U.S. Department of Justice Office of Justice Programs Bureau of Justice Statistics Compendium of Federal Justice Statistics, 2002 September 2004, NCJ 205368 --------------------------------------------------------------- This file is text only without graphics and many of the tables. A Zip archive of the tables in this report in spreadsheet format (.wk1) and the full report including tables and graphics in .pdf format are available from: http://www.ojp.usdoj.gov/bjs/abstract/cfjs02.htm This report is one in a series. More recent editions may be available. To view a list of all in the series go to http://www.ojp.usdoj.gov/bjs/pubalp2.htm#cfjs --------------------------------------------------------------- Lawrence A. Greenfeld Director, BJS This Bureau of Justice Statistics Report was prepared by the Urban Institute under the supervision of Steven K. Smith and Mark Motivans of the Bureau of Justice Statistics (BJS). This report was prepared under BJS grant number 98-BJ-CX-K015. Principal staff at the Urban Institute were Laura Winterfield, William Adams, Avi Bhati, Kamala Mallik Kane, Barbara Parthasa rathy, Sam Wolf, Christine Arriola, and Yan Yuan. Layout and design were by Mick Schomm er. Tom Hester of BJS provided editorial review. This report is made possible through the cooperation of the following Federal agencies and their staffs: The United States Marshals Service (USMS), the Drug Enforcement Administration (DEA), the Administrative Office of the United States Courts (AOUSC), the Executive Office for U.S. Attorneys (EOUSA),and the Federal Bureau of Prisons (BOP). The staff who provided expert advice about the source records include: Joe Briggs (USMS); Steven Schlesinger and Catherine Whitaker(AOUSC); Barbara Tone, Siobhan Sperin, and Laurie Hamilton (EOUSA); Sue Allison and Nancy Miller (BOP); and Lou Reedt and Christine Kitchens (U.S. Sentencing Commission). The contents of this document do not necessarily reflect the views or policies of the Bureau of Justice Statistics or the U.S. Department of Justice. BJS authorizes any person to reproduce, publish, translate, or otherwise use all or any part of the material in this publication; citation to source, however, is appreciated. An electronic version of this report and the data underlying graphics may be found on the BJS Internet Home Page (http://www.ojp.usdoj.gov/bjs/). The BJS-sponsored Federal Justice Statistics Resource Center (FJSRC)Internet Home Page (http://fjsrc.urban.org)provides online access to the Federal Justice database. Users may download data from the Federal Justice database for independent analysis or use the online query system to quickly obtain customized statistics. To order additional copies of this report or CD-ROM's containing the Federal Justice database, call the BJS Clearing house at 1-800-851-3420. Contents Highlights, Arrest Prosecution Pretrial release Adjudication Sentencing Appeals Corrections Introduction, Organization of the Compendium Modifications in the 2002 Compendium Notes to reader System overview, Chapter 1. Arrests for Federal offenses, Arrests by offense categories Arrests by law enforcement agencies Arrests by Drug Enforcement Administration agents Arrests across demographic groups Tables, Chapter notes, Chapter 2. Prosecution, U.S. attorneys' decisions Suspects in matters received Suspects in matters concluded Tables, Chapter notes, Chapter 3. Pretrial release, Types of pretrial release Factors relating to release or detention Pretrial outcomes by offense categories Pretrial outcomes across demographic groups Length of pretrial detention Tables, Chapter notes, Chapter 4. Adjudication, Defendants in criminal cases filed Defendants in criminal cases terminated Case processing times Convictions by U.S. magistrates Characteristics of convicted defendants Tables, Chapter notes, Chapter 5. Sentencing, Offenders convicted and sentences imposed Average prison sentences imposed Relationship between sentence imposed and mode of conviction Characteristics of offenders sentenced to prison Tables, Chapter notes, Chapter 6. Appeals, Appeals filed Appeals terminated Tables, Chapter notes, Chapter 7. Corrections, Federal offenders under supervision Outcomes of offenders completing supervision Characteristics of offenders completing supervision Admissions, releases, and standing population of Federal prisoners Federal prisoners: First releases and time served Characteristics of Federal prisoners Tables, Chapter notes, Methodology, The Federal justice database Table construction and interpretation Offense classifications Source agencies for Compendium data tables Glossary, Tables Chapter 1. Arrests for Federal offenses, October 1, 2001 - September 30, 2002 Suspects arrested for Federal offenses, by offense Suspects arrested for Federal offenses and booked by U.S. Marshals Service, by Federal law enforcement agency Characteristics of Federal arrestees booked by U.S. Marshals Service Characteristics of suspects arrested by Drug Enforcement Administration agents, by type of drug Characteristics of suspects arrested by Drug Enforcement Administration agents, by weapon use at time of arrest Chapter 2. Prosecution, October 1, 2001 - September 30, 2002 Suspects in matters received by U.S. attorneys, by offense Disposition of suspects in matters concluded, by offense Suspects in matters concluded and declined, by investigating agency Basis for declination of prosecution by U.S. attorneys Disposition of matters declined for prosecution by U.S. attorneys, by offense Mean and median processing times from receipt to filing or declination, by offense Chapter 3. Pretrial release, October 1, 2001 - September 30, 2002 Outcomes of initial and detention hearings Type of pretrial release, by offense Form of pretrial detention, by offense Type of pretrial release, by defendant characteristics Form of pretrial detention, by defendant characteristics Pretrial detention hearing outcomes, by offense Pretrial detention hearing outcomes, by defendant characteristics Behavior of defendants released prior to trial, by offense Behavior of defendants released prior to trial, by type of release Behavior of defendants released prior to trial, by defendant characteristics Length of pretrial detention, by form of release or detention, and most serious offense charged Chapter 4. Adjudication, October 1, 2001 - September 30, 2002 Defendants in criminal cases commenced, by offense Disposition of criminal cases terminated, by offense Time from filing to disposition of criminal cases terminated Dispositions by U.S. magistrates Characteristics of convicted offenders Chapter 5. Sentencing, October 1, 2001 - September 30, 2002 Sentence types in criminal cases terminated, by offense Type and length of sentences imposed, by offense Sentences imposed on convicted offenders, by offense of conviction and method of disposition Convicted offenders sentenced to incarceration, by offense and offender characteristics Average incarceration sentence lengths imposed, by offense and offender characteristics Median incarceration sentence lengths imposed, by offense and offender characteristics Chapter 6. Appeals, October 1, 2001 - September 30, 2002 Criminal appeals filed, by type of criminal case and offense Criminal appeals filed and criminal appeals terminated, by offense Criminal appeals terminated, by type of criminal case and offense Disposition of criminal appeals terminated on the merits, by offense Criminal appeals cases terminated on the merits, by nature of offense Chapter 7. Corrections, October 1, 2001 - September 30, 2002 Federal offenders under supervision, by offense Characteristics of Federal offenders under supervision, by offense Outcomes of probation supervision, by offense Characteristics of offenders terminating probation supervision Outcomes of supervised release, by offense Characteristics of offenders terminating supervised release Outcomes of parole, by offense Characteristics of offenders terminating parole Admissions and releases of Federal prisoners, by offense Characteristics of Federal prison population, by major offense category Number of first releases from Federal prison, by release method and sentence length Average time to first release and percent of sentence served, for prisoners released by standard methods Mean time served to first release, by length of sentence imposed, offense, and type of case, for prisoners released by standard methods Percent of sentence served to first release, by length of sentence imposed, offense, and type of case, for prisoners released by standard methods Characteristics of first releases from prison, by offense, all releases Mean time served to first release from Federal prison for prisoners released by standard methods, by offense and offender characteristics Figures Highlights, October 1, 2001 - September 30, 2002 Average length of sentences imposed, by offense Admissions to Federal Bureau of Prisons, releases, and prisoners at yearend, by offense Average time to first release, standard releases, by offense System overview, October 1, 2001 - September 30, 2002 System overview, Federal criminal case processing Federal criminal case processing outcomes Chapter 1. Arrests for Federal offenses, October 1, 2001 - September 30, 2002 Type of drug involved in drug arrests, by race of arrestee Type of immigration arrests Suspects arrested by the DEA, by State of arrest Chapter 2. Prosecution, 21 October 1, 2001 - September 30, 2002 Suspects in matters concluded: Percentage of suspects that were prosecuted in U.S. district court Suspects in matters referred, by Federal judicial district Chapter 3. Pretrial release, October 1, 2001 - September 30, 2002 Percent of defendants released prior to case disposition, by offense category Detention hearings held, by offense category Pretrial release rates, by defendant characteristics Pretrial release violation and revocation rates, for cases terminated Chapter 4. Adjudication, October 1, 2001 - September 30, 2002 Disposition of cases terminated Average time from filing to disposition of cases terminated, by offense Chapter 5. Sentencing, October 1, 2001 - September 30, 2002 Rates of incarceration and probation for offenders convicted and sentenced in cases terminated Incarceration and probation sentence lengths (in months) of offenders convicted and sentenced in cases terminated Average prison sentences imposed on defendants convicted at trial or by guilty plea Chapter 6. Appeals, October 1, 2001 - September 30, 2002 Types of criminal appeal cases Appeals cases filed, by type of offense Disposition of criminal appeals terminated Chapter 7. Corrections, October 1, 2001 - September 30, 2002 Violation rates of offenders terminating probation, supervised release, or parole, by type of violation Violation rates of offenders completing probation, supervised release, or parole, by category of offense Rates of violation of conditions of supervision, by level of education Mean lengths of sentence imposed and time served(in months)for offenders released from prison for the first time Characteristics of first releasees Federal offenders under supervision, by Federal judicial district Highlights The number of suspects investigated by U.S. attorneys increased between 2001 and 2002, from 121,818 to 124,335. About three-quarters of those for which the investigation was concluded were prosecuted -- either before a U.S. district court judge (62%)or before a U.S. magistrate (11%) -- and 27% of those investigated were not prosecuted by U.S. attorneys. The number of defendants prosecuted in Federal courts increased slightly between 2001 and 2002, from 86,728 to 90,407. The number of offenders under Federal correctional supervision increased 77% between 1990 and 2002. At the end of fiscal year 2002, the number of offenders in Federal prison or on community supervision was 250,398 compared to 141,790 during 1990. At the end of fiscal year 2002, the number of Federal inmates serving a sentence of imprisonment was 143,031. The number under community supervision was 107,367. Over 70% of those under community supervision were on post-incarceration supervised release (73,229) or parole (3,561). Arrest During 2002, 124,074 suspects were arrested by Federal law enforcement agencies for violations of Federal law. Twenty-seven percent of those arrested and booked by the U.S. Marshals Service were for drug offenses, 21% for immigration offenses, 18% for supervision violations, 14% for property offenses, 7% for public-order offenses, 6% for weapon offenses, 4% for violent offenses, and 3% to secure and safeguard a material witness. About 70% of suspects booked by the U.S. Marshals Service were arrested by Department of Justice agencies, while Treasury Department agencies accounted for 12% of all arrests. Within the Department of Justice, the U.S. Marshals Service made 39% of the arrests; the Immigration and Naturalization Service, 33%; the Drug Enforcement Administration 14%, and the Federal Bureau of Investigation 13%. Prosecution During 2002 U.S. attorneys initiated criminal investigations involving 124,335 suspects, and they concluded their investigations of 124,081 suspects. Thirty-one percent of the suspects were investigated for drug, 22% for property, 19% for public order, 14% for immigration, 9% for weapon, and 5% for violent offenses. Of the suspects in criminal matters concluded, U.S. attorneys prosecuted 76,314 in U.S. district courts and 14,093 were disposed of before U.S. magistrates. During 2002, U.S. attorneys declined 27% of matters concluded. Suspects in criminal matters involving immigration or drug offenses were more likely to be prosecuted in a U.S. district court (85% and 77%, respectively) than were suspects involved in weapon (70%), violent (54%), property (54%), or public-order offenses (30%). Suspects involved in property offenses (such as fraud), violent offenses, or public-order offenses were more likely to be declined for prosecution (about 40%)than were suspects investigated for weapon (28%), drug (18%), or immigration (3%)offenses. Pretrial release Of 78,060 pretrial cases commenced in 2002, 37% of defendants were released after either an initial or detention hearing, while 62% were detained, and less than 1 % were dismissed. During 2002, 45% of the 71,572 defendants who terminated pretrial services were released at some time prior to their criminal trial. Defendants charged with property offenses or public-order offenses were more likely to be released prior to trial (79% and 69%, respectively)than were defendants charged with weapon (43%), drug (41%), violent (35%), or immigration (9%)offenses. The proportion of defendants released prior to their trial decreased from 62% during 1990 to 45% during 2002. Defendants having a prior criminal history of serious or violent crimes were less likely to be released than those without a prior criminal history; defendants with a greater number of prior convictions were less likely to be released than those with fewer prior convictions. About 25% of the defendants with a prior violent felony conviction were released before trial, while 61% of defendants with no prior convictions were released. Forty-six percent of defendants with one prior conviction were released, as compared to 37% of defendants having two to four prior convictions and about 28% of defendants having five or more prior convictions. Eighty-one percent of defendants released prior to trial completed their periods of release without violating the conditions of their release. Nineteen percent of defendants released violated the conditions of their release, and 8% of defendants had their release revoked. Defendants charged with weapon or drug offenses were more likely to commit at least one violation of their conditions of