. In addition, the economic status of corporate offenders could be a determining factor in the decision to prosecute or not. Thus, Blacks theory c ould be useful in determining if discretionary decisions to pros ecute corporate crime cases vary by the statuses of corporate offenders. By the time a case reaches the se ntencing phase the degree of variation among offenders is almost non-existent and consequently the amount of variation in legal outcomes is minimal as well. At the federal level, Pontell, Calavita, and Tillman (1994) analyzed responses in legal outcomes to the savings and loans scandals of the 1980s and found prosecutors were reluctant to pursue white-collar and corporate crime cases. In part, the complex nature of financial cases and attempting to uncover convoluted business trans actions and schemes makes the investigation process laborious and time-consuming. In addi tion, federal prosecutors typically have high caseloads and other offenses that garner greater priority in comparison to fraud cases. Moreover, if corporate crime cases reach the trial stage, prosecutors confront difficulties in terms of presenting elaborate crimes in laymen terms to bot h juries and judges. A ccordingly, the intricate nature of these crimes impedes the probabilities of conviction because such cases are difficult to fully investigate and comprehend. In one of the most comprehensive studies of prosecutorial discretion, Benson and Cullen (1998) examined what factors affect state prosecuto rs decisions to pursue corporate crime cases. Overall, they found that state prosecutors display considerable trepidation in initiating the legal process against corporate offenders. Indeed, they identified at least five common considerations
119 that ultimately hinder or restrict prosecutors de sire to commence criminal litigation, including a lack of available resources, criminal justice priorities, legal constraints, and political and economic considerations. Many prosecutors are reluctant to devote scarce resources toward corporate crime cases because they require a considerable amount of investigation and preparation. Corporations also have adroit atto rneys who can delay the legal process through the filing of motions or through attempts to limit prosecutorial access to critical information. In other situations, prosecutors noted that street offenses such as drug and violent cases receive greater priority or attention in comparison to corporate crimes. With regards to legal constraints, prosecutors mentioned that it is difficult to transf er or apply traditional individual-level statutes or laws toward corporations. Also, because of th e structural complexity of large corporations it is difficult to pinpoint liability, es tablish probable cause, or criminal intent against corporations. Furthermore, Benson and Cullen found that pros ecutors generally succumb to partisan and economic pressures primarily because they occupy politically appointed or elected positions. Since corporations play an important role in the political and economic arenas (principally through campaign contributions and vital jobs or resources to the surrounding communities) prosecutors may be reluctant to pursue cases ag ainst certain corporate offenders. Undoubtedly, their careers are tied to the cases they do and do not pursue. In the wake of the controversial firing of seven federal prosecutors who purportedl y failed to aggressively pursue cases against Democratic interests this seems like a stark reality (Johnston, Lipton, and Zeleny, 2007). As such, this dissertation may have be en better served if it focused on the political characteristics of prosecutors as a measure of dist rict cultural climate instead of judges. Instead of inquiring if districts with a greater number of Republican appointed judges im pose less law, a more relevant question would have been to what degree do Rep ublican prosecutors impose less law or have
120 lower rates of corporate prosecutions compared to Democratic prosecutors. Since it is customary for presidents to appoint all federal prosecutors, an appropriate test w ould require a comparison of corporation prosecutions during Republican and Democratic presidencies. Overall, it appears prosecutors are highly influenced by the so cial statuses of corporate offenders. Prosecutors seem more likely to pursue cases against sm aller, financially poor corporations because it easier to identify the o ffender(s), gather evidence, establish intent or liability, ensure convictions, and ultimately is a better use of their resources and time (Benson & Cullen, 1998; Pontell, Calavita, & Tillman, 1994; Piquero & Davis, 2004). By contrast, larger corporations with greater fina ncial resources complicate the le gal process by making prosecution too time-consuming or costly. In the context of Blacks theory, future research could examine how the economic, integrational, cultural, organizational, and nor mative (respectability) statuses of corporate offenders and districts impacts the behavior of law or the quantity of prosecutions directed against corporations.
