Working Paper: CEPR ID: DP14736
Authors: Imran Rasul; Brendon McConnell
Abstract: A vast literature uses ingroup biases to explain animus towards others. The notion can be extended to multi-identity societies, where social preferences are defined over one ingroup and multiple outgroups. We use a novel research design to recover the structure of social preferences across outgroups in a high stakes setting. We investigate whether increased animosity towards Muslims post 9-11 had spillover effects on Black and Hispanic individuals in the federal criminal justice system. Using linked administrative data tracking defendants from arrest through to sentencing, we find that as 9-11 increased animosity towards Muslims, sentence and pre-sentence outcomes for Hispanic defendants significantly worsened. Outcomes for Black defendants were unchanged. We underpin a causal interpretation of our findings by providing evidence to support the identifying assumptions underlying the research design. The findings are consistent with judges and prosecutors displaying social preferences characterized by contagious animosity from Muslims to Hispanics. To understand why increased animosity towards Muslims post 9-11 could spillover onto Hispanics, we draw on work in sociology to detail how Islamophobia and immigration have become intertwined in American consciousness since the mid 1990s, but were forcefully framed together in the aftermath of 9-11. We narrow the interpretation of the results as being driven by social preference structures using decomposition analysis, and correlating sentencing differentials to judge characteristics, including their race/ethnicity. Our findings provide among the first field evidence of contagious animosity, so that social preferences across outgroups are interlinked and malleable
Keywords: No keywords provided
JEL Codes: D91; J15
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| increased animosity towards Muslims post-9/11 (F52) | worsened sentencing outcomes for Hispanic defendants (K37) |
| increased animosity towards Muslims post-9/11 (F52) | increase in sentence length for Hispanic defendants (K40) |
| increased animosity towards Muslims post-9/11 (F52) | Hispanic defendants more likely to receive initial offenses carrying statutory minimums (K37) |
| worsened sentencing outcomes for Hispanic defendants (K37) | contagious animosity from Muslims to Hispanics (J15) |
| Hispanic defendants charged post-9/11 (K37) | statutory minimum sentence longer than whites (K14) |