release (31% and 30%, respectively), while defendants charged with weapon or violent offenses were more likely to have their release revoked (13% and 12%,respectively) than were other defendants. Defendants released during 2002 were more likely to violate the conditions of their pretrial release than those released during 1990. During 2002, 20% of those released at some point prior to trial violated a condition of their release. During 1990, 12% violated their release conditions. Adjudication During 2002, 87,727 defendants were charged in Federal courts with a criminal offense, 87% of whom were charged with felonies. Of the defendants charged with felonies, 40% were prosecuted for drug, 21% for property, 17% for immigration, 11% for weapon, 6% for public- order, and 4% for violent offenses. The number of defendants charged with a felony immigration offense increased by 14% between 2001 and 2002, from 11,504 to 13,101. The number charged with a felony weapon offense increased by 25%, from 6,495 to 8,104. Criminal cases were concluded against 80,424 defendants during 2002,87% of whom had been charged with felonies. The proportion of defendants convicted in the Federal courts increased from 81% during 1990 to 89% during 2002. The proportion of convicted defendants who pleaded guilty increased from 87% during 1990 to 96% during 2002. About 92% of defendants charged with felonies were convicted. The conviction rate was similar for the major offense categories: 95% of defendants charged with immigration offenses, 92% of both drug and violent defendants, 91% of property defendants, 89% of weapon defendants, and 88% of public-order defendants. Sentencing Defendants convicted during 2002 were more likely to be sentenced to prison than those convicted during 1990. During 2002 about 75% of defendants were sentenced to prison compared to 60% of those sentenced during 1990. Ninety-three percent of felony violent offenders received prison terms, as did 92% of felony weapon offenders, 91% of felony drug offenders, 89% of felony immigration offenders, 66% of felony public-order offenders, and 59% of felony property offenders. The 53,682 offenders sentenced to prison received, on average, 57.1 months of imprisonment. Offenders sentenced for felony violent offenses, felony weapon offenses, and felony drug offenses received longer average prison terms (88.5, 83.9, and 76 months, respectively) than those convicted of felony property, immigration, public-order offenses (25, 27.9, and 38.5 months, respectively). While the proportion of defendants sentenced to prison is at an all- time high, average prison sentences have declined from the peak attained during 1992. During 1992 the average prison term imposed was 62.6 months; for violent felony offenders, the average term imposed was 94.8 months; for drug felony offenders, the average term was 84.1 months. Appeals Between 1994 and 2002, the number of appeals received by the U.S. Courts of Appeals remained relatively stable -- between about 9,000 and 11,500 annually. However, the proportion of criminal defendants appealing some aspect of their conviction decreased from 21% during 1994 to 16% during 2002. During 2002, 11,569 criminal appeals were filed, a 3% increase from FY2001 (11,281). Forty-nine percent of the appeals filed challenged both the conviction and sentence imposed. Only 4% of appeals were filed by the Government. Of the 11,695 appeals terminated during 2002, 75% (or 8,770) were terminated on the merits. In 83% of the appeals terminated on the merits, the district court ruling was affirmed, at least in part. Corrections Community supervision Between 1990 and 2002, the number of offenders on community supervision increased by 27%, from 84,801 during 1990 to 107,367 during 2002. While nearly equal proportions of offenders were serving terms of probation and post-incarceration supervision (parole or supervised release) during 1990, during 2002, over 71% were serving a term of post-incarceration supervision(68% supervised release and 3% parole) while 28% were on probation. Drug offenders comprised 12% of offenders on probation, 54% of offenders serving terms of supervised release, and 45% of offenders on parole. Property offenders comprised 38% of offenders on probation, 24% of offenders serving terms of supervised release, and 8% of offenders on parole. A total of 15,116 offenders terminated probation during 2002. Most of these offenders (80%) completed their terms of probation successfully. Twelve percent of probationers terminating supervision during 2002 committed technical violations; 7% committed new crimes. A total of 27,678 offenders completed terms of supervised release during 2002. Of these offenders, 62% successfully completed their terms without violating conditions of release; 23% committed technical violations; and 13% committed new crimes. A total of 1,817 offenders completed terms of parole during 2002. Of these offenders, 58% successfully completed their terms without violating conditions of release; 23% committed technical violations; and 13% committed new crimes. Prison Between 1990 and 2002, the number of inmates serving a sentence of imprisonment increased by 151%, from 56,989 during 1990 to 143,031 during 2002. During 2002, 50,440 prisoners were received by the Bureau of Prisons from U.S. district court commitments. An additional 17,437 prisoners were returned to Federal prison for violating conditions of probation, parole, or supervised release, or were admitted to Federal prison from elsewhere than a U.S. district court. Drug offenders -- who comprised 42% of persons admitted into Federal prison -- comprised the largest percentage of persons in prison (57%) at the end of 2002. During 2002, 44,339 prisoners were released for the first time from Federal prison after commitment by a U.S. district court. Of these, 39,568 were released by standard methods and 4,771 were released by extraordinary means (death, treaty transfer, sentence commutation, or drug treatment). An additional 16,493 prisoners were released from subsequent commitments to Federal prison. Average time served by Federal offenders increased from 24 months during 1994 to 32 months during 2002. The proportion of the sentence served increased from 65% during 1990 to 91% during 2002. Violent, weapon, and drug offenders were among those offenders who served the longest prison terms (59 months for violent offenders and 43 months each for weapon and drug offenders). Introduction This Bureau of Justice Statistics (BJS) report presents an overview of case processing in the Federal criminal justice system. The data presented are compiled from the BJS Federal Justice Statistics Program (FJSP) database. The FJSP database includes data provided by the U.S. Marshals Service, Drug Enforcement Administration, Administrative Office of the U.S. Courts, Executive Office for the U.S. Attorneys, Federal Bureau of Prisons, and U.S. Sentencing Commission. The Administrative Office of the U.S. Courts, in addition to providing data describing defendants in cases processed by the Federal judiciary, provides data describing defendants processed by the Federal pretrial services agencies and the Federal probation and supervision service. The BJS Federal Justice Statistics database is archived on CD-ROM, copies of which are available from the BJS Clearinghouse. The data can be downloaded from the Federal Justice Statistics Resource Center at . Each agency reports on those defendants it processed during a given year in an annual statistical report. These reports are often incomparable across agencies due to the varying methods the agencies use to report case processing activities. As reported by an interagency working group, headed by BJS, the differences in the case processing statistics are attributable, in part, to the differing needs and missions of the agencies. The working group found the following differences in reported statistics: * the universe of cases reported during a given period -- some agencies report on those case processing events that occurred during a particular period, whereas others report on those events recorded during a particular period; and * many of the commonly used case processing statistics -- suspect/defendant processed, offense committed, disposition, and sentence imposed -- are defined differently across agencies. BJS, through its FJSP, has recognized the incomparability of these annual statistical reports and has attempted to reconcile many of the differences identified by the working group. For instance, by combining databases from several years, BJS is able to report on those cases that actually occurred during the reporting period. Commonly used case processing statistics are made comparable across stages by applying uniform definitions to data obtained from each agency. Because definitions in the FJSP are consistent with those categories in BJS programs describing State defendants convicted, sentenced, or imprisoned, the comparison of Federal and State case processing statistics is facilitated. The 2002 Compendium, 17th in a series which includes 1984, 1985, 1986, 1988, 1989, 1990, 1992, 1993, 1994, 1995, 1996, 1997, 1998, 1999, 2000, and 2001 describes defendants processed at each stage of the Federal justice system -- arrest by Federal law enforcement agencies (chapter 1), investigation and prosecution by the U.S. attorneys (chapter 2), pretrial release or detention (chapter 3), adjudication in the U.S. district courts (chapter 4), sentencing (chapter 5), appeal of the conviction and/or sentence imposed (chapter 6), and corrections (chapter 7) -- for the 12-month period ending September 30, 2002 (the Federal fiscal year). Prior to 1994, the Compendium was reported on a calendar-year basis. The tables presented report events that occurred during the Federal fiscal year -- October 1, 2001-- September 30, 2002. Generally, the tables include both individual and organizational defendants. Organizational defendants are not included in tables describing pretrial release and detention, tables showing defendants sentenced to incarceration, or tables describing offenders under post-conviction community supervision. Juvenile offenders charged as adults are included in the reported statistics. Felony and misdemeanor distinctions are provided where possible(see "Offense classifications" in Methodology). Organization of the Compendium Each chapter of the Compendium describes a major stage in the pro- cessing of criminal suspects and defendants. Each chapter contains Chapter notes that describe the universes of data used in the tables and information relevant to the interpretation of individual tables. The Compendium contains the following: Chapter 1 describes arrests made by Federal law enforcement agencies for violations of Federal law, including the characteristics of arrestees. Chapter 2 describes decisions made by Federal prosecutors in screening criminal matters and the characteristics of defendants in cases prosecuted or declined for prosecution. Chapter 3 describes the pretrial release and detention practices of the Federal judiciary, including the characteristics of defendants detained or released pending trial. Chapter 4 describes actions by the Federal judiciary in adjudicating defendants in cases filed by the U.S. attorneys, including the offense charged and characteristics of defendants convicted. Chapter 5 describes the sentences imposed by the Federal judiciary on convicted defendants, including the characteristics of defendants sentenced. Chapter 6 describes appeals of criminal convictions and sentences imposed in the Federal courts, including the original offense charged. Chapter 7 describes defendants under Federal correctional supervision -- probation, parole, and supervised release -- including the outcome of the supervision (successful completion or violations), admissions to and releases from Federal prison, and time served by Federal inmates. The Methodology section describes the procedures followed in analyzing data and developing tables. The Glossary contains definitions for terms used in the Compendium. Since many terms used in the text and tables have specialized meanings (either because they refer to Federal law or because of reporting procedures by the Federal agencies supplying the data), readers are encouraged to check the glossary for exact definitions of tabulated data. Modifications in the 2002 Compendium The FY2002 data marked the introduction of two new detailed offense categories: "Wildlife offenses" and "Environmental offenses". According to the reporting practices of this report, these two detailed offenses are grouped under the major offense category, "Public-order offenses, Other" in the tables. The "Wildlife offenses" category supersedes the "Migratory birds" category that appeared in prior Compendia and includes other wildlife-related offenses as well. Notes to reader The tables in the Compendium were constructed to permit valid comparisons within each table and to allow the reader to compare percentages (but not raw totals) across tables. It should be understood, however, that the total number of subjects/defendants shown in a particular table may not equal the number of subjects/ defendants involved in a particular stage of processing, since some records could not be linked and some data sources did not include information on particular data elements classified in a particular table. Data notes indicate the exact universe for individual tables. The Compendium is a statistical presentation of Federal criminal justice information with limited analyses of trends or explanatory factors underlying the statistics. Analyses of Federal justice statistics may be found in special reports and other publications, some of which are cited in the Compendium. Assessment of changing patterns in the Compendium tabulations may depend on detailed examination of subcategories not shown in the tabulations or may require other sources of information, such as knowledge of legislation or Federal agency procedures. Chapter 1 Arrests for Federal offenses Discussion Tables October 1, 2001 - September 30, 2002 Suspects arrested for Federal offenses and booked by USMS, by offense\ Suspects arrested for Federal offenses and booked by USMS, by Federal law enforcement agency Characteristics of Federal arrestees booked by USMS. . . . . . . Characteristics of suspects arrested by Drug Enforcement Administration agents, by type of drug Characteristics of suspects arrested by Drug Enforcement Administration agents, by weapons involved at time of arrest Chapter notes A recent Bureau of Justice Statistics survey found that, in 2002, 67 Federal agencies employed over 93,000 full-time officers with the authority to carry firearms and make arrests.***Footnote 1: Brian A. Reaves and Lynn M. Bauer, Federal Law Enforcement Officers, 2002, Bureaus of Justice Statistics, Washington, DC (NCJ-199995). This report preceded realignments specified in the Homeland Security Act of 2002.*** Three-fifths of these officers were employed by four agencies: the Immigration and Naturalization Service (INS), Bureau of Prisons (BOP), Customs Service, and the Federal Bureau of Investigation (FBI). Certain military officers, Federal air marshals, and Central Intelligence Agency Security Protective Service officers are also authorized to make arrests, but were not included in the survey. In 2002 the U.S. Marshals Service employed over 2,600 officers with arrest and firearm authority***Footnote 1:Brian A. Reaves and Lynn M. Bauer, Federal Law Enforcement Officers, 2002, Bureaus of Justice Statistics, Washington, DC(NCJ-199995). This report preceded realignments specified in the Homeland Security Act of 2002.