121 APPENDIX A NUMBER OF CASES PER FEDERAL DIST RICT (2001-2004) Federal district Cases Federal district Cases Federal district Cases Alabama North 5 Kansas 1 Ohio North 25 Alabama Middle 1 Kentucky East 5 Ohio South 17 Alabama South 1 Kentucky West 3 Oklahoma North 3 Alaska 2 Louisiana East 27 Oklahoma East 0 Arizona 3 Louisiana Middle 10 Oklahoma West 1 Arkansas East 5 Louisiana West 8 Oregon 7 Arkansas West 2 Maine 6 Pennsylvania East 27 California North 30 Maryland 11 Pennsylvania Middle 7 California South 14 Massachusetts 7 Pennsylvania West 4 California East 8 Michigan East 24 Rhode Island 4 California Central 41 Michigan West 7 South Carolina 11 Colorado 3 Minnesota 9 South Dakota 2 Connecticut 13 Mississippi North 0 Tennessee East 9 Delaware 4 Mississippi South 3 Tennessee Middle 2 District of Columbia 11 Missouri East 14 Tennessee West 1 Florida North 2 Missouri West 7 Texas North 24 Florida Middle 42 Montana 3 Texas South 17 Florida South 44 Nebraska 3 Texas East 7 Georgia North 16 Nevada 3 Texas West 8 Georgia Middle 3 New Hampshire 3 Utah 3 Georgia South 0 New Jersey 15 Vermont 7 Hawaii 0 New Mexico 1 Virginia East 23 Idaho 4 New York North 9 Virginia West 10 Illinois North 12 New York South 38 Washington East 1 Illinois Central 2 New York East 25 Washington West 13 Illinois South 10 New York West 7 West Virginia North 0 Indiana North 6 North Carolina East 13 West Virginia South 8 Indiana South 6 North Carolina Middle 2 Wisconsin East 2 Iowa North 2 North Carolina West 3 Wisconsin West 5 Iowa South 4 North Dakota 2 Wyoming 0 Total 813
122 APPENDIX B GRAPHS OF SLOPE RELATIONSHIPS 0 3.75 7.50 11.25 15.00Predicted Mean Logged Total Fine Amount -0.63 0.37Ability to Pay Figure B-1. Predicted slopes of ability to pay and logged total fine amount in a sample of districts 0 3.75 7.50 11.25 15.00Predicted Mean Logged Total Fine Amount -0.63 0.37Ability to Pay Figure B-2. Predicted slopes of ability to pay and logged total fine amount in all districts
123 10.00 11.25 12.50 13.75 15.00Predicted Mean Logged Total Fine Amount -0.32 0.68Number of Counts Figure B-3. Predicted slopes of number of counts and logged total fine amount in a sample of districts 10.00 11.25 12.50 13.75 15.00Predicted Mean Logged Total Fine Amount -0.32 0.68Number of Counts Figure B-4. Predicted slopes of number of counts and logged tota l fine amount in all districts
124 10.00 11.25 12.50 13.75 15.00Predicted Mean Logged Total Fine Amount -3.63 -1.63 0.37 2.37Culpability Score Figure B-5. Predicted slopes of number of counts and logged total fine amount in a sample of districts 10.00 11.25 12.50 13.75 15.00Predicted Mean Logged Total Fine Amount -4.13 -1.63 0.87 3.37 5.87Culpability Score Figure B-6. Predicted slopes of number of counts and logged tota l fine amount in all districts
125 LIST OF REFERENCES Adm inistrative Office of the United States Courts. 2007. 2006 Annual Report of the Director: Judicial Business of th e United States Courts Washington, DC: US Government Printing Office. Albonetti, Celesta. 1997. Sentencing under th e Federal Sentencing Guidelines: Effects of defendant characteristics, guilty pleas, and departures on sentence outcomes for offenses, 1991-1992. Law and Society Review 31:789-822. Alexander, Cindy, R. J. Arlen, and Mark A. Cohen. 1999. Regulating corporate criminal sanctions: Federal guidelines and th e sentencing of public firms. Journal of Law and Economics 42:393-422. Alliance of Justice. 2007. Federa l judiciary by appointing president. http://allianceforjustice.org. Alschuler, A lbert W. 2005. Disparity: The normative and empirical fa ilure of the Federal Guidelines. Stanford Law Review 58:85-119. Alumbaugh, Steve and C. K. Rowland. 