Abstract

The California Racial Justice Act provides a novel basis for challenging racial disparities in charging, conviction, and sentencing, even in the absence of explicit intent to discriminate. However, the lack of accessible data demonstrating a “significant difference” in outcomes for “similarly situated” defendants across racial groups has hindered the Act’s implementation. This paper introduces Paper Prisons Racial Justice Act data tool (www.paperprisons.org/rja) to address this gap by providing, in its first release, comprehensive aggregate criminal offender record information (CORI) data from the California Department of Justice on racial disparities at the county, offense, and stage-of-conviction levels from 2010 to 2021. The paper describes the provenance, processing, and production of this data. The tool covers all misdemeanor and felony offenses in California and provides statistical information for five racial groups. Using this comprehensive data, we examine how courts should interpret the RJA’s requirement that comparisons be made among “similarly situated” defendants in the “same county.” We find that for many offenses, especially in smaller counties, and for less populous racial/ethnic groups, and for later stages of prosecution, there are often insufficient cases to meet privacy-preserving minimum sample size requirements for release of the data. For example: for 16 counties with small Black populations, not a single offense has a large enough number of arrest incidents in 2019 for the data to be reportable. Moving to later in the criminal cycle, we find that no comparisons to Black defendants with respect to felony convictions are possible for over 75% of counties. The data situation is markedly worse for Native Americans. Only five counties have enough incidents to cover more than half of the incident arrests. If the RJA is to achieve its goal of eliminating racial bias statewide, its evidentiary standards must be interpreted flexibly and allow for the aggregation of data, for example, across geographies, time periods, and offense categories when necessary to accrue adequate sample sizes. The language of the statute and early court decisions suggest such flexibility is permissible.