*** All suspects arrested by Federal agencies are transferred to the custody of the U.S. Marshals Service for processing, transportation, and detention.***Footnote 2: The Federal agency making the arrest may be different from the Federal agency initiating the investigation involving the arrestee. Statistics describing agencies initiating investigations are shown in table 2.3 of chapter 2, "Prosecution."*** During 2002, the U.S. Marshals Service received 124,074 suspects for processing from the Federal law enforcement agencies, including over 34,000 suspects arrested by deputy U.S. marshals, and over 12,000 arrested by the Drug Enforcement Administration (DEA). Number arrested and booked by offense categories During 2002, 124,074 suspects were arrested by law enforcement agencies for violations of Federal law and subsequently booked by the U.S. Marshals Service. Of those suspects arrested and booked, 27% were for drug offenses, 21% for immigration offenses, 18% for supervision violations, 14% for property offenses, 7% for public-order offenses, 6% for weapon offenses, 4% for violent offenses, and 3% to secure and safeguard a material witness. Examining drug arrests further, the greatest portion of drug arrests (37%) involved cocaine (crack or powder); 22%, marijuana; 13%, amphetamine (or methamphetamine); 6%, heroin; and the remainder involved other drugs or drug paraphernalia. Examining immigration offenses further, most (84%) of arrests for immigration offenses were for illegal entry into the United States; 12% involved alien smuggling; 2% involved false claims of citizenship by entrants to the United States; and 2% involved other immigration violations. Number arrested and booked by law enforcement agencies During 2002, 70% of all arrests for Federal offenses were made by components of the Department of Justice and Department of Treasury components accounted for another 12% of all arrests. In addition, 5% of arrests were made by State and local agencies, and 7% of persons were arrested after voluntarily reporting to the U.S. Marshals following a summons. Within the Department of Justice, 39% of arrests were made by the U.S. Marshals Service, 33% were made by the Immigration and Naturalization Service, 14% were made by the Drug Enforcement Administration, 13% were made by the Federal Bureau of Investigation, and less than 1% were made by other Department of Justice components. Within the Treasury Department, 52% of arrests were made by the U.S. Customs Service, 32% were made by the Bureau of Alcohol, Tobacco, and Firearms, 13% were made by the Secret Service, and 4% were made by the Internal Revenue Service. Suspects arrested and booked across demographic groups Most (86%) of the suspects arrested by Federal law enforcement agencies during 2002 were male. Over two-thirds (70%) were white and about one-quarter (26%) were black. Individuals between the ages of 21 and 40 years comprised 70% of all those arrested. More than one-third (37%) of those arrested were identified as noncitizens. About two-thirds (68%) of the suspects arrested for drug offenses were white, and most of the remaining third (30%) were black. The racial distribution of suspects differed by the type of drug involved. Whites constituted the majority of suspects arrested for most drug types, including amphetamines (93%), marijuana (89%), heroin (63%), and other drugs (66%). The exception was cocaine-related drug arrests, with blacks comprising 58% of suspects and whites comprising 41%. -------------------------------------------------------------- Arrests reported by DEA agents (tables 1.4-1.5) During 2002, agents of the Drug Enforcement Administration arrested 29,145 suspects. Not all of these suspects were booked for violations of Federal law by the U.S. marshals: some were referred to U.S. attorneys for a prosecutorial decision while others were referred to a State prosecutor (see Chapter notes). Suspects arrested by DEA agents were predominantly white (70%), non-Hispanic (58%), male (84%), and U.S. citizens (74%). The greatest proportion of arrestees were between 21 and 30 years old (44%). Cocaine was the predominant drug involved in arrests made by DEA agents, with cocaine powder comprising 25% of all arrests and crack cocaine, 15%. Twenty-two percent of DEA arrests involved methamphetamines; 19% involved marijuana; 8% involved opiates; and 11% involved other drugs or non-drug offenses. The majority of suspects arrested for each type of drug were white, with the exception of crack cocaine, for which the majority of suspects were black. Five percent of all suspects were armed at the time of their arrest. Subjects armed at time of arrest differed across demographic groups. Males were twice as likely to be armed at arrest compared to females (5% versus 3%). Blacks (5%) were more likely than whites (4%) to be armed at arrest. Non-Hispanics (5%) were more likely than Hispanics (4%) to be armed at arrest, and U.S. Citizens (5%) were more likely to be armed at arrest than noncitizens (4%). If armed at arrest, suspects from all demographic subgroups were more likely to be armed with a handgun than any other weapon. Of the 1,351 suspects who were armed, 87% were armed with a handgun, 22% were armed with a shotgun or rifle, and 9% were armed with another weapon. ---------------------------------------------------------------- Chapter notes 1) Tables 1.1-1.3 were derived from the U.S. Marshals Prisoner Tracking System. Only records of arrests made during October 1, 2001, through September 30, 2002, were selected. Information on individual cases, offenses, and detainees were used to construct the tables. 2) Tables 1.4-1.5 were derived from the DEA Defendants Statistical System. Only records of arrests made during October 1, 2001, through September 30, 2002, were selected. Counts of DEA arrests from tables 1.4-1.5 will be higher than those reported as DEA arrests by the U.S. Marshals because some suspects arrested by DEA agents are transferred to State or local jurisdiction and are, therefore, never booked by the U.S. Marshals Service. 3) The arrest of offense classification used by the U.S. Marshals were aggregated into the categories shown in the tables. These categories are similar, but may not be directly comparable, to categories used in other tables of this Compendium. Chapter 2 Prosecution Federal criminal cases may be brought by the U.S. Attorney's Office, by the Criminal Division of the U.S. Department of Justice, or by other authorized agencies such as the U.S. Marshals Service. The 93 U.S. attorneys serve as the chief Federal law enforcement officers within their respective districts. Investigations are most commonly referred to a U.S. attorney by a Federal investigative agency (primarily the Federal Bureau of Investigation; the Drug Enforcement Administration; the Immigration and Naturalization Service; the Customs Bureau; the Bureau of Alcohol, Tobacco and Firearms; and the Secret Service) or by a State or local investigative agency. Investigations may also be initiated -- and cases brought directly -- by U.S. attorneys or by the Criminal Division of the U.S. Department of Justice. This chapter reports on suspects that were at least partially investigated by U.S. Attorneys. U.S. attorneys' decisions After criminal investigations are initiated and criminal suspects are referred to them, U.S. attorneys may file charges against defendants in a U.S. district court, or they may file cases before U.S. magistrates, who have the authority to adjudicate misdemeanor offenses (18 U.S.C. sec. 3401); because of the relatively less serious nature of these cases, and because they are handled by magistrates, the U.S. attorneys count these cases as criminal matters disposed by U.S. magistrates. U.S. attorneys may also decline to file charges for reasons such as weak or insufficient evidence, minimal Federal interest, lack of resources, or lack of criminal intent. Matters that are declined may be referred to another authority for prosecution or be settled through alternative resolution procedures. The decision to prosecute a suspect in a criminal matter depends upon many factors, including the Attorney General's priorities, U.S. attorney priorities and resources, laws governing each type of offense, and the strength of evidence in each case. During 2002, 62% of the suspects investigated by U.S. attorneys were prosecuted in U.S. district court, and U.S. magistrates handled an additional 11%. Suspects were prosecuted at different rates depending on the offense involved. Suspects in immigration matters were the most likely to be prosecuted; 85% had charges filed against them in U.S. district court and an additional 12% had charges filed before a U.S. magistrate. The rate of prosecution was similar for drug offenses, with 77% of suspects having charges filed in district court and 5% having charges filed in magistrate court. About two-thirds of suspects in weapon and violent offense matters were prosecuted, with most (70% and 54%, respectively) being prosecuted in district court. Public-order and property offenses were the least likely to be prosecuted. Thirty percent of public-order suspects were prosecuted in U.S. district court, and 33% were concluded by magistrates. Fifty-four percent of property offense suspects were prosecuted in district court. Suspects in matters received During 2002 there were 124,335 suspects in new matters received for investigation by U.S. attorneys. Of these suspects, about one-third (31%) were investigated for drug offenses; about one-fifth (22%) for property offenses; 19% for public-order offenses; 14% for immigration offenses; 9% for weapon offenses; and 5% for violent offenses. Investigating and prosecuting terrorism threats and related persons and enterprises were the highest priority for 2002, although violent crimes -- particularly those involving firearms -- as well as identifying, disrupting, and dismantling drug trafficking organizations continued to be high priorities. Other areas of special emphasis during 2002 included corporate fraud and civil rights prosecution -- especially hate crimes arising out of terrorist attacks on the United States.***Footnote 1: Statistical Report, United States Attorneys' Offices, Fiscal Year 2002. U.S. Department of Justice, Executive Office for United States Attorneys, Washington, DC.*** Disposition of suspects in matters concluded Upon receiving a matter, a U.S. attorney may either decline it for prosecution immediately or investigate further. Upon conclusion of the investigation, U.S. attorneys may file criminal charges in a U.S. district court, refer the matter to a U.S. magistrate, or decline the case for Federal prosecution. Of the 124,081 suspects in criminal matters concluded during 2002, 62% were prosecuted in U.S. district court, 11% were referred to U.S. magistrates, and 27% were declined for Federal prosecution.***Footnote 2: The number of suspects in criminal matters concluded is not comparable to the 1993 and prior compendia. See Chapter notes, item 2, p. 33, for details.*** Suspects prosecuted in U.S. district court -- U.S. attorneys prosecuted 76,314 suspects in U.S. district courts during 2002. The likelihood of prosecution in U.S. district court varied widely across offense categories. Suspects involved in immigration and drug offenses were more likely to be prosecuted than suspects in other types of offenses. U.S. attorneys filed charges in district court against most(85%)of the suspects involved in immigration offenses and most (77%) of the suspects involved in drug offenses; 70% of those in weapon offenses; 54% of those in violent offenses; 54% of those in property offenses; and 30% of those in public-order offenses. Differences in rates of prosecution within major categories were substantial. For example, among violent offenses, 78% of robbery suspects were prosecuted, while 23% of murder suspects, 46% of assault suspects, 37% of kidnaping suspects, and approximately 42% of sexual abuse suspects were prosecuted in U.S. district courts. Among suspects investigated for property offenses, those in counterfeiting matters had the highest rate of prosecution (64%), while those investigated for arson and explosives had one of the lowest rates of prosecution (45%). Among suspects investigated for public-order offenses, those in antitrust (60%), tax law (56%), and customs law matters (48%) were much more likely to be prosecuted than suspects in escape (29%) and civil rights (4%) matters. Suspects in matters concluded by U.S. magistrates -- During 2002 U.S. attorneys referred 11% of criminal matters investigated to U.S. magistrates. Cases concluded by U.S. magistrates are, by statute, misdemeanors.***Footnote 3: At the time that U.S. attorneys receive them, matters may not be classifiable as felonies or misdemeanors. If a U.S. magistrate disposes of a case -- or what the U.S. attorneys call a "matter" -- its offense level has been determined, by statute, to be a misdemeanor.*** Public-order offenses made up the majority (54%) of matters concluded by U.S. magistrates, including conspiracy, aiding and abetting, and jurisdictional offenses (33%) and escape violations (9%). Immigration (14%), drug (14%), and property offenses (13%) were also common. Violent and weapon offenses comprised the smallest portion of the matters disposed by magistrates (2% and 1% respectively). For most offenses, the likelihood that a U.S. magistrate concluded the matter was relatively low. Offenses having comparably high rates of disposal by magistrates included traffic offenses (98%); liquor offenses (81%); conspiracy, aiding and abetting and jurisdictional offenses (84%); other property violations (70%); and escape violations (49%). Suspects in matters concluded and declined, by investigating agency -- During 2002, 55% of suspects in matters concluded were investigated by components of the Department of Justice; the Treasury Department investigated 20% of suspects in matters concluded. In addition, independent Federal agencies and State/local agencies investigated 10% of suspects in matters concluded. Among the investigating agencies, the Immigration and Naturalization Service (4%), the National Park Service (4%), and the U.S. Army (5%) had among the lowest declination rates, while the Health and Human Service's Office of Inspector General (62%), the Small Business Administration (66%), and the Environmental Protection Agency (60%) were among those with the highest declination rates. The declination rate for matters investigated by components of the Department of Justice (28%) was about the same as the overall declination rate. Suspects in matters declined for Federal prosecution Twenty-seven percent of the matters concluded by U.S. attorneys in 2002 were declined for Federal prosecution. Suspects in property and public-order matters were the most likely to have their matters declined (40% and 38%, respectively). Forty-one percent of matters involving violent offenses were declined, as were 28% of weapon and 18% of drug matters. Matters involving an immigration offense were the least likely to be declined (3%). The decision to decline Federal prosecution is based on a number of factors, including the lack of a prosecutable offense, an alternative resolution, or case- and suspect-related reasons. Of the 33,674 declinations during 2002, 27% occurred because of case-related reasons, mostly due to weak evidence (23%); 20% occurred because there was no crime or criminal intent was lacking; and 26% occurred for other reasons, such as agency requests (9%) and lack of resources (6%). Not all suspects whose matters were declined by U.S. attorneys avoided prosecution. Twenty-one percent of the suspects in matters declined were referred to another authority for prosecution. An additional 3% were settled