Variation in the Effects of Race, Ethnicity, and Immigration Offenses on Federal Sentencing Decisions, 1999-2012

• In Event: Race, Ethnicity, and Criminal Sentencing

Abstract

The focus of this study is a test of whether the effects of offender race and ethnicity are constant across the full distribution of sentence length decisions. I analyze federal sentencing data from 1999 thru 2012 using quantile regression models, which allow for testing whether offender and case characteristics are constant (or variable) across the full distribution of sentence length decisions. I rely on typical offender and case characteristics, consistent with other research on federal sentencing decisions. There are four key findings regarding the effects of race, ethnicity and immigration offenses. First, the black-white differential in sentence length generally increases in magnitude while moving from less severe to more severe punishments. Second, the Hispanic-white differential changes direction – among less severe sentences, Hispanics are punished more severely than whites, but among more severe sentences, whites are punished more severely than Hispanics. Third, immigration offenses have a suppression effect on race-ethnicity differentials in sentence lengths among less serious sentence lengths, but among more serious sentence lengths, the inclusion of immigration offenses enhances the magnitude of the race-ethnicity differentials. Fourth, there is evidence of temporal variability in the effects of race and ethnicity, but no consistent trends of increasing or decreasing disparities.

Authors

• Chester Britt, Northeastern University

• Emmanuelle Klossou, Stevenson University