U.S. Department of Justice Office of Justice Programs Bureau of Justice Statistics ******************************************************** This file is text only without graphics and many of the tables. A Zip archive of the tables in this report in spreadsheet format (.csv) and the full report including tables and graphics in .pdf format are available on BJS website at: http://www.bjs.gov/index.cfm?ty=pbdetail&iid=5436 ********************************************************* Special Report Drug Offenders in Federal Prison: Estimates of Characteristics Based on Linked Data Sam Taxy, Julie Samuels, and William Adams, Urban Institute Publicly available BOP datasets do not include information about prisoners’ criminal history, offense details, and sentencing decisions, making it difficult to assess the criminal background of the offenders and the nature of their offenses. This report uses a new dataset that links the BOP population at fiscal yearend 2012 (i.e., stock population), with sentencing information from the U.S. Sentencing Commission (USSC) to describe the federal prison population in greater detail. The dataset includes the 94,678 sentenced offenders in federal prison whose most serious offense was drug-related, who were serving time on a new U.S. district court commitment, and who had valid links to USSC data. Of these, almost all were drug trafficking offenders, while the remainder were sentenced for other drug offenses, primarily possession. The 94,678 drug offenders drug offenders included in the dataset are referred to as drug offenders throughout this report. The Methodology and the Dataset composition and definition of drug offender textbox provide more information about the linking methodology, file composition, exclusionary criteria, and link rates. The counts and percentages based on the BOP variables presented in this report may differ slightly from other published Bureau of Justice Statistics (BJS) or BOP reports because only an analysis of linked observations is reported here. This report examines drug offenders in terms of their demographics, basic offense characteristics, and sentence imposed. Although the USSC dataset contains information related to both offense characteristics (e.g., quantity of drugs) and the justice system response (e.g., whether an offender was convicted of an offense carrying a mandatory minimum penalty and whether that penalty was applied at sentencing), the focus here is on the offenders and basic offense characteristics. The data were disaggregated by the primary type of drug involved (drug type is a variable only available in the USSC datasets). Other BOP and USSC variables of interest—demographics, criminal history, weapon involvement in the instant offense (the most recent offense for which the individual was sentenced and incarcerated), and sentence length—were analyzed by drug type. ************************************************* ************ HIGHLIGHTS ************ * This study is based on 94,678 offenders in federal prison at fiscal yearend 2012 who were sentenced on a new U.S. district court commitment and whose most serious offense (as classified by the Federal Bureau of Prisons) was a drug offense. * Almost all (99.5%) drug offenders in federal prison were serving sentences for drug trafficking. * Cocaine (powder or crack) was the primary drug type for more than half (54%) of drug offenders in federal prison. * Race of drug offenders varied greatly by drug type. Blacks were 88% of crack cocaine offenders, Hispanics or Latinos were 54% of powder cocaine offenders, and whites were 48% of methamphetamine offenders. * More than a third (35%) of drug offenders in federal prison at sentencing, had either no or minimal criminal history— lowest criminal history category. * Nearly a quarter (24%) of drug offenders in federal prison used a weapon in their most recent offense. * The average prison sentence for federal drug offenders was more than 11 years. * Across all drug types, crack cocaine offenders were most likely to have extensive criminal histories (40%), used a weapon (32%), and received longer prison terms (170 months). ************************************************* The number of federally sentenced prisoners in the Federal Bureau of Prisons (BOP) increased 84% between fiscal year (FY) 1998 and 2012, and the number of drug offenders in federal prison grew 63% during this time*** Footnote 1 See Federal Criminal Case Processing Statistics on the BJS website at www.bjs.gov/fjsrc ***. At fiscal yearend 2012, offenders whose most serious offense (as defined by the BOP) was a drug offense accounted for about half (52%) of the federally sentenced prison population (table 1). Previous analyses focusing on the growth of the prison population from FY 1998 to FY 2010 have shown that 42% of the growth in the federally sentenced population was due to an increase in the number of drug offenders, and the largest contributor to that growth was length of time served for drug offenses*** Footnote 2 Mallik-Kane, K., Parthasarathy, B., & Adams, W. (2012). Examining growth in the federal prison population, 1998–2010 (NCJ 239785). Washington, DC: Urban Institute***. Because drug offenders comprise about half of federal prison population and sentence length for this subpopulation is the greatest source of federal prison population growth, this report examines these offenders in greater detail. *********************************** Powder and crack cocaine offenders made up more than half of all BOP drug offenders *********************************** More than half (54%) of drug offenders in the federal prison system had a form of cocaine (powder or crack) as the primary drug type (table 2). Methamphetamine offenders (24%) accounted for the next largest share, followed by marijuana (12%) and heroin (6%) offenders. Offenders convicted of crimes involving other drugs (including LSD, some prescription drugs, and MDMA or ecstasy) made up 3% of offenders. *********************************** Three-quarters of BOP drug offenders were black or Hispanic *********************************** About three-quarters of drug offenders in federal prison were either non-Hispanic black or African American (39%) or Hispanic or Latino (37%); nearly a quarter (22%) were non-Hispanic white offenders (table 3). In addition, drug offenders in federal prison were overwhelmingly male (92%), about a quarter (24%) were noncitizens, and nearly 80% were age 30 or older. *********************************** Race and citizenship varied greatly by drug type *********************************** The vast majority of crack cocaine offenders (88%) were non- Hispanic black or African American. By comparison, more than half (54%) of powder cocaine offenders were Hispanic or Latino and 13% were non-Hispanic white. In addition, more than half (59%) of marijuana offenders and almost half (48%) of heroin offenders were Hispanic or Latino. A large proportion of methamphetamine offenders were non-Hispanic white (48%) or Hispanic or Latino (45%). Although most drug offenders were U.S. citizens, citizenship status varied across drug type subgroups. Almost all crack cocaine (97%) and other drug (89%) offenders were U.S. citizens. Aside from those, citizenship rates ranged from 71% for heroin offenders to 65% for marijuana offenders. Age and sex varied more subtly across drug types. The percentage of male drug offenders by type of drug ranged from 88% of methamphetamine to 96% of crack cocaine offenders. The percentage of persons age 30 or older ranged from 73% of other drug to 82% of powder cocaine offenders. *********************************** More than a third of federal drug offenders had the lowest level of criminal history at sentencing *********************************** More than a third (35%) of federal drug offenders were in the lowest USSC criminal history category, Category I, which includes offenders with no previous term of imprisonment and others with minimal criminal histories. This was the most frequently applied category (table 4). Almost 8 in 10 (79%) federal drug offenders in Category I had no points assigned at sentencing, meaning they had no relevant criminal history as used to compute categorization. This accounted for more than 25% of all drug offenders in the BOP. In comparison, more than 22% of federal drug offenders were in the highest criminal history category (Category VI), which indicates the most extensive criminal histories. ************************************************* *************************************** USSC Criminal History Categorization *************************************** The U.S. Sentencing Commission (USSC) aggregates available criminal history information into broader criminal history categories to be used at sentencing. This information includes whether the offender has been previously sentenced to a term of imprisonment, the length of each previous sentence imposed, whether the most recent offense was committed soon after a release from incarceration, whether the most recent offense was committed while under criminal justice supervision (e.g., probation), and whether there were any crimes of violence that were not previously captured in any of the previous other categories. ************************************************* Criminal history points are assigned in differing levels for each of these categories, and the points are collapsed into six levels of criminal history—zero points indicates that there is no history in any of the above domains, and higher point totals reflect more extensive criminal histories. Criminal history category may also be adjusted based on other aggravating circumstances, such as being a career offender or a repeat and dangerous sex offender against minors. Judges may depart upward or downward regarding the criminal history category if the final category substantially misrepresents the seriousness of the offender’s criminal history.*** Footnote 3 For more information, see Office of the General Counsel, U.S. Sentencing Commission, Criminal History Primer, Washington, DC: U.S. Sentencing Commission. 2014***. ************************************* Crack cocaine offenders had the most extensive criminal histories *********************************** The extent of criminal history for drug offenders varied by primary drug type. Crack cocaine offenders had more extensive criminal histories than any other type of drug offenders, and more than 40% of crack offenders in federal prison were in the highest criminal history category (Category VI), followed by heroin offenders (24%). The proportion of offenders in Category I ranged from 39% for heroin offenders to 49% for powder cocaine offenders. Within each drug type, the majority of offenders in Category I had no criminal history points at the time of sentencing. Across all drug offenders, powder cocaine (41%) and marijuana (36%) offenders were the most likely to have no criminal history points applied at sentencing and be in Category I. Almost a quarter of heroin offenders (24%) were classified under the most serious criminal history category (Category VI), the second highest share of any drug type after crack cocaine. *********************************** About a quarter of federal drug offenders were sentenced for weapons involvement *********************************** Nearly 1 in 4 (24%) federal drug offenders in prison at yearend 2012 who were sentenced since 1998 were sentenced for having a weapon involved in the instant offense (table 5).