through alternative resolution procedures. Suspects involved in drug offenses constituted the largest number who were referred by U.S. attorneys to other authorities for prosecution or resolution. Thirty-one percent of the drug suspects whose matters were declined prosecution by U.S. attorneys were referred for other prosecution or resolution, as were 26% of the immigration, 22% of public-order, 22% of violent, 30% of weapon, and 20% of property offense suspects. Within major offense categories, the likelihood that suspects in matters declined for prosecution were referred to other authorities for prosecution or alternative resolution varied among specific offenses. Within property offenses, referral occurred with 37% of persons involved in counterfeiting, compared to 18% of suspects in fraud cases. Within public-order offenses, 62% of escape suspects were referred to other authorities or had their case settled through alternative procedures, compared to 6% of suspects in civil rights cases. Processing times -- The average number of months from receipt of a matter to its conclusion by a U.S. attorney as a case filing or declination or its disposal by a U.S. magistrate was 9.2 months (a median of 1.8 months). However, the time to process matters varied with the outcome of the matter. U.S. attorneys took an average of 4.2 months to file charges in U.S. district court, but over half of the cases were filed within 1 month. Declinations took longer, with an average duration of almost 21 months, though half of the declinations occurred within 15 months. The processing time for matters disposed by U.S. magistrates also differed by the outcome. Matters resulting in a conviction took an average of 3.6 months to conclude, with half of the cases concluding within 1.9 months. Matters that did not result in a conviction took longer to conclude -- an average of 13.6 months, with half concluding within 2.8 months. Processing times also varied with the type of offense involved. Suspects involved in immigration offenses were processed most quickly, with an average case processing time of 1.8 months and a median time of less than 1 month. Matters involving weapon, drug, and violent offenses took longer to process (5.1 to 7.1 months, on average, and median times from 0.9 to 2.5 months), while public-order and property offense matters took the longest to conclude, with average times of 12.7 and 15 months, and median times of 4.1 and 8.1 months, respectively. Chapter notes 1) Tables 2.1-2.6 were created from the Central System data files of the EOUSA. For table 2.1, only records which showed a matter received during 2002 were selected. For tables 2.2-2.6, only records which show a matter terminated by reason of declination, disposition by a U.S. magistrate, or filed as a case in U.S. district court from October 1, 2001, through September 30, 2002, were selected. In all of these tables, matters "declined immediately" -- those cases in which the U.S. attorney spent less than 1 hour investigating -- were excluded. 2) The number of suspects in matters investigated in table 2.1, and the number of suspects in matters concluded in tables 2.2 and 2.6 are not directly comparable to counts in the 1993 and prior compendia. In those years, these statistics included appellants in appeals filed or appeals terminated. In table 2.1 of this Compendium and its corresponding tables in 1994 and subsequent compendia, the number of suspects in matters received is limited to suspects in criminal matters that were filed as cases in U.S. district courts, handled by U.S. magistrates, or declined for prosecution. In the 1993 and prior compendia, the number of suspects in criminal matters received included criminal appeals. To obtain a number of suspects comparable to the statistic reported in the 1993 and prior compendia, add to the data in table 2.1 of this Compendium, the number of appellants in appeals that were handled by U.S. Attorneys. In tables 2.2 and 2.6 of this Compendium and their corresponding tables in the 1994 and subsequent compendia, the number of suspects in matters concluded also is limited to those suspects whose matters were concluded by a case filing in a U.S. district court, by declination, or by disposition by a U.S. magistrate. In the 1993 and prior compendia, this number included appellants in criminal appeals terminated. To obtain a number that is comparable to the number reported in the 1993 and prior compendia, the number of appellants in appeals terminated and handled by U.S. attorneys needs to be added to the totals in table 2.2. 3) Offenses in the Central System data files are classified by the title and section of the United States Code for the most serious offense investigated, as determined by the assistant U.S. attorney responsible for the matter. For tables in this Compendium, these citations were translated into the corresponding four-digit offense codes used by the Administrative Office of the United States Courts(AOUSC). These four- digit codes were then aggregated into the categories shown in the tables. U.S. Code citations often do not permit detailed classification of drug offenses by the type of criminal activity involved. A person charged with conspiracy is classified under the substantive offense alleged (for example, conspiracy to defraud is classified as fraud) unless type of conspiracy cannot be determined from the U.S. Code citation. Unclassified conspiracies are included among "conspiracy, aiding and abetting, traffic, and jurisdictional offenses." The most serious offense investigated is based on the criminal lead charge as determined by the assistant U.S. attorney responsible for the criminal matter. 4) In the tables in this chapter, the "all offenses" row or "total declinations" row includes records whose offense category (basis for declination)is missing or indeterminable. The percentage distributions in these tables, however, are based on the number of observations with nonmissing offenses (basis for declination) values. The number of missing values are reported either on a separate line or in a footnote. 5) Additional data for "lack of resources" as a reason for declination were discovered in 1999; these cases previously were coded as "unknown or indeterminate" in 1998 and prior years. Therefore, in table 2.4, the apparent increase in "lack of resources" as a declination reason (between 1998 and future years) is due in large part to this additional information and reclassification of declination reasons from "unknown and indeterminate." Chapter 3 Pretrial release The Bail Reform Act of 1984 (18 U.S.C. sec. 3141) sets the terms for release and detention of defendants facing charges in Federal courts. At the time defendants first appear before a judicial officer, they may be (1) released on personal recognizance or unsecured bond; (2) released subject to conditions imposed by the court, possibly including traditional bail; (3) temporarily detained without bail (limited to 10 working days) to permit deportation, exclusion, or the revocation of previously granted conditional release; or (4) detained pending the outcome of a detention hearing.***18 U.S.C. sec. 3142(e) (1984)*** According to the statute, a defendant must be brought before a judicial officer without "unnecessary delay" upon arrest for an "initial appearance". The judicial officer, who may be a judge, but is generally a U.S. magistrate, determines whether the defendant will be released or detained prior to trial. If a hearing is required, the release decision may be delayed no longer than 3 days (or 5 days if requested by the defendant). The recommendation to detain the defendant is made by the U.S. attorney and/or the pretrial services officer. If the court concurs with the recommendation to detain the defendant, a pretrial detention hearing is held to show cause why the defendant should be detained pending adjudication of the charges. According to the Bail Reform Act of 1984, preventive detention is applicable in instances where the defendant was charged with (1) a crime of violence, (2) an offense with a statutory maximum sentence of life imprisonment or death, (3) a drug offense with a statutory maximum sentence of 10 years or more imprisonment, or (4) any felony offense if the defendant had been convicted on two or more occasions of an offense described above or a similar State-level offense. A detention recommendation may be made if there is a serious risk that the defendant would flee in an attempt to avoid prosecution or would obstruct justice or intimidate witnesses or jurors. Types of pretrial release Defendants may be released without financial conditions according to the following: Personal recognizance -- defendant is released subject to no financial or other conditions. Unsecured bond -- no money is required to be posted before release, but defendant is liable for full bail amount if he or she fails to appear. Conditional release -- any combination of restrictions that are deemed necessary to guarantee the defendant's appearance at trial or the safety of the community. Non-financial conditions commonly place restrictions on the defendant's movements, associations, and/or actions. They may also involve employment or treatment for medical, psychological, or substance abuse conditions. Defendants may also be released on financial conditions. Financial conditions include: (1) deposit bond (the defendant is required to post a percentage of the total bail amount, usually 10%),(2)surety bond (the defendant is released subject to guarantees by a third person that the full amount will be paid), or (3) collateral bond (collateral equal to the full bail amount required to be posted by the defendant before release). Financial conditions may occur in combination with nonfinancial conditions. Factors relating to release or detention In deciding whether to release a defendant and in setting release conditions, the court is directed to consider the nature and circumstances of the offense charges, the weight of evidence against the defendant, the defendant's character, physical and mental condition, family ties, employment, financial resources, length of residence in the community, community ties, past conduct, history relating to drug or alcohol abuse, criminal history, and record concerning appearance at court proceedings. These factors are considered at an initial hearing where the defendant may be released, detained, held for a detention hearing, or the case may be continued or dismissed. At the detention hearing, a defendant may be released, given preventive detention, temporarily detained, detained with no bail set or made, or the case may be continued, dismissed, or transferred. Of the 78,060 defendants whose cases commenced in 2002, 37% (or 28,993) were released after either an intitial hearing or a detention hearing, and 62% of defendants were detained. Of the 71,572 defendants who terminated pretrial services during 2002, 45% (32,140) were released for some period of time prior to trial. Of those released defendants, 81% completed pretrial services without violating the conditions of their release. In 2002, 73% (52,515) of defendants terminating pretrial services were detained for some period of time prior to the disposition of their case. This includes those who were detained prior to having a preliminary hearing with the magistrate, at which point they could have been either released or ordered to have a detention hearing. Of the defendants detained, 54% were ordered detained by the court until case disposition. Pretrial detention hearings were held for 38,764 defendants. Of these, 28,301, or 73%, were ordered detained. Of the 32,140 defendants released prior to trial, 20% violated a condition of their release. The majority (92%) of these violations were technical violations of the bail conditions. Defendants released on deposit bond were more likely (28%) than other defendants to incur some violation of the conditions of their release. Conversely, defendants given conditional release violated their release at the lowest rates (less than 1%). Eight percent of all released defendants had their release revoked. Those charged with violent offenses were detained longer on average (112 days) than defendants charged with other offenses (not shown in a table). Defendants who were never released and were detained by the court were detained longer than were defendants detained for financial conditions (111 days) Outcomes for pretrial cases commenced Releases and detentions after the first 2 hearings -- Of the 78,060 cases commenced in 2002, 37% were released after either an initial hearing or a detention hearing, 62% were detained, and less than 1% were dismissed. Seventy-six percent of defendants receiving pretrial release were released after an initial hearing; 73% of detained defendants were detained after a detention hearing. After these two hearings, less than 1% of incoming defendants (279) were awaiting a release decision (not shown in a table). Of those charged with property offenses, 73% were released, as were 63% of those charged with public-order offenses, 37% of weapon offense defendants, 31% of drug offense defendants, and 28% of violent offense defendants. Only 5% of those charged with immigration offenses were released after these two hearings. Outcomes for pretrial cases terminated Releases -- Of cases terminated in 2002, 45% were released. The percent released varied greatly across and within offense categories. Seventy-nine percent of property offenders were released, compared with only 9% of immigration offenders. Among violent offenders, 22% of those charged with robbery were released compared to 55% of those charged with sexual abuse and 62% of those charged with assault. Among public-order defendants, 11% of those charged with an escape offense were released compared to 96% of those charged with tax law violations, 92% of traffic defendants, and 69% of nonviolent sex offense defendants. Forty-three percent of those charged with weapon offenses were released, while only 9% of immigration defendants were released. Forty-one percent of all drug defendants were released. Drug trafficking defendants were less likely to be released than those charged with nontrafficking offenses (40% compared to 56%). How defendants were released prior to trial varied among offense categories. Within each major offense category except for violent and immigration offenses, the highest percentage of defendants released was through unsecured bond. Thirty-five percent of violent defendants were released of whom 34% were released on unsecured bond and 48% on personal recognizance -- compared to 79% of all property defendants released, of whom 55% were released on unsecured bond and 33% on personal recognizance. Detentions -- Among the major offense categories, defendants charged with immigration, drug, or violent offenses were the most likely (98, 84% and 82% respectively) to be detained for some period of time. However, violent defendants were more likely than immigration or drug defendants to be detained by the court (64% versus 59% and 53%). Of those charged with weapons offenses, 78% were detained, with 57% being detained by the court. Public-order and property and defendants were less likely to be detained than were other defendants. During 2002, 44% of public-order defendants were detained, with 47% of these detained by the court. Forty percent of defendants charged with property offenses were detained, with 42% of these detained by the court. Seventy percent of defendants charged with violent offenses had a detention hearing. This rate varied by offense category: 65% of drug or immigration defendants, 62% of weapons defendants, 31% of those charged with public-order offenses, and 