1990. The links between platform-based appointment criteria and trial judges abortion judgments. Judicature 74:153-162. Akers, Ronald L. 1999. Criminological Theories, 3rd edition Los Angeles, CA: Roxbury. Aullette, J and Raymond Michalowski. 1993. Fire in hamlet: a case study of state-corporate crime. Pp. 171-206 in Political Crime in Contemporary America, ed. K. Tunnell. New York: Garland Publishing. Austin, Jay E., John M. Carter II, Bradley D. Klein, and Scott E. Schang. 2004. Judging NEPA: A hard look at judi cial decision making under the Nationa l Environmental Policy Act. Environmental Law Institute. Washington, DC. Beck, Matthew E. and Matthew E. Obrien. 2000. Corporate cr iminal liability. American Criminal Law Review 37:261-285. Benson, Michael L. and Francis T. Cullen. 1998. Combating Corporate Crime: Local Prosecutors at Work. Boston, MA: Northeaste rn University Press. Benson, Michael L. and Esteban Walker. 1988. Sentencing the white-collar offender. American Sociological Review 53:294-302. Black, Donald. 1976. The Behavior of Law. New York: Academic Press. ----------. 1995. The epistemology of pure sociology. Law and Social Inquiry 20:829-870. Borg, Marian J. and Karen F. Parker. Mobilizing law in urban areas: Th e social structure of homicide clearance rates. Law and Society Review 35:435-466.
126 Braithwaite, John and David Biles. 1 980. Empirical verification and Blacks The Behavior Of Law American Sociological Review 45:334-338. Britt, Chester L. 2000. Social context and ra cial disparities in punishment decisions. Justice Quarterly 17:707-732. Bryk, Anthony S. and Stephen W. Raudenbush. 1992. Hierarchical Linear Models Newbury Park, CA: Sage. Calavita, Kitty. 1983. The demise of the occupati onal safety and health administration: A case study in symbolic action. Social Problems 30:437-448. Calavita, Kitty, Robert Tillman, and Henry N. P ontell. 1997. The savings and loan debacle, financial crime, and the state. Annual Review of Sociology 23:19-38. Calavita, Kitty and Henry N. Pontell. 1990. Heads I win, tails you lose. C rime and Delinquency 36:309-341. Canes-Wrone, Brandice. 2003. Bureaucratic de cisions and composition of the lower courts. American Journal of Political Science 47:205-214. Carp, Robert A., Kenneth L. Manning and Ronald Stidham. 2004. The decision-making behavior of George W. Bu shs judicial appointees. Judicature 88:20-28. ----------. 2001. President Clinton s district judges: Extreme liber als or just plain moderates? Judicature 84:282-288. Carp, Robert A. and Ronald Stidham. 2001. Judicial Process in America 5th edition. Washington, D.C.: Congre ssional Quarterly Inc. ----------. 1983. Policymaking and Politics in the Federal District Courts Knoxville, TN: The University of Tennessee Press. Center for American Progress. 2005. Progressive Priorities: An Action Agenda for America. Washington, DC: Center for American Progress. Chambliss, William J. and Robert Seidman. 1982. Law, Order, and Power, 2nd edition. Reading, MA: Addison-Wesley. Chemerinsky, Erwin. 2003. Federal judicial selec tion in the new millennium: ideology and the selection of federal judges. University of California-Davis Law Review 36:619-631. Chiricos, Theodore G. and Miriam A. DeLone. 1992. Labor surplus and punishment: a review and assessment of theory and evidence. Social Problems 39:421-446. Clarke, Philippa & Blair Wheaton. 2007. A ddressing Data Sparseness in Contextual Population Research. Sociological Methods & Research 35: 311-351.
