Assessing disparity in the federal court processing of immigration cases

In recent decades, the United States has experienced a wave of immigration, an economic recession, and several terroristic attacks. In response, the government has scapegoated and blamed undocumented immigrants of color for recent social ills. As a result, a large share of government resources has been allocated to the enforcement and processing of immigration violations. Consequently, the number of immigration cases processed in U.S. federal courts has spiraled to nearly 50% of bookings and 34% of federal sentencing cases. Yet, immigration offenses have received little empirical attention in the courts and sentencing literature due in part to differences in the way immigration offenses are processed compared to other federal offense types, and relatedly, the empirical difficulties immigration offenses pose for analysis. Nevertheless, the increased representation of immigration offenses in federal courts, along with the punitive rhetoric and heightened social control targeting undocumented immigrants of color, warrants a comprehensive assessment of how immigration cases are processed in U.S. federal courts. Accordingly, this dissertation seeks to identify inequality in the processing of immigration cases by examining: 1) cumulative disadvantage within immigration cases; 2) contextual disparity and how social context interacts with ethnicity to influence multiple federal court outcomes within immigration cases; and 3) ethnic disparity within immigration cases over time.

Data come from the Federal Justice Statistics Program Data Series, the U.S. Census, the Uniform Crime Reports, Leip’s Atlas of U.S. Presidential Elections, the National Judicial Center, and the U.S. Department of Justice. The quantitative analysis addresses the first question by employing a cumulative disadvantage approach where multiple decision points are considered and the effects of prior stages on subsequent outcomes. The quantitative analysis proceeds to address the second question by using multilevel modeling for multiple court outcomes. The longitudinal analysis is separately conducted on sentence length for 18-year data, from 1994 through 2012, to assess racial and ethnic disparity over time.

The results indicate that cumulative disadvantage is present within immigration cases, that social context influences certain decision points, and that ethnic disparity has diminished over time in some districts.