29% of those charged with property offenses were ordered to have a detention hearing. Once a defendant was ordered to have a detention hearing, the chances of being detained were less variable across offense categories. The highest rate of defendants ordered detained were those charged with immigration offenses (88%). Of the 70% of defendants charged with violent offenses and given a pretrial detention hearing, 75% were ordered detained. In 2002, 72% of those charged with weapon offenses, 69% of drug offenses, 65% of public-order offenses, and 59% of property offenses were ordered detained following a pretrial hearing. Violations -- Of released defendants terminating pretrial services during 2002, those charged with weapon or drug offenses were more likely than other defendants (31% and 30% compared with 20% overall) to incur at least one violation during the release period. Among other major offense categories, those charged with public-order offenses were least likely (8%)than others to violate conditions of their release. Those charged with weapon, violent, or drug offenses were more likely than others to have their release revoked (13% of all released weapon defendants, 12% of released violent defendants, and 11% of drug defendants compared with 8% overall). Five percent of released property defendants and released immigration defendants were revoked. Of the 32,140 defendants released, those on deposit bond were more likely (28%) than other defendants (20% overall) to incur a violation of the conditions of their release. Defendants with the lowest rate of release violations were those given conditional release (1%). This group was also the least likely to have a release revocation. Only 0.2% of defendants on conditional release who committed a violation were revoked. Pretrial outcomes across demographic groups Releases -- Females were more likely (72%) than males (40%) to be released during 2002. Males were more likely to have a financial condition imposed on them than females (21% versus 13%). Sixty-two percent of all defendants identified as non-Hispanic were released during 2002 compared to 21% of Hispanics. Non-Hispanic releasees were twice as likely as Hispanics to be released on personal recognizance. Sixty-four percent of Native Americans and 59% of Asian defendants were released compared with 50% of black defendants and 42% of white defendants. The higher the education level of the defendant, the greater the probability of release. Thirty-nine percent of defendants with less than a high school education were released in 2002, while 79% of those who had completed college were released. Releasees with a college degree, when compared to those with less education, were less likely to receive financial conditions (16% versus 26%) and more likely to be released on personal recognizance (30% versus 25%) than those who had less education. Defendants with a history of drug abuse were released less frequently than those defendants with no known drug history (50% versus 59%). However, among all released defendants, drug abuse history does not appear to have had a major effect on the form of pretrial release. Detentions -- Males were more likely (78% versus 52%) than females to be detained. Of all detainees, 75% of the men and 56% of the women were detained following a detention hearing. Hispanics had a substantially higher probability of being detained (93%) compared with non-Hispanics (60%). Almost all of those charged with an immigration offense (98%) -- of which non-citizen Hispanics comprise the majority -- were detained. Younger defendants were more likely than older ones to be detained. Seventy-five percent of defendants between ages 16 and 18 and 74% of defendants ages 19-20 were detained, compared to 60% of those over 40. However, among all defendants detained, the chances of being detained following a detention hearing were similar across all age groups. Not only were male defendants more likely than female defendants to have a detention hearing (58% versus 36%), they were also more likely (75% versus 56%) to be ordered detained as a result of the hearing. The likelihood of having a detention hearing increased with the number of prior convictions and severity of the defendant's criminal history. Furthermore, 58% of those with a history of drug abuse had detention hearings as compared to 44% of those without such a past. However, those with a history of drug abuse were as likely to be detained after a hearing. Violations -- Released males were more likely than females to incur a violation during the pretrial release period (21% versus 16%) and were slightly more likely to have their release revoked. Native Americans were more likely to incur at least one violation than blacks, whites, or Asians (29% versus 22% for blacks, 19% for whites, and 13% for Asians). Native American defendants also were more likely to have their release revoked (17% versus 8% for blacks, 7% for whites, and 5% for Asians). Defendants without a high school diploma were more likely to have at least one violation than defendants with college degrees (28% versus 8%). Defendants with no prior criminal history were 2 times less likely to have a pretrial release violation than those with a prior conviction. Released defendants with a known drug history were almost 3 times more likely to incur a violation than those with no known drug history and to have their release revoked (15% versus 4%). Length of pretrial detention Among defendants detained for any time prior to case disposition, those charged with violent offenses were detained the longest (an average of 112 days) compared to drug (83 days), weapon (81 days), immigration (64 days), public-order (60 days), or property (46 days), defendants (not shown in a table). Of defendants released, those released conditionally spent the greatest average number of days detained (61 days) compared to those released on personal recognizance (43 days), unsecured bond (34 days), or financial conditions (29 days). This trend varied only slightly across all major offense categories -- in general, defendants released by conditional means were detained more days than defendants who were released by any other means. Of defendants not released, those detained by the court were detained longer, on average, than those held on financial conditions (111 days compared with 79 days). This pattern was true across all major offense types. Chapter notes 1) All tables in chapter 3 were created from data in the Pretrial Services Information Act System data base, which is maintained by the Pretrial Services Agency (PSA) within the AOUSC. The data describe pretrial defendants processed by Federal pretrial service agencies within each district. Defendants who received pretrial services through a local, non-Federal agency, such as the District of Columbia, are not included. The data describe 78,060 defendants whose cases commenced and 71,572 defendants who terminated pretrial services during October 1, 2001, through September 30, 2002, and whose cases were filed by complaint, indictment, or information. In these tables, the totals ("all offenses") ncluded records whose offense or other attributes were missing or indeterminable. The percentage distributions were based on non-missing values, and missing values were reported in a separate row or in a footnote. Offenses in the PSA are based on the most serious charged offense, as determined by the probation officer responsible for the interview with the defendant. The probation officer classifies the major offense charged into AOUSC four-digit offense codes. For defendants charged with more than one offense on an indictment, the probation officer chooses as the major charged offense the one carrying the most severe penalty or, in the case of two or more charges carrying the same penalty, the one with the highest offense severity. The offense severity level is determined by the AOUSC, which ranks offenses according to the maximum sentence, type of crime, and maximum fine amount. These four-digit codes are then aggregated into the same offense categories as those used in chapter 2. For drug offenses, the type of drug activity -- trafficking or possession is obtained by the probation officers from their reading of the indictment or other charging documents. The AOUSC citation manual provides probation officers with detailed instructions on how to code drug crimes. 2) In tables 3.1-3.5, the percentages showing the methods of release or methods of detention were based on the number of defendants released or the number of defendants detained. In tables 3.6 and 3.7, the percentages were based on the number of defendants who had hearings and were ordered detained. This method departs from the 1993 and prior compendia, in which the percentages were based on the number of defendants terminating pretrial services. Chapter 4 Adjudication Federal criminal cases may result in conviction of the defendant following plea or trial, acquittal of the defendants after trial, or dismissal of the case.***Footnote 1: Unless otherwise noted, data describe felony and misdemeanor cases in U.S. district courts and include Class A misdemeanors handled by U.S. magistrates. The data also include Class B misdemeanors that are handled by U.S. district court judges, which occur infrequently.*** Defendants in criminal cases commenced During 2002, 87,727 defendants had criminal cases commenced against them in U.S. district courts, and 76,163 (87%) of those defendants were charged with felonies. The 30,673 drug felony defendants comprised 35% of all criminal defendants in cases commenced and 40% of all felony defendants. Felony property and immigration defendants comprised 18% and 15%, respectively, of all defendants in cases commenced. The 11,493 misdemeanor defendants constituted 13% of all defendants in cases commenced; 45% of those misdemeanor defendants were charged with traffic violations. Defendants in criminal cases terminated Cases were terminated against 80,424 defendants during 2002,70,225 (87%) of whom were felony defendants. During 2002 the 29,300 drug-related felony defendants comprised 36% of all defendants in cases terminated and 42% of all felony defendants. Eighty-nine percent of all defendants, 92% of all felony defendants and 71% of misdemeanor defendants were convicted. Conviction rates for major felony offense categories were in the 87% to 95% range. Within major offense categories, however, the conviction rates varied more widely. For example, within the public-order category, the conviction rate ranged from 51% for national defense violations to 69% for postal law violations to 96% for tax law violations. Most defendants who were convicted pleaded guilty. Overall, almost 96% of those convicted pleaded guilty, while only 4% were convicted at trial. For felony offenses, 96% pleaded guilty to their charges. For the major felony offense categories, guilty pleas were registered for 99% of immigration offenders, 96% for each of drug and property offender, 94% and 93% for public order and violent offenders, respectively, and 93% of weapon offenders. During 2002, only 4% (3,500) of the 80,424 defendants in cases terminated exercised their right to a trial. Eight percent of weapon offenders went to trial as compared with 7% of violent offenders, 6% of public order offenders, 4% each of property and drug offenders, and 1% of immigration offenders. Of defendants who exercised their right to a trial, 2,726 (78%) were convicted either by a jury or a bench trial. The felony trial conviction rate was 82%, while the misdemeanor conviction rate was 56%. Among felony offenses, trial conviction rates ranged from 78% for public-order defendants to 88% for immigration defendants. Case processing times The data in table 4.3 cover the interval from the time a case is filed in U.S. district court through sentencing for those convicted as well as the interval from case filing through disposition for those not convicted, or those whose cases are dismissed. Title I of the Speedy Trial Act of 1974, as amended, sets time requirements for processing criminal cases in Federal courts. Except for certain exclusions, indictment must occur within 30 days of arrest; defendants are guaranteed at least 30 days to prepare for trial; and the Government must be ready for trial within 70 days. The Speedy Trial Act does not specify an interval from trial to sentencing,***Footnote 2: 2The Sentencing Reform Act of 1984 specifies minimum intervals for the disclosure and objections to the presentence report. (See rule 31, Federal Rules of Criminal Procedure.)*** and the time limits set by the act exclude several enumerated periods of pretrial procedure such as time spent awaiting rulings on motions. ***Footnote 3: 18 U.S.C. 3161 et. Seq.*** Additionally, defendants may waive their rights within the Speedy Trial Act. As a result, the actual processing time for most defendants can be longer than the statutory limits without violating the provisions of the act. Overall, the average time for processing defendants was 9.5 months. For felony defendants, the overall processing time average was 10.2 months; for those charged with misdemeanor crimes, the average processing time was 4.6 months. Defendants who pleaded guilty were processed, on average, 5.0 months quicker than defendants who went to trial. Overall, defendants whose cases went to trial took the longest to process (13.8 months), on average. Among major felony offense categories, case processing times were similar to the overall pattern, except when comparing the time until dismissal to the time of trial. Defendants convicted of public-order and drug offenses took a greater amount of time for dismissal (19.6 and 20.3 months, respectively) on average, than for trial (17.9 and 17.5 months, respectively). Convictions by U.S. Magistrates During 2002 U.S. magistrates disposed of 10,045 misdemeanor criminal defendants, 71% of whom were convicted. Public-order offenders comprised nearly 55% of the defendants disposed by U.S. magistrates, with property offenders comprising another 22%. Drug offenses comprised a little over 14% of U.S. magistrates' cases. Characteristics of convicted defendants Among defendants convicted, there were 6 times as many men as women (86% versus 14%); over twice as many whites as blacks(70% versus 26%); and 1« times as many non-Hispanics as Hispanics (60% versus 40%). Most were U.S. citizens (66%), had graduated high school or completed some higher education (54%), and had some criminal history (60%). Persons over the age of 30 represented 56% of defendants. Chapter notes 1) Tables 4.1-4.4 were derived from the Administrative Office of U.S. Courts (AOUSC) criminal master data files. Only records with cases filed in U.S. district court (table 4.1) or cases that terminated in U.S. district court during October 1, 2001, through September 30, 2002, were selected. Offenses were classified according to the most serious offense charged. In the case of multiple offenses, the offense carrying the most severe potential penalty was selected. In this Compendium, carjacking offenses are classified as robberies, based on title and section of the U.S. Code. In compendia prior to the 1997 Compendium, they were classified as motor vehicle thefts, based on the AOUSC offense classifications. 