127 Clinard, Marshall B., and Richard Quinney. 1973. Criminal Behavior Systems: A Typology New York: Holt, Rinehart, and Winston. Clinard, Marshall B., and Yeager, Peter. C. 1980. Corporate crime. New York: Free Press. Cohen, Mark. A. 1992. Environmental crime a nd punishment: Legal/economic theory and empirical evidence on enforcement of federal environmental statutes. The Journal of Criminal Law and Criminology 8:1054-1108. ----------. 1991. Explaining judicial behavior or whats unconstitutional about the sentencing commission. The Journal of Law, Economics, and Organization 7:183-199. ----------. 1989. Corporate crime and punishme nt: A study of social harm and sentencing practices in the federal courts, 1984-1987. American Criminal Law Review 26:605-662. Congressional Quarterly. 2002. Presidential Elections, 1789-2000 Washington, DC: Congressional Quarterly Press. Desio, Paula J. 2004. Introduction to organi zational sentencing and the U.S. Sentencing Commission. Wake Forest Law Review 39:559-563. Dixon, Jo. 1995. The organizational context of criminal sentencing. American Journal of Sociology 5:1157-1198. Doyle, Daniel P. and David F. Luckenbill. 1991. Mobilizing Law in Response to Collective Problems: A Test of Blacks Theory of Law. Law and Society Review 103-116. Eisinger, Peter K. 1988. The Rise of the Entrepreneurial State. Madison, WI: University of Wisconsin Press. Ermann, M. David and Richard J. Lundman. 2002. Corporate and Governmental Deviance, 6th edition New York: Oxford University Press. Ferrell, O. C., Debbie Thorne LeClair, and Linda Ferrell. 1998. The Federal Sentencing Guidelines for Organizations: A Fr amework for Ethical Compliance. Journal of Business Ethics 17: 353-363. Flemming, Roy B., Peter F. Nardul li, and James Eisenstein. 1992. The Craft of Justice Philadelphia, PA: University of Pennsylvania Press. Friedrichs, David. 2004. Enron Et Al.: Paradigmatic White Collar Crime Cases for the New Century. Critical Criminology. 12: 113-132. Geis, Gilbert. 1967. White collar crime: The heavy electrical equipment antitrust case of 1961, Pp. 139-151 in Criminal Behavior Systems: A Typology edited by M. Clinard and R. Quinney. New York: Holt, Rinehart and Winston.
128 ----------. 2002. White-C ollar Crime Pp. 37-52 in Controversies in White-Collar Crime edited by G. W. Potter. Cincinnati, OH: Anderson Publishing Co. George, Tracey E. 2001. Court Fixing. Arizona Law Review 43: 9-62. Giles, Michael W., Virginia A Hettinger, and Todd Peppers. 2001. Picking Federal Judges: A Note on Policy and Partisan Selection Agendas. Political Science Quarterly. 54: 623641. Goff, Colin. 1993. Sanctioning Corporate Criminals, Pp. 239-262 in Understanding Corporate Criminality edited by M. B. Blankenship. New York: Garland Publishing, Inc. Goldman, Sheldon. 2005. Judicial Confirmation Wa rs: Ideology and the Battle for the Federal Courts. University of Richmond Law Review 39: 871-908. ----------. 1997. Picking Federal Judges: Lower Cour t Selection From Roosevelt Through Reagan New Haven, CT: Yale University Press. ----------. 1989. Reagans judicial legacy: Completi ng the puzzle and summing up. Judicature. 72: 318-330. Goldman, Sheldon, Elliot Slotnic k, Gerard Gryski, and Sara Schiavoni. 2005. W. Bushs Judiciary: The First Term Record. Judicature 88: 244-275. Goldman, Sheldon and Thomas P. Jahnige. 1985. The Federal Courts as a Political System. New York: Harper and Row. Goodman, Robert. 1979. The Last Entrepreneurs. New York: Simon and Schuster. Gordon, D. M., R. Edwards, and M. Reich. 1982. Segmented work, divided workers Cambridge: Cambridge University Press. Gottfredson, Michael and Mich ael Hindelang. 1979. A Study of The Behavior of Law American Sociological Review 44: 3-18. Gould, Lewis L. 2003. Grand Old Party New York: Random House. Grant II, Don Sherman and Ramiro Martinez Jr. 1997. Crime and the Restructuring of the US Economy: A Reconsideration of the Class Linkages. Social Forces 75: 769-799. Grant II, Don Sherman and Michael Wallace. 1994. The Political Economy of Manufacturing Growth and Decline across the American States, 1970-1985. Social Forces 73: 33-63. Greenberg, David F. 1983. Donald Black s Sociology of Law: A Critique. Law and Society Review 17 (2): 337-368.
