Gender In-Group Bias: Evidence from Judicial Decisions

Abstract

How does gender affect judicial decisions? Whether and to what extent judges have preferential treatment to plaintiffs of the same gender? In this paper, we provide novel evidence of gender in-group bias in judicial decisions. We use a naturally occurring dataset of online transcripts of judicial decisions and exploit effectively random assignment of cases to judges. The analysis provides evidence in support of gender bias: a claim with a male plaintiff is 3.6\% more likely to be accepted if it is assigned to a male judge, as opposed to a female judge. The bias is slightly larger in regions with initially more biased gender attitudes. Judges' working experience or workload does not affect the magnitude of bias. We provide evidence in support of in-group gender bias explanation and rule out the possibility that the results are driven by differences in the information available to the judge or differences in plaintiffs' behaviors. These findings raise policy concerns about the fairness of trials in courts with a small proportion of female judges.