Presentence detention in U.S. District Courts: the effects of race/ethnicity, gender, and social context

Abstract

Despite the substantial research conducted on racial and ethnic disparities in the criminal justice system, the majority of research focuses on final sentencing decisions. Less attention has been devoted to earlier stages in criminal processing, such as the presentence detention process. In fact, the analyses that did investigate presentence detention status mainly used state-level data, and there is only a handful of studies that have examined race/ethnicity and gender effects on detention and bail in the federal criminal justice system (e.g., Albonetti, 1989; Bak, 1998: Reitler, et. al., 2012; Spohn, 2009). Furthermore, recent analyses have uncovered that extralegal disparities in presentence detention outcomes, particularly racial and ethnic disparities, vary across court jurisdictions (e.g., Levin, 2008; Pinchevski & Steiner, 2013; Eisenstein & Jacob, 1977). Thus, the purpose of this study is to explore the effects of defendants’ race/ethnicity and gender, and social context, on defendants’ pretrial status. This study employs a multilevel modeling strategy and uses criminal sentencing data (N=130,120) from the U.S. Sentencing Commission (USSC) for fiscal years 2008 to 2010, across 89 U.S. District Courts, including supplemental data drawn from the U.S. Census Bureau and the Federal Judicial Center’s Federal Court Management Statistics. The initial findings indicate that Black and Hispanic defendants are less likely than Whites, and females are more likely than males, to be released on bail and rereleased on their own recognizance (ROR). Regarding the district level, ROR is less likely for defendants processed in districts with a higher crime rate and higher socioeconomic disadvantage. On the other hand, bail is more likely for defendants processed in districts with larger courts (more judges). In cross-level interactions increase in percent Black, increase in crime rate, and increase in socioeconomic disadvantage are positively associated with bail for Black defendants, while increase in percent Hispanic and court size are positively associated with bail for Hispanic defendants. Consistent with the courts as communities and focal concerns perspectives, this study found that the presentence process is influenced by local practices, norms, and concerns for organizational efficiency. These distinctive features of court jurisdictions interact with race/ethnicity and affect defendants' presentence status. Future research is needed to better assess the detention process with increased access to federal data. In addition, examination of intersectionality of race/ethnicity, gender and age at the presentence detention stage is required.