129 Gruner, Richard S. 1992. Just Punishment and Adequate Deterrence for Organizational Misconduct: Scaling Economic Penaltie s under the New Corporate Sentencing Guidelines. Southern California Law Review 66:225-288. Hagan, John, Ilene H. Nagel, and Celesta Albonetti. 1980. The Differential Sentencing of White-Collar Offenders in Ten Federal District Courts. American Sociological Review 45: 802-820. Hagan, John, Ilene H. Nagel, and Celesta Albone tti. 1982. The social organization of whitecollar sanctions: A study of pros ecution and punishment, Pp. 259-275 in White-collar and economic crimes edited by P. Wickman and T. Dailey. Lexington, MA: Lexington. Hembroff, Larry A. 1987. Testing Blacks Theory of Law. American Journal of Sociology 93: 322-347. Heydebrand, Wolf and Carrol Seron. 1990. Rationalizing Justice: The Political Economy of Federal District Courts. New York: State University of New York Press. Hofer, Paul J., Kevin R. Blackwell, and R. Barry Ruback. 1999. The Effect of the Federal Sentencing Guidelines on Inter-J udge Sentencing Disparities. Journal of Criminal Law and Criminology 90: 239-321. Hox, Joop J. 2002. Multilevel Analysis: Techniques and Applications Mahwah, NJ: Lawrence Erlbaum Associates. Izraeli, Dove and Mark S. Schwartz. 1998. What can we Learn from the U.S. Federal Sentencing Guidelines for Organizational Ethics? Journal of Business Ethics. 17: 1045-1055. Jacobs, David and Ronald Helms. 2001. Toward a political sociology of punishment: Politics and changes in the incarcerated population. Social Science Research 30: 171-194. ----------. 1996. Toward a Polit ical Model of Incarceration: A Time-Series Examination of Multiple Explanations for Prison Admission Rates. American Journal of Sociology 102: 323-357. Johnston, David, Eric Lipton, and Jeff Zeleny. 2007. White House Said to Prompt Firing of Prosecutors. New York Times March 13: Section A, page 1. Kasarda, John and Michael D. Irwin. 1991. National Business Cycle and Community Competition for Jobs. Social Forces 69: 733-761. Kautt, Paula. M. 2002. Location, location, locati on: Interdistrict and in tecircuit variation in sentencing outcomes for federa l drug-trafficking offenses. Justice Quarterly, 19: 633-671.
130 Kautt, Paula and Spohn, Cassie. 2002. Crack-ing down on black drug offenders? Testing for interactions between offender race, drug type, and sentencing strategy in federal drug sentences. Justice Quarterly, 19: 1-35. Kotz, David M. 2003. Neoliberalism and the Soci al Structure of Accumulation Theory of LongRun Capital Accumulation. Review of Radical Political Economics 35: 263-270. Kramer, Ronald C., Raymond J. Michalowski, an d David Kauzlarich. 2002. The origins and development of the concept and theo ry of state-corporate crime. Crime and Delinquency 48: 263-282. Kruttschnitt, Candace. 1980-1981. Social Status and Sentences of Female Offenders. Law and Society Review 15: 247-265. ----------. 1982. Women, crime, and dependency. Criminology 19: 495-513. Lambiras, Jon J. 2003. White-Collar Crime: Wh y the Sentencing Disparity Despite Uniform Guidelines? Pepperdine Law Review. 30: 459-528. Laufer, William S. 2002. Corporate prosecuti on, cooperation, and the trading of favors. Iowa Law Review 87: 643-667. Lessan, Gloria T. and Joseph F. Sheley. 1992. Does Law Behave? A Macrolevel Test of Blacks Propositions on Change in Law. Social Forces 70: 655-678. Lofquist, William S. 1993. Legislating Organizational Probation: State Capacity, Business Power, and Corporate Crime Control. Law and Society Review 27: 741-773. Lyles, Kevin L. 1997. The Gatekeepers: Federal District Courts in the Political Process Westport, CT: Praeger. Mann, K., Stanton Wheeler, and A Sarat. 1980. Sentencing the white-collar offender. American Criminal Law Review 17: 479-500. Menendez, Albert J. 2005. The Geography of Presidential Elections in the United States, 18682004. Jefferson, NC: McFarland and Co Inc. Michalowski, Raymond J. and Susan M. Carls on. 2000. Crime, Punishment, and Social Structures of Accumulation. Journal of Cont emporary Criminal Justice. 16: 272-292. Mokhiber, Russell. 1988. Corporate Crime and Violence San Francisco, CA: Sierra Club Books. Mooney, Linda A. 1986. The Behavior of Law in a Private Legal System. Social Forces 64 (3): 733-750. Murphy, Diana E. 2002. The Federal Sentencing Guidelines for Organizations: A decade of promoting compliance and ethics. Iowa Law Review 87: 697-719.