2) Table 4.5 was created by matching the AOUSC master data files with the U.S. Sentencing Commission (USSC) monitoring system files and the Pretrial Services Agency (PSA) data files. These latter two data files contain information on the characteristics of defendants. The USSC monitoring system files are limited to records of defendants sentenced under the Federal sentencing guidelines only. These include defendants convicted of felonies or Class A misdemeanors. Excluded from the USSC data were defendants convicted only of Class B or C misdemeanors or infractions, defendants whose offenses were committed before November 1, 1987, and juvenile offenders. Juveniles charged as adults are included in table 4.5. Some of the defendants excluded from the USSC data files were included in the PSA data. (See Chapter 3 Chapter notes for more information on the PSA data.) Table 4.5 indicates the number of records for which relevant data were available. Percentage distributions were based on records with known values of defendant characteristics. 3) A criminal record, as reported in table 4.5,is limited to prior adult convictions. For some defendants in this table, it is further limited to the portion that is relevant for calculating sentences under the Federal sentencing guidelines. In general, this is limited to sentences imposed within a 15-year period prior to the current offense and offenses committed within the United States. For most defendants, the criminal history used to calculate sentencing guideline ranges includes their entire adult criminal history. Chapter 5 Sentencing Depending upon the type and severity of the offense, convicted offenders may be sentenced to incarceration, probation, a fine, or a combination of sanctions such as a split or mixed sentence (imprisonment as well as a period of probation supervision). (For a definition of mixed sentences, see Glossary, p. 113.) The Federal sentencing guidelines require a term of supervised release following service of any prison sentence of more than 1 year. In addition, courts have the discretion to impose supervised release in any other case. Except where otherwise indicated, tables in this chapter are based on the most serious offense of conviction. They are not directly comparable with tables in earlier chapters that are based on the most serious offense investigated or most serious offense charged. (See "Offense classifications" in Methodology, p. 109.) Offenders convicted and sentences imposed Of the 71,798 offenders sentenced during 2002, 53,682 (75%) were sentenced to prison; 12,190 (17%) were sentenced to probation ***Footnote 1: Offenders given an intermediate sanction such as intermittent confinement or community confinement that also included probation supervision are counted among offenders given probation.***; and 2,897 (4%) were ordered to pay only a fine. During 2002, 52,246 defendants convicted of felonies were sentenced to prison representing 97% of all prison sentences and 83% of all convicted felons. Felony drug, violent, weapon, and immigration offenders were more likely to receive prison sentences (91%, 93%, 92%, and 89% respectively) than were property (59%) or public-order (66%) offenders. Among public-order offenders, persons convicted of nonviolent sex offenses (91%) and escape (90%) were almost as likely to receive prisons sentences as were violent, drug, weapon, and immigration offenders. Among property offenders, persons convicted of arson or the use of explosives (76%) were the most likely to receive prison sentences. Overall, 17 % of convicted offenders were sentenced to probation. The percentage of misdemeanants sentenced to probation (47%) was nearly 4 times that of convicted felons (13%). Persons convicted of food and drug offenses were the most likely to receive probation (78%), followed by persons convicted of postal law violations and larceny misdemeanors (74%). While 4% of all convicted felons were ordered to only pay a fine, 30% of the felons convicted of antitrust violations were so ordered. Fines were generally reserved for misdemeanor offenders, 31% of whom received fines. Average prison sentencesimposed For all offenders sentenced to prison terms, the average term of imprisonment imposed was 57.1 months. Persons convicted of felonies received an average of 58.4 months. The average length of prison sentence length imposed varied among major felony offense categories. Property offenders received the shortest sentence, on average, while violent offenders received the longest (25.0 months compared to 88.5 months). Weapon offenders received an average prison term of 83.9 months, and drug offenders received 76.0 months. Lesser terms were administered for public-order offenders receiving 38.5 months, on average, and immigration offenders receiving an average prison term of 27.9 months. The average sentence length for violent offenses (58.4 months) was almost 6 times longer than average sentence length for misdemeanor offenses (9.8 months). Relationship between sentence imposed and mode of conviction During 2002, offenders convicted by trial were more likely to receive prison sentences than those who pleaded guilty. Seventy-four percent of the offenders convicted by guilty plea received some prison time, while 87% of offenders convicted at trial received prison sentences. For violent offenders, 93% of those convicted by plea received prison sentences, as did 97% of those convicted at trial. For property offenders, 58% of those convicted by plea received prison sentences, compared to 86% of those convicted at trial. For public-order offenders, 65% of those convicted by plea, as contrasted with 90% of those convicted at trial, went to prison. The average prison term imposed on defendants convicted at trial was longer than the term imposed on defendants convicted by plea. Defendants convicted at trial received 144.4 months, on average (a median of 84 months), while those convicted by plea received an average of 53.3 months (a median of 33 months). Drug offenders convicted at trial received an average of 195.8 months as compared to 71.4 months for drug offenders convicted by a guilty plea. Violent offenders who went to trial received an average sentence of 204.4 months as compared to the 81.3 months for those convicted by guilty plea. Those convicted at trial for weapon offenses received, on average, 188.5 months, compared to the 75.4 months for weapon offenders convicted by guilty plea. Characteristics of offenders sentenced to prison The majority of persons convicted during 2002 were male (86%), white (70%), non-Hispanic (60%), and U.S. citizens (66%). The percentage sentenced to prison was not uniform across all categories of age, race, sex, education, and ethnicity. Similarly, the average sentences imposed were not uniform across these categories. However, conclusions about the effect that characteristics may have had on sentencing cannot be drawn from the aggregate patterns in the data. Such an assessment would require detailed information for each category showing the presence or absence of factors that are legally intended to affect sentencing, such as offense severity, weapon use, role in the offense, victim injury, dollar loss, and so forth. One such study of the factors affecting sentencing found that "nearly all of the aggregate differences among sentences for whites, blacks, and Hispanics during 1989-90 can be attributed to characteristics of offenses and offenders that current law and sentencing guidelines establish as legitimate considerations in sentencing decisions." ***Footnote 2: Douglas C. McDonald and Kenneth E. Carlson, Sentencing in the Federal Courts: Does Race Matter? Washington, DC: U.S. Department of Justice, Bureau of Justice Statistics, December 1993 (NCJ 145328), p. 1.*** Average sentences imposed were longest for males (59.2 months), blacks (84.5 months), non-Hispanics (67.2 months), and U.S. citizens (66.9 months). These categories of offenders (males, blacks, non-Hispanics, and U.S. citizens) are also more highly concentrated in the offense types associated with longer average sentences. In general, average prison terms were longer for violent, drug and weapon offenses than for property, public-order and immigration offenses. These three offense categories (violent, drug, and weapons)are generally considered to be more serious under the Federal sentencing guidelines than are property and public-order offenses. Chapter notes 1) Tables 5.1-5.3 were generated from the AOUSC criminal master data files. Only records of defendants sentenced between October 1, 2001 and September 30, 2002 were selected. Offenses shown in these tables -- offenses of conviction -- are based on the longest actual sentence imposed. 2) Tables 5.4-5.6 were generated from AOUSC criminal master data files supplemented with linked data on offender characteristics from two sources: The United States Sentencing Commission (USSC) monitoring system files (which are limited to records of defendants sentenced under the Federal sentencing guidelines); and the Pretrial Services Information Act System database, maintained by the Pretrial Services Agency (PSA). These two supplemental data files contain information on the characteristics of offenders. Table 5.4 indicates the number of records for which relevant data were available. Percent distributions were based on records with known values of defendant characteristics and offenses. Means and medians in tables 5.5 and 5.6 were based on the number of records shown in table 5.4 3) Criminal record, as reported in tables 5.4-5.6, is limited to prior adult convictions. For some defendants in these tables, it is further limited to the portion of their criminal record that is relevant for calculating sentences under the Federal Sentencing Guidelines. In general, this is limited to sentences imposed within 15 years of the current offense and sentences for offenses committed within the United States. For most defendants, the criminal history used to calculate sentencing guideline ranges is the same as their adult criminal history. Chapter 6 Appeals Prior to implementation of the Sentencing Reform Act of 1984, only criminal convictions could be appealed. However, the Sentencing Reform Act provided for the appellate review of sentences imposed given that the sentence was (1) imposed in violation of the law; (2) imposed as the result of an incorrect sentencing guideline application; (3) outside the recommended guideline sentencing range; or(4)imposed for an offense for which no sentencing guideline exists and is plainly unreasonable. Both the defendant and the Government have the right to appeal an imposed sentence.***18 U.S.C. Sec.3742*** Appeals filed During 2002 the U.S. Court of Appeals received 11,569 criminal appeals, a small increase (up 3% from 2001) when compared to that of 2000 to 2001 (23%). Of all appeals, 4% were filed by the Government (not shown in a table). Almost 43% of all appeals filed (for which the offense is known) were for drug convictions. Appeals for property offense convictions comprised 16% of all appeals filed, immigration offenses comprised 15% as well, weapons offenses were 13%, and 6% of appeals filed were for violent offenses. Eighty-four percent of all appeals filed in 2002 were guidelines-based appeals. Sixteen percent of appeals filed were appeals of convictions for crimes committed before 1987, when the sentencing guidelines took effect. Forty-nine percent of all appeals challenged both the conviction and the sentence imposed. Nearly 3 times as many appeals challenged only the sentence imposed as those that challenged only the conviction (26% versus 9%). The distribution of guideline-based appeals by type of appeal filed (sentence, conviction, or both) was relatively constant across the major offense categories underlying the appeal. However, 63% of post-guideline appeals filed for immigration offenses appealed both the conviction and the sentence, compared to 58% for all criminal offense categories. Appeals terminated A total of 11,695 appeals terminated during 2002. Sixteen percent were appeals of convictions for offenses sentenced under laws in effect prior to the sentencing guidelines. The remaining 84% of appeals terminated were guidelines-based appeals. Of these, 58% were appeals of both the conviction and sentence, 11% were appeals of the conviction only, and 31% were appeals of the sentence only. In nearly half of all appeals terminated in 2002(for which the offense is known), drug offense convictions underlie the appeal (44%); the underlying offense in 16% of the appeals was for a property offense; 14% were for an immigration offense; 12% were for weapon offense appeals; and 6% were appeals for violent offense convictions. Robbery offenses composed 54% and murder composed 17% of the underlying offenses for appeals of a violent conviction. Sixty-eight percent of property offense appeals were for fraud convictions. Of the 11,695 appeals terminated during 2002, 75% were terminated on the merits, while the remainder were terminated on procedural grounds. An appeal terminated on the merits is an appeal that is terminated based on the factors of the case, not because of technical or procedural reasons. For appeals based on a conviction, 75% of drug offenses and 69% of public-order offenses were terminated on the merits. Also terminated on the merits were 81% of appeals based on a weapon conviction, 78% of appeals based on an immigration or a violent conviction, and 74% of appeals based on a property conviction. Within these major offense categories, 66% of appeals based on murder convictions were terminated on the merits, as were 82% of robbery convictions, and 74% of fraud convictions. In 79% of appeals terminated on the merits, the district court ruling was affirmed. In another 4% it was partially affirmed. Among appeals terminated on merits, 81% of all drug and weapon appeals were affirmed, 75% of all immigration and public order appeals were affirmed, as were 76% of violent offense appeals. District court decisions were reversed or remanded back to the court in 8% of the cases. Property and public order offenses had the highest rate of reversal or being remanded back to the court(11% and 13%). Nine percent of weapon offense appeals were reversed or remanded, as were 8% of violent offense appeals. Less than 8% of appeals were reversed or remanded for drug or immigration offenses. Overall, appeals were dismissed in 8% of cases. Among major offense categories, the highest dismissal rate was for immigration offenses (16%), an increase of 8 percentage points from fiscal year 2001 when only 8% of immigration appeals were dismissed. Chapter notes Tables in chapter 6 were created from the AOUSC appeals database. Records of appeals filed or terminated during October 1, 2001, through September 30, 2002, were selected. Data include records of 11,569 criminal appeals filed during fiscal year 2002 and 11,695 appeals terminated during the same period. The unit of analysis in the appeals data is a single case. Most records in the appeals data report on a single appellant. Appeals were classified into the offense categories that represent the underlying offense of conviction. Offenses represent the statutory offense charged against a defendant in a criminal appeal. Offenses were classified by the AOUSC into their four-digit offense codes. These were then aggregated into the offense categories shown in the tables. Chapter 7 Corrections Federal offenders under supervision As of the end of fiscal year 2002, there were 107,367 offenders under active Federal supervision, of which most (91%) were felons. Seventy- two percent of these offenders received one of two forms of post- incarceration supervision: supervised release (73,229) or parole (3,561). The remainder (30,577) were under probation supervision. Most of the probationer population had been convicted of either a property felony (38%) or some type of misdemeanor offense (31%). Fifty-four percent of offenders under supervised release and 45% of parolees had been convicted of a drug offense. Among the 107,367 offenders under active Federal supervision at the end of fiscal year 2002, 78% were