***Footnote 4 Although USSC datasets go back to 1994, some weapons variables are unavailable prior to 1998. To ensure continuity within comparisons of weapons, offenders sentenced prior to 1998 were excluded***. Of these drug offenders, about a quarter (26%) received a mandatory minimum sentence for a separate charge of conviction for using a weapon during the commission of the offense (pursuant to 18 U.S.C. § 924(c)); and about three- quarters (73%) had a weapon specific offense characteristic that was not a separate charge of conviction but was still included in the sentence computation (not shown). A majority of the weapon- involved offenders received a sentence enhancement, accounting for 18% of the total drug offender population. Another 6% of the drug offender population were convicted of a 924(c) weapon offense.***Footnote 5 All figures reported for weapon involvement exclude offenders who have both types of weapon; they made up less than 0.5% of drug offenders***. *********************************** Less than a third of crack cocaine offenders had their sentence increased due to weapons involvement in the offense *********************************** Crack cocaine offenders were the most likely to receive a mandatory minimum sentence for weapon involvement under 924(c) (9%) or to have guidelines adjusted for a weapon-specific offense characteristic (23%) totaling to almost a third with weapon involvement (32%). Methamphetamine offenders were the next most likely to have weapon enhancements—more than 25% had some kind of sentencing related to weapons. In comparison, marijuana offenders were the least likely to have these weapon enhancements—3% had a mandatory minimum sentence under 924(c) for gun involvement and 12% had a specific offense characteristic. *********************************** The average prison sentence of drug offenders in federal prison was 11.3 years *********************************** The total prison term imposed on federal drug offenders was more than 11 years on average. The median prison term was 10 years. Almost two-thirds (65%) of offenders were sentenced to a term of imprisonment of more than 5 years up to and including 20 years. Thirteen percent of federal drug offenders were sentenced to a term of imprisonment greater than 20 years, while less than 1% were sentenced to a term of 1 year or less (table 6). *********************************** Marijuana offenders had the shortest prison terms *********************************** Marijuana offenders had the shortest average term of imprisonment of all drug types. The average sentence was more than 7 years (88 months) and the median was 5 years (60 months). A majority of marijuana offenders received a sentence of more than 1 year up to and including 5 years in prison. In comparison, crack cocaine offenders had the longest average and median prison terms. On average, these offenders were sentenced to more than 14 years (170 months), while the median sentence was more than 12.5 years (151 months). A majority (62%) of crack cocaine offenders were sentenced to more than 10 years in prison. ************* Methodology ************* These analyses are based on a data file built from the population of new U.S. district court commitments in the Federal Bureau of Prisons (BOP) at fiscal yearend 2012. These prisoners made up 87% of the federal prison population at that time. Pre-sentencing admissions, offenders sentenced prior to the Sentencing Reform Act of 1984, and others who were not new U.S. district court commitments (e.g., supervision violators, D.C. Superior Court commitments, or state boarders) were excluded. For more information about offenders who were included or excluded for the purposes of linking, see Dataset composition and definition of drug offender below. For the remaining 188,988 prisoners, probabilistic linking methods were used to link each BOP record with a U.S. Sentencing Commission (USSC) record corresponding to that prisoner’s shared identifying information. USSC files from FY 1994 to FY 2012 were used for these purposes. Linking the BOP and the USSC datasets is useful because the BOP does not publicly provide certain data fields that are available in the USSC dataset, including criminal history information, offense characteristics, and sentencing information (e.g., whether an offender was convicted of an offense carrying a mandatory minimum penalty and whether that penalty was applied at sentencing). For drug offenders, the USSC also provides information about up to five drug types involved in the instant offense (the most recent offense for which the individual was sentenced and incarcerated), along with the weight of each type. This report uses the USSC primary drug type data. Probabilistic matching techniques were applied across personally identifiable information such as name, date of birth, and sentencing year to link records between the two different data sources.