131 Myers, Martha A. 1980. Predicting the Behavior of Law: A Test of Two Models. Law and Society Review 14: 836-857. Nagel, Ilene H., and Winthrop M. Swenson. 1993. The Federal Sentencing Guidelines for Corporations: Their Development, Theore tical Underpinnings, and Some Thoughts About Their Future. Washington University Law Quarterly 71 205-259. Parker, Jeffrey S. and Raymond A. Atkins. 1999. Did the corporate criminal sentencing guidelines matte? Some prelim inary empirical observations. Journal of Law and Economics. 42: 423-453. Pearce, Frank and Steve Tombs. 2002. States, Corporations, and the New World Order. Pp. 185-222 in Controversies in White -Collar Crime edited by G. W. Potter. Cincinnati, OH: Anderson Publishing Co. Piquero, Nicole Leeper and Jason L. Davis. 2004. Extralegal Factors and the Sentencing of Organizational Defendants: An Examinati on of the Federal Sentencing Guidelines. Journal of Criminal Justice. 32: 643-654. Podgor, Ellen S. and Jerold H. Israel. 1997. White Collar Crime in a Nutshell, 2nd Edition St Paul, MN: West Publishing Co. Pontell, Henry N., Kitty Calavita, and Robert Tillman. 1994. Corporat e Crime and Criminal Justice System Capacity: Government Response to Financial Institution Fraud. Justice Quarterly 11: 383-410. Quinney, Richard. 1980. Class, State, and Crime, 2nd Edition New York: Longman ----------. 1974. Critique of Legal Order: Crime Control in Capitalist Society. Boston: Little, Brown, and Company. Revesz, Richard L. 1997. Environmental Re gulation, Ideology, and the D.C. Circuit. Virginia Law Review 83: 1717-1772. Richardson, Richard J and Kenneth N. Vines. 1970. The Politics of Federal Courts: Lower Courts in the United States Boston, MA: Little, Brown. Rodriguez, Laurie J. and Davi d E. Barlow. 1999. Structural conditions and the United States Sentencing Commission. Crime, Law, and Social Change 32: 169-202. Rowland, C.K., and Bridget Jeffery Todd. 1991. Where you stand depends on who sits: Platform Promises and Judicial Gatekeep ing in the Federal District Courts. Journal of Politics : 53: 175-185. Savchak, Elisha Carol, Thomas G. Hansford, Donald R. Songer, Kenneth L. Manning, and Robert A. Carp. 2006. Taking it to the Next Level: The El evation of District Court Judges to the US Court of Appeals. American Journal of Political Science 50: 478-493.
132 Savelsberg, Joachim. 1992. Law that does not fit Society: Sentencing Guidelines as a Neoclassical Reaction to the Dile mmas of Substantivized Law. American Journal of Sociology 97: 1346-81. Scherer, Nancy. 2005. Scoring Points: Politicians, Activists, and The Lower Federal Court Appointment Process. Stanford, CA: Stanford University Press. Schanzenbach, Max. 2005. Have Federal Judg es Changed Their Sentencing Practices? The Shaky Empirical Foundations of the Feeney Amendment. Journal of Empirical Legal Studies 2: 1-48. Schwartz, Herman. 1988. Packing the Courts. New York: Charles Scribners Sons. Segal, Jeffrey A. and Harold Spaeth. 1993. The Supreme Court and the Attitudinal Model. New York: Cambridge University Press. Shapiro, Susan A. 1985. The road not taken: The elusive path to criminal prosecution for white-collar offenders. Law and Society Review 19: 179-217. Sheldon, Charles H. and Linda S. Maule. 1997. Choosing justice: The recruitment of state and federal judges. Pullman, Washington: Washington State University Press. Shover, Neal, Donald A. Cle lland, and John Lynxwiler. 1986. Enforcement or Negotiation: Constructing a Regulatory Bureaucracy. Albany, NY: State University of New York Press. Smith, Kevin B. 2004. The politics of punishme nt: Evaluating politic al explanations of incarceration rates. The Journal of Politics 66: 925-938. Smith, Douglas A. 1987. Police Response to Interpersonal Violence: Defining the Parameters of Legal Control. Social Forces. 65: 767-782. Snider, Laureen. 2003. Researching Corporate Crime, Pp. 49-68 in Unmasking the Crimes of the Powerful edited by S Tombs and D Whyte. New York: Peter Lang. ----------. 2000. The Sociology of Corporate Crime: An obituary. Theoretical Criminology. 4 (2): 169-206. ----------. 1993. Regulating Co rporate Behavior, Pp. 177-210 in Understanding Corporate Criminality edited by M. B. Blankenship. New Yo rk: Garland Publishing, Inc. ----------. 1990. Cooperative m odels and corporate crime. Crime and Delinquency 36: 373-390. Solum, Lawrence B. 2005. Judicial Selection: Ideology Versus Character. Cardozo Law Review 26: 659-689.
