Political Accountability and Stereotyping in Prosecutions

• In Event: Government Accountability to Citizens

Abstract

Racial disparities in the criminal justice system have been well documented. In the political economy literature, two standard theoretical explanations are given for these disparities. In the first, disparities are driven by the preferences of criminal justice officials (e.g. animus or favoritism). In the second, disparities are driven by the beliefs of criminal justice officials about the prevalence of crime or the guilt of criminal suspects. The latter explanation---often termed "statistical discrimination"---hinges on the kind and amount of information available to officials. In this paper, I study statistical discrimination by criminal prosecutors, whose successes and failures in court depend on their costly investigations of the "facts on the ground." A key feature of modern prosecutors' offices is their hierarchical structure: while a jurisdiction's lead prosecutor is often elected, most rank-and-file staff members, such as assistant prosecutors, are not. Assistant prosecutors typically answer to the lead prosecutor, who may or may not share their objectives. I analyze a formal model in which criminal cases are assigned to low-level assistant prosecutors and come with some rudimentary information about the defendant and the crime being prosecuted. The official decides whether to invest time and energy to increase their certainty about the defendant's guilt, which can reduce the likelihood of convicting an innocent person or acquitting a guilty person. However, their decisions are subject to review by the lead prosecutor who uses her discretion to intervene in cases that are important for her own political objectives. I show that the costliness of a rank-and-file prosecutor's investigation, combined with the political incentives created by the hierarchical organization, can generate underinvestment in investigating cases and cause prosecutors to rely more heavily on stereotypes about defendants. As a result, the model demonstrates how institutions can reinforce the persistence of inefficient stereotypes in the criminal justice system.

Author

• Ryan Hubert, University of California, Davis