***Footnote 6 For more detailed information about the linking methodology employed, see Kelly, J. (2012). Federal Justice Statistics Program Data Linking System (NCJ 239536). Washington, DC: Urban Institute***. Of the 188,988 federally sentenced prisoners included in the initial dataset, 97% had valid matches in the USSC data (table 7). There are many reasons why a person may not be matched to a USSC record. The two primary technical reasons are (1) a person is in prison for Class B or C misdemeanors not covered by the U.S. Sentencing Guidelines ***Footnote 7 The U.S. Sentencing Guidelines do not apply to Class B and C misdemeanor offenses handled by U.S. magistrate judges; however, these offenders may receive short prison terms and therefore would be included in the BOP data***. and (2) a person was sentenced pursuant to the U.S. Sentencing Guidelines but prior to 1994, when the Bureau of Justice Statistics (BJS) introduced the Federal Justice Statistics Program standard analysis files and dyad link files. A Jaro-Winkler score threshold of 0.70 was used by the matching algorithm; names that matched below that level were excluded from this dataset. A comparison of linked and unlinked records by primary offense category shows that two offense types--violent offenses and immigration offenses--were over-represented in the unlinked population relative to the linked population (table 8). This is consistent with possible reasons for exclusion, as violent offenders typically receive very long sentences (certain violent offenders sentenced to long prison terms prior to 1994 will be unlinked in this dataset) and a significant share of immigration offenders may be Class B and C misdemeanants not sentenced in U.S. district court pursuant to the federal sentencing guidelines, and therefore would not be included in the USSC dataset or in the linked file. ************************************************* *********************************** Dataset composition and definition of drug offender *********************************** This report focuses on the subset of 94,678 federally sentenced prisoners in the Federal Bureau of Prisons (BOP) on a new U.S. district court commitment whose most serious instant offense (as classified by the BOP) was a drug offense, and who had valid links to an observation in the United States Sentencing Commission (USSC) database. The universe of offenders analyzed in this report was a product of (1) the linking methodology and (2) the definition of drug offenders used in the report. The base dataset used for linking includes only federally sentenced offenders in BOP for a new U.S. district court commitment. Offenders serving time in prison for a violation of probation or supervised release were excluded from the dataset, although they were sentenced at an earlier time. Violators are different from new commitments, differing in terms of age, race, sex, and citizenship (table 9). Many noncitizens are not released back to the United States and are therefore not actively supervised by federal probation, thus accounting for the lower proportion of noncitizens among supervision violator returns to prison as compared to new court commitments. ************************************************* This report uses the BOP’s classification of most serious offense, which differs from that of the USSC, to define drug offenders in federal prison. In comparison, the USSC uses the statutory maximum to determine the most serious offense. As a result, this report does not include all persons who have committed a federal drug offense and instead focuses on prisoners whom BOP classified as drug offenders because the drug offense carried the longest sentence imposed. To identify drug offenders for this analysis, BOP’s most serious offense categorization (which relies on the longest sentence imposed) was used. Alternative methods of identifying drug offenders in BOP custody could have expanded the definition to include secondary charges or to rely on USSC’s categorization of most serious offense, which primarily depends on the statutory maximum. The USSC criteria would have yielded a slightly different population of prisoners considered to be drug offenders (table 10). Another definition of drug offender could be any individual serving time for a drug offense, rather than any offender whose most serious offense is drug related. This would include other offenders who have more serious primary charges, as categorized by both the BOP and USSC, for which they were sentenced in addition to a secondary or tertiary drug charge. This criterion would have yielded 102,012 persons who were serving time for a drug offense, although that offense may not have been the most serious charge of conviction. Most of the remaining offenders had a weapons offense as the most serious offense (table 11). ******************************************************** The Bureau of Justice Statistics of the U.S. Department of Justice is the principal federal agency responsible for measuring crime, criminal victimization, criminal offenders, victims of crime, correlates of crime, and the operation of criminal and civil justice systems at the federal, state, tribal, and local levels. BJS collects, analyzes, and disseminates reliable and valid statistics on crime and justice systems in the United States, supports improvements to state and local criminal justice information systems, and participates with national and international organizations to develop and recommend national standards for justice statistics. William J. Sabol is director. This report was written by Sam Taxy, Julie Samuels, and William Adams of Urban Institute. Mark Motivans and Rhonda Keith provided statistical verification and review. Jill Thomas edited the report. Tina Dorsey produced the report. October 2015, NCJ 248648 ******************************************************** ************************************************* Office of Justice Programs Innovation * Partnerships * Safer Neighborhoods www.ojp.usdoj.gov ************************************************* ********************************************** 10/22/2015 9:25pm JER **********************************************