133 Songer, Donald R., Susan W. Johnson, and R onald Stidham. 2003. Presidential Success Through Appointments to the Unite d States District Courts. The Justice System Journal. 24: 283-300. Steffensmeier, Darrell and Stephen Demuth. 2000. Ethnicity and Sentencing Outcomes in US Federal Courts: Who is Punished More Harshly? American Sociological Review 65: 705-729. Steffensmeir, Darrell, Jeffery T. Ulmer, and J ohn H. Kramer. 1998. The Interaction of Race, Gender, and Age in Criminal Sentencing: The Punishment Cost of Being Young, Black, and Male. Criminology 36: 763-798. Stucky, Thomas D., Karen Heimer, and Joseph B. Lang. 2005. Partisan Politics, Electoral Competition and Imprisonment: An An alysis of States over time. Criminology 43: 211-247. Sunstein, Cass R., David Schkade, and Lisa Mi chelle Ellman. 2004. Ideological Voting on Federal Courts of Appeals: A Preliminary Investigation. Virginia Law Review 301354. Surrency, Erwin C. 1987. History of the Federal Courts. New York: Oceana Publications. Sutherland, E. H. 1949. White collar crime. New York: Dryden Press. Ulmer, Jeffery T. 2005. The Localized Uses of Federal Sentencing Guidelines in Four U.S. District Courts: Evidence of Processual Order. Symbolic Interaction 28: 255-279. Ulmer, Jeffery T. and Brian Johnson. 2004. Sent encing in Context: A Multilevel Analysis. Criminology 42: 137-177. Ulmer, Jeffery T. and John H. Kramer. 1998. The use and Transformation of Formal DecisionMaking Criteria: Sentencing Guidelines, Organizational Contexts, and Case Processing Strategies. Social Problems. 45: 248-267. US Sentencing Commission (USSC). 2007. Guidelines Manual. Washington, DC: Office of Publishing and Public Affairs. US National Archive and Record Administration (NARA). 2005. Constitution of the United States. http://www.archives.gov/nati onal-archives-experience/charters/ constitution_transcript.html. Weidner, Robert R., Richard S. Frase, and Jennifer S. Schultz. 2005. The Impact of Contextual Factors on the Decision to Imprison in Large Ur ban Jurisdictions: A Multilevel Analysis. Crime and Delinquency 51: 400-424. Yarnold, Barbara M. 1992. Politics and the Courts New York: Praeger.
134 Yeager, Peter C. 1991. The Limits of Law: The Public Regulation of Private Pollution New York: Cambridge University Press.
135 BIOGRAPHICAL SKETCH Jason Davis graduated from Hephzibah High School near Augusta, Georgia in 1991. He earned his bachelors degree (s ociology major) from Augusta St ate University in 1997. Upon graduation, he attended the Univer sity of Florida and earned a ma sters degree in sociology in 1999. Jason continued his graduate education at the University of Florida and pursued a doctorate in sociology with a major emphasis in criminology and a secondary concentration in organizations. Jason is currently employed as an instructor of criminology at the University of South Florida, St Petersburg, where he lives with his wife and son.