Federal criminal sentencing: race-based disparate impact and differential treatment in judicial districts

Race-based inequity in federal criminal sentencing is widely acknowledged, and yet our understanding of it is far from complete. Inequity may arise from several sources, including direct bias of courtroom actors and structural bias that produces racially disparate impacts. Irrespective of these sources, inequity may also originate from different loci within the federal system. We bring together the questions of the sources and loci of inequity. The purpose of our study is to quantify race-based disparate impact and differential treatment at the national level and at the level of individual federal judicial districts. We analyze over one-half million sentencing records publicly available from the United States Sentencing Commission database, spanning the years 2006 to 2020. At the system-wide level, Black and Hispanic defendants receive average sentences that are approximately 19 months longer and 5 months longer, respectively. Demographic factors and sentencing guideline elements account for nearly 17 of the 19 months for Black defendants and all five of the months for Hispanic defendants, demonstrating the disparate impact of the system at the national level. At the individual district level, even after controlling for each district's unique demographics and implementation of sentencing factors, 14 districts show significant differences for minoritized defendants as compared to white ones. These unexplained differences are evidence of possible differential treatment by judges, prosecutors, and defense attorneys.