The U.S. criminal legal system is notoriously racialized. Though Black and Latinx people make up less than 30% of U.S. residents, they constitute more than 50% of the nearly two million people currently in U.S. prisons and jails. For decades, research has indicated that one group of decision-makers has had an outsized influence on these numbers: prosecutors. From whom to charge to what sentences to recommend, no actor plays a greater role in determining who goes to prison in this country. Highly subjective and lacking in formal guidance and accountability, prosecutorial decisions are especially vulnerable to racial bias. They are also cloaked in secrecy. Data about how and why prosecutors make decisions often does not exist or is shielded from public view. As a result, it has been nearly impossible to determine the extent to which prosecutors' decisions contribute to racial disproportionality in the criminal legal process, let alone whether such decisions are the product of racial bias. This Essay argues that prosecutors' offices must collect, maintain, and publish standardized data on the bases of their charges, declinations, plea offers, and resolutions if we are to ever understand and address vectors of racial bias in the criminal legal system. Contextualizing this "call for data" within two case studies-one on the racialized impact of felony murder and accomplice liability murder laws and the other on the California Racial Justice Act-we demonstrate how prosecutorial data transparency would enable researchers, advocates, and policymakers to better identify and remediate racial bias in decision-making. Data transparency would also promote prosecutorial accountability both internally and externally. Legislative efforts to implement data transparency must address privacy and surveillance concerns, especially since prosecutorial data transparency would expand a carceral source of information. At the same time, the consequences of data opacity are already shaping case outcomes. In this way, data transparency provides one remedy for currently unchecked systems, and serves as a step towards data justice.