male; 63% were white; 81% were of non-Hispanic origin; 37% were over age 40 (compared to 31% who were between 31 and 40 years of age and 32% who were 30 or younger); 37% had a high school diploma only (compared to 31% who had at least some college and 33% who had less than a high school diploma); and 61% had no known drug history. Outcomes of offenders terminating supervision Probation -- During 2002, 15,116 offenders concluded one or more terms of active probation. Overall, 80% of offenders successfully completed their term of probation, another 18% violated their conditions of probation, and the remaining 2% were administrative closures. Of offenders terminating probation, about 7% committed a new crime. Most committed technical violations, including drug use (4%) or absconding (3%). Offenders convicted of violent offenses were less likely than others to successfully complete a term of probation without a violation. During 2002, 32% of probationers convicted of violent offenses violated their conditions of probation, as did 29% of weapon, 22% of drug, 21% of immigration, and 16% of property offenders. Violent and weapon probationers were also more likely to commit new crimes (12%) than were probationers convicted of weapon (10%), drug (7%), or property (5%) offenses. Supervised release -- During 2002, 27,678 offenders concluded terms of supervised release. Overall, 62% of them successfully completed their term of supervised release; 23% committed technical violations, such as drug use (8%) or absconding (5%); and 13% of these offenders violated their supervision by committing a new crime. The remaining 2% had their supervision administratively terminated. Immigration offenders were less likely than others to complete a term of supervised release without a violation. Sixty-four percent of immigration offenders violated conditions of supervised release, while 48% of weapon, 33% of drug, and 29% of property offenders violated supervised release. Immigration offenders were also more likely to commit new crimes (38%) than offenders convicted of violent (17%), weapon (17%), drug (12%), or property (10%) offenses. Parole -- During 2002, 1,817 offenders concluded terms of parole. Overall, 58% of these offenders successfully completed their term of parole. Thirteen percent of these parolees violated their supervision by committing a new crime; 23% committed technical violations, such as drug use (9%) or absconding (5%); and 7% completed their parole term through an administrative case closure. Offenders convicted of violent offenses were less likely than others to complete a term of parole without a violation. Forty-nine percent of violent offenders violated conditions of parole, while 34% of public order and property, and 26% of drug offenders violated parole. Violent and property offenders were also more likely to commit new crimes (16%) than offenders convicted of drug or public order(10%)offenses. Characteristics of offenders terminating supervision Probation -- Among offenders whose term of probation was concluded, 72% were male; 67% were white and 84% were of non-Hispanic origin; 35% were less than age 31 (compared to 39% over age 40); 26% had less than a high school diploma (compared to 38% whose highest education was a high school diploma and 37% who had at least some college education); and 79% had no known drug history. Younger offenders, those with a history of drug abuse, and those with a lower educational background had higher probabilities of violating the conditions of probation. Offenders ages 16 to 20 were more likely (42%) to violate conditions of supervision than probationers in other age groups -- 26% of those ages 21 to 30 and 14% of those over age 30. Probationers with a history of drug abuse were 5 times more likely to violate probation for drug use than offenders who were not drug abusers. Probationers with known histories of drug abuse were also more likely to violate supervision for other reasons (fugitive status or new crimes) than were other probationers. Probationers with lower levels of education were more likely to violate the conditions of probation than those with higher levels of education. Those who did not graduate high school had a 28% violation rate, those whose highest education was a high school diploma had a 18% violation rate, those with some college had a 12% violation rate, and those with a college degree had only a 7% violation rate. Supervised release -- Among those who concluded a term of supervised release, 84% were male; 63% were white; 79% were of non-Hispanic origin; 26% were less than age 31 (40% were over age 40); 39% had only some high school (35% whose highest education was a high school diploma and 26% had at least some college); and 54% had no known drug history. Younger offenders, those with a history of drug abuse, and those with a lower educational background had higher probabilities of violating the conditions of supervised release. Offenders under age 30 were more likely to violate conditions of supervision than offenders in other age groups -- 54% of those under age 31 violated a condition of their supervision compared to 22% of those over age 40. Releasees with a history of drug abuse were nearly 4 times more likely to terminate their supervision for technical violations of drug use as were releasees who were not drug abusers. Releasees with known histories of drug abuse were also more likely to violate supervision with technical violations or new crimes. Similarly, releasees with lower levels of education were more likely to violate conditions of supervised release than those with higher levels of education. Those who did not graduate from high school had a 45% violation rate, those whose highest education was a high school diploma had a 35% violation rate, those with some college had a 25% violation rate, and those with a college degree had a 11% violation rate. Parole -- Among those who concluded a term of parole, 97% were male; 59% were white; 87% were of non-Hispanic origin; 82% were over age 40; 46% had only some high school (35% whose highest education was a high school diploma and 18% had at least some college); and 53% had no known drug history. Offenders with a history of drug abuse and those with a lower educational background had higher probabilities of violating the conditions of parole. Releasees with a history of drug abuse were 3 times more likely to terminate their supervision for technical violations of drug use than were releasees who were not drug abusers. Releasees with known histories of drug abuse were also more likely to violate supervision with technical violations or new crimes. Releasees with lower levels of education were usually more likely to violate conditions of parole than those with higher levels of education. Those whose highest education was a high school diploma had a 38% violation rate, those who did not graduate from high school had a 39% violation rate, those with some college had a 26% violation rate, and those with a college degree had a 17% violation rate. Admissions, releases, and standing population of Federal prisoners The Federal prison population grew by 7,045 persons during fiscal year 2002, increasing from 135,986 sentenced prisoners to 143,031.***Footnote 1: Tables 7.9-7.16 exclude prisoners sentenced by a District of Columbia Superior Court judge for violations of the DC Criminal Code.*** The greatest growth was among drug, weapons, and immigration offenders. In 2002 the number of drug offenders in prison grew by 4,111, weapon offenders by 1,648, and immigration offenders by 712. Federal prisoners: First releases and time served During 2002, 44,339 prisoners were released from Federal prison for the first time after serving a sentence imposed by a U.S. district court.***Footnote 2: Tables 7.11-7.16 include only prisoners first released by the BOP during 2002.*** Of these, 39,568 were released by standard means. These standard releasees served an average of 31.6 months in prison -- 91% of the sentence imposed, on average. ***Footnote 3: Average time to first release is the number of months in Bureau of Prisons facilities minus credits for time spent in jail prior to final disposition or sentencing. Percent of sentence served is the average percentage of each individual prisoner's percent of total sentence obligation served until first release.*** The remaining 4,771 first releases were released by extraordinary means: 3,563 were released early due to participation in a drug treatment program; 324 were deported; and 884 were released for other reasons -- including commutation, death, or completion of an intensive confinement program. Drug offenders made up the majority of those released by extraordinary means (84%). Time served until first release varied by offense: Violent offenders served an average of 58.8 months; drug offenders and weapons offenders served an average of 42.9 months; the average number of months served by immigration offenders was 19.2; and property offenders served an average of 16.5 months. While violent, weapon, and drug offenders served more time before first release, property, immigration, and public-order offenders served more of their sentences than violent and drug offenders. Old law offenders -- those who committed crimes before implementation of Federal Sentencing Guidelines in November 1987 - that were released in 2002 are not shown separately in any table. Less than 1%, or 265, of all released offenders were old law offenders. New law offenders released during 2002 served, on average, 30.9 months in prison (91% of the sentence imposed). Since the Guidelines became effective as of November 1987, new law offenders sentenced to terms of imprisonment greater than 204 months could not have been released by standard means as of the end of fiscal year 2002.***Footnote 4: The Federal Sentencing Guidelines allow a prisoner up to 54 days per year for good behavior, so a prisoner sentenced to 191 months in prison in November 1987 could have served 166 months of his or her sentence and been released in 2001.*** They could have been released only for extraordinary reasons such as death, commutation, or treaty transfer.***Footnote 5: Prisoners released for extraordinary reasons have been excluded from all time served and percent of sentence served calculations.*** Characteristics of Federal prisoners Standing population -- Of prisoners in Federal prison at the end of 2002, 93% were male, 58% were white, 39% were black, 32% were Hispanic, and 72% were U.S. citizens. Fifty-five percent of the yearend population were 31 years of age or older. This portrait of the Federal prison population was similar across offenses, with a few exceptions. Ninety-five percent of immigration offenders were white, 89% were Hispanic and only 4% were U.S. citizens. More than 92% of violent offenders were non-Hispanic and U.S. Citizens. Prisoners released -- Of prisoners released during 2002, 88% were male, 73% were white, 24% were black, 45% were Hispanic, and 58% were U.S. citizens. Fifty-five percent were age 31 or older. These patterns were most similar for those released for drug offenses, and least similar for offenders released for immigration, weapon, and violent offenses. Ninety-six percent of released immigration offenders were male, 96% were white, 3% were black, 91% were Hispanic and 5% were U.S. Citizens. Time served to first release -- Of prisoners released by standard methods, males served more time than females (an average of 32.7 months versus 22.5 months); blacks served more time than whites (44.2 months versus 27.5 months); and non-Hispanics served more time than Hispanics (36.1 months versus 26.4 months). Older persons served more time than younger (until over age 40), and U.S. citizens served more time than noncitizens. However, across offenses, the overall pattern does not apply. For example, blacks served an average of 17 months longer than whites, but among property offenders released from prison, whites served the same amount of time as blacks. Black offenders served an average of 57 months for drug offenses, compared to 37 months for white offenders. Among immigration offenders, non-citizens were in prison an average of 5 months longer than U.S. citizens before being released. Chapter note 1) Offenders were classified according to their most serious offense at conviction. In cases involving multiple offenses, the AOUSC offense severity hierarchy was applied. The most serious offense was the one with the most severe penalty imposed. If equal prison terms were imposed, or there was no imprisonment, the offense with the highest severity code, as determined by the U.S. Title and Code Criminal Offense Citations Manual, was selected. 2) Tables 7.1 and 7.2 were created using the probation, parole, and supervision data files of the Federal Probations Supervision Information System (FPSIS), which is maintained by the AOUSC. Yearend pending cases (that is, records with offenders who were under active supervision as of the end of the fiscal year, September 30, 2002) were selected. Corporate defendants were excluded from table 7.1. 3) Tables 7.3-7.8 were created using the FPSIS data files as well. Only records with one or more terminations of active supervision during October 1, 2001, through September 30, 2002, were selected. Each termination was counted separately. Technical violations and terminations for new crimes are shown only if supervision terminated with incarceration or removal from active supervision for reasons of a violation. Corporate defendants were excluded from tables 7.3-7.8. 4) Table 7.9-7.16 were created from the Federal Bureau of Prisons (BOP) data files. Tables 7.9 and 7.10 show sentenced prisoners in the custody of the BOP at the end of fiscal year 2002. Tables 7.11-7.16 include only prisoners committed by U.S. district courts or released by the BOP during fiscal year 2002. In these tables, a single person may be counted more than once if that person was committed into, or released from, the BOP more than one time during fiscal year 2002, or if that person appears in more than one column in a table, e.g., commitment and population. Offense categories in these tables were based on combinations of offense designations used by the BOP. They are similar, but may not be directly comparable, to the categories used in other tables of this Compendium. Offenses for some prisoners admitted or released were not able to be classified; these offenders were included in the totals and are shown as "other prisoners" on the last line of tables. In these tables offenses were classified according to the offense associated with the longest single sentence actually imposed. Classifications in other tables may have been based on the longest potential sentence allowed by law. 5) Tables 7.9-7.16 distinguish between prisoners committed by U.S. district courts for violations of the U.S. Code and other prisoners. Prisoners released after commitment from U.S. district court are called "first releases." The other admissions and releases include offenders who returned to prison after their first release (such as probation, parole, or supervised release violators), offenders convicted in other courts (such as military courts), and persons admitted to prison as material witnesses or for purposes of treatment, examination, or transfer to another authority. Tables 7.9-7.16 exclude prisoners sentenced by a District of Columbia Superior Court judge for violations of the DC Criminal Code. Offenders who entered or left a prison temporarily -- such as for transit to another location, for health care, or to serve a weekend sentence -- were not counted as admitted or released. Persons who were detained for deportation by the Immigration and Naturalization Service and who were not criminal offenders were also not included in the tables. Table 7.11 distinguishes between standard and extraordinary methods of release from prison. "Standard" releases include the usual methods of exiting from prison, such as full-term sentence expirations, expirations with good time, mandatory releases, and releases to parole. Extraordinary releases are unusual exits from prison, such as death, commutation, and transfer to another facility. 6) Tables 7.11-7.16 include only prisoners committed by U.S. district courts and released by the BOP during fiscal year 2002. Prisoners committed by U.S. district court -- but not for a violation of the U.S. Code, or whose offense could not be classified -- were included in the total but are not shown separately. Table 7.9, however, shows these prisoners separately. Other prisoners -- such as probation and parole violators and prisoners committed by other courts, such as military courts or District of Columbia Superior Court -- were excluded from tables 7.11-7.16, as were other persons admitted to Federal prison but not committed from a U.S. district court. 7) In tables 7.11-7.16, time served was calculated for prisoner's arrival into jurisdiction of the BOP until first release from prison, plus any jail time served and credited. The calculation is the same as that currently used by the BOP. Prisoners serving consecutive sentences may have total imposed sentences exceeding the longest single sentence length. Accordingly, time actually served may exceed the longest single imposed sentence. The percent of sentenced served is the average of each individual prisoner's percent of sentence served. Because other publications may include different groups of prisoners, calculate time served differently, or use a different offense classification, data in tables in this Compendium may differ from estimates of time served in previous publications by the BOP or in publications based on other data sources. Time served, as reported in tables 7.11-7.16 in this Compendium, may not be directly comparable to the calculation of time served in the 1993 and prior compendia. The methodology in this report uses additional information to identify prison commitment dates and account for jail credit. In previous reports, jail credits were overestimated in some cases. In addition, time served estimates reported in tables 7.12-7.14 and 7.16 include only prisoners released by standard methods (expirations, mandatory releases, and releases to parole). Prisoners released by extraordinary means (death, commutation, transfer, and so on) are excluded from the time served calculations in these tables. 8) Table 7.9 shows all persons admitted to, or released from, the jurisdiction of the BOP during fiscal year 2002, except those prisoners who were committed by a District of Columbia Superior Court judge. Prisoners counted in this column are the same as prisoners included in tables 7.11, 7.12, and 7.13. Table 7.9 shows in separate columns, as well as in the total, prisoners committed by a U.S. district court -- but not for a violation of the U.S. code, or whose offense could not be classified. Other columns in table 7.9 include prisoners who were committed by other courts, returned to prison for violation of the conditions of supervision, or were received for examination, treatment, or transfer to another jurisdiction. Offenders who returned to prison for a violation of the conditions of supervision without a new court commitment, were classified according to the offense with the longest single sentence originally imposed at conviction. Methodology The Federal justice database Source of data The source of data for all tables in this Compendium is the Bureau of Justice Statistics (BJS) Federal justice database. The database is presently constructed from source files provided by the U.S. Marshals Service, the Drug Enforcement Administration, the Executive Office for U.S. Attorneys, the Administrative Office of the U.S. Courts, the United States Sentencing Commission, and the U.S. Bureau of Prisons. The Administrative Office of the U.S. Courts, in addition to providing data describing defendants in cases processed by the Federal judiciary, provides data describing defendants processed by the Federal pretrial services agencies and the Federal probation and supervision service. Federal law prohibits the use of these files for any purposes other than research or statistics. A description of the source agency data files is provided in the table at the end of this section. Some records in the Federal justice database are matched according to a statistically weighted combination of names, other personal identifiers, dates of court appearances, types of offenses, and other relevant information contained in the files. Using the matched data files, it is possible to combine information about two or more stages of the processing of a criminal matter or case, for example from adjudication to probation or parole supervision. Reporting period Wherever possible, matters or cases have been selected according to the event which occurred during fiscal year 2002 (October 1, 2001, through September 30, 2002). Files which are organized by their source agencies according to fiscal year nonetheless include some pertinent records in later years' files. For example, tabulations of suspects in matters concluded during fiscal year 2002 in this Compendium have been assembled from source files containing records of 2002 matters concluded which were entered into the data system during fiscal years 2002 or 2003. Availability of data items The availability of particular items of information is affected by the data source. For example, data on prosecutors' decisions prior to court filing are provided for cases investigated by U.S. attorneys but not for those handled by other litigating divisions of the U.S. Department of Justice. Criminal Division cases enter the data base once they are filed in U.S. district court, however. Many items of social and demographic information come from presentence investigation records, supervision records, or sentencing records, and are available only for arrested defendants who were convicted and/or began serving a sentence involving supervised release. This particularly affects gender, race, ethnicity, and prior record information. Table construction and interpretation The tables presented report events that occurred during the Federal fiscal year -- October 1, 2001 -- September 30, 2002. Generally the tables include both individual and organizational defendants. Organizational defendants are not included in tables describing defendants under pretrial release and detention, defendants sentenced to incarceration, and offenders under post-conviction supervision. Juvenile offenders are included in the reported statistics. Unit of analysis The unit of analysis in chapters 1 through 6 is a combination of a person (or corporation) and a matter or case. For example, if a single person is involved in three different criminal cases during the time period specified in the table, he or she is counted three times in the tabulation. Similarly, if a single criminal case involves a corporate defendant and four individual defendants, it counts five times in the tabulation. In chapter 7 the unit of analysis for incarceration, probation, parole, or other supervised release is a person entering custody or supervision, or a person leaving custody or supervision. For example, a person convicted in two concurrent cases and committed once to the custody of the Federal Bureau of Prisons in the indicated time period is counted as one admission to a term of incarceration. A person who terminates probation twice in the indicated time period, such as with a violation and again after reinstatement, is counted as two terminations of probation. Interpretation The tables in the Compendium are constructed to permit the user to make valid comparisons of numbers within each table and to compare percentage rates across tables. The total numbers of subjects in Compendium tables that are based on records linked between two files are generally less than the total number of records in either source file. Accordingly, comparisons of absolute numbers across two or more tables in this volume and other data sources are not necessarily valid. Offense classifications Procedure The offense classification procedure used in this Compendium is based on the classification system followed by the Administrative Office of the U.S. Courts. Specific offenses are combined to form the BJS categories shown in the Compendium tables.***These categories correspond to the Bureau of Justice Statistics crime definitions and, to the extent possible, are organized and presented consistent with BJS publications on State criminal justice systems.*** Offense categories for Federal arrestees in chapter 1 are based on the FBI's National Crime Information Center (NCIC) offense classifications, which are converted into U.S. Marshals' four-digit offense codes, which, in turn are aggregated into the offense categories shown in the tables. These categories are similar, but may not be directly comparable to the BJS offense categories used in other chapters of this Compendium. For data from the Executive Office for U.S. Attorneys, which include United States Code citations but do not include the Administrative Office offense classifications, United States Code titles and sections are translated into the Administrative Office classification system and then aggregated into the offense categories used in the tables. Offense categories for prisoners in chapter 7 are based on combinations of offense designations used by the Bureau of Prisons. They are similar to the categories in other chapters and other tables in chapter 7, but may not be directly comparable. Felony/misdemeanor distinctions Felony and misdemeanor distinctions are provided where possible. Felony offenses are those with a maximum penalty of more than 1 year in prison. Misdemeanor offenses are those with a maximum penalty of 1 year or less. Felonies and misdemeanors are further classified using the maximum term of imprisonment authorized. Section 3559, U.S. Code, Title 18 classifies offenses according to the following schedule: Felonies Class A felony -- life imprisonment, or if the maximum penalty is death. Class B felony -- 25 years or more. Class C felony -- less than 25 years but more than 10 years. Class D felony -- less than 10 years but more than 5 years. Class E felony -- less than 5 years but more than 1 year. Misdemeanors Class A misdemeanor -- 1 year or less but more than 1 month. Class B misdemeanor -- 6 months or less but more than 30 days. Class C misdemeanor -- 30 days or less but more than 5 days. Infraction -- 5 days or less, or if no imprisonment is authorized. In this Compendium, felony and misdemeanor distinctions are provided where the data permit these distinctions. Chapters 1 and 2 do not use this distinction because many suspects cannot be so classified at the arrest and investigation stages in the criminal justice process. Chapter 3 no longer reports this distinction because the Pretrial Services Agency no longer gathers this information. Chapters 4 and 5 distinguish between felony and misdemeanor offenses, as do tables 7.1 7.6. Tables 7.7 7.11 follow the convention of other BJS publications by separately tabulating offenders whose actual imposed sentences are less than or equal to 1 year and those whose actual sentences are greater than 1 year. Most serious offense selection Where more than one offense is charged or adjudicated, the most serious offense (the one that may or did result in the most severe sentence) is used to classify offenses. The offense description may change as the criminal justice process proceeds. Tables indicate whether investigated, charged or adjudicated offenses are used. In chapter 2 the most serious offense is based on the criminal lead charge as determined by the assistant U.S. attorney responsible for the criminal proceeding. In chapter 3 the major charged offense is based upon the Administrative Office's offense severity classification system, as determined by the pretrial officer responsible for the case. To select this offense, the officer ranks offenses according to severity based on maximum imprisonment, type of crime, and maximum fines. In chapter 4 the most serious offense charged is the one that has the most severe potential sentence. For chapter 5 conviction offenses are based on statutory maximum penalties. In chapter 6 offenses are classified by the offense of conviction. In tables 7.1 7.6, the most serious offense of conviction is either the one having the longest sentence imposed or, if equal sentences were imposed or there was no imprisonment, it was the offense carrying the highest severity code as determined by the Administrative Office's offense severity code ranking. In tables 7.7 7.13, prisoners are classified according to the offense which bears the longest single incarceration sentence. Offense categories For offense categories in all text tables, the following conditions apply: "Murder" includes nonnegligent manslaughter. "Sexual abuse" includes only violent sex offenses. "Fraud" excludes tax fraud. "Larceny" excludes transportation of stolen property. "Other property felonies" excludes fraudulent property offenses, and includes destruction of property and trespass. "Tax law violations" includes tax fraud. "Obscene material" denotes the mail or transport thereof. "All other felonies" includes felonies with unknown or unclassifiable offense type. "Misdemeanors" includes misdemeanors, petty offenses, and unknown offense levels. "Drug possession" also includes other drug misdemeanors. Glossary Acquittal -- legal judgment that a criminal defendant has not been proven guilty beyond a reasonable doubt of the charges against him. Affirmed -- in the appellate courts, the decree or order is declared valid and will stand as rendered in the lower court. Agriculture violations -- violations of the Federal statutes on agriculture and conservation: for example, violations of the Agricultural Acts, Insecticide Act, and the Packers and Stockyards Act, 1921; also violations of laws concerning plant quarantine and the handling of animals pertaining to research (7 U.S.C., except sections on food stamps (fraud); also 16 U.S.C. sections relating generally to violations in operating public parks, such as trespassing for hunting, shooting, and fishing). Antitrust violations -- offenses relating to Federal antitrust statutes, which aim to protect trade and commerce from unlawful restraints, price fixing, monopolies (for example, 15 U.S.C. sec sec 1, 3, 8, 20, and 70(i)), and discrimination in pricing or in furnishing services or facilities (15 U.S.C. sec sec 13(c) and 13(e)). Appeal -- a review by a higher court of a judgment or decision of a lower court. Appeals, U.S. Court of -- an intermediate Federal court, inferior to the U.S. Supreme Court, but higher than the U.S. district court. The function of the U.S. court of appeals is to review the final decisions of the district courts, if challenged. There are 13 courts of appeal in the Federal system representing the 12 judicial circuits and the Federal circuit (28 U.S.C. sec 41). Appellant -- the party which takes an appeal from 1 court or jurisdiction to another; opposite of appellee. Appellee -- the party against whom the appeal is taken; opposite of appellant. Arson -- willfully or maliciously setting, or attempting to set, fire to any property within the special maritime and territorial jurisdiction of the United States as defined in 18 U.S.C. sec 7 (18 U.S.C. sec 81; and 49A U.S.C. sec sec 1804 and 1809). (See also, "Explosives.") As