Essays on Criminal Justice Reform and Racial Inequality

Abstract

This dissertation contains essays on criminal justice reform and racial inequality in the US. In Chapter 1, I study the money bail system, a system that requires people to post financial collateral for pretrial release from jail. Money bail advocates argue that its usage is critical for averting misconduct, while skeptics counter that its effects on misconduct are small and not worth the human costs of pretrial detention. To test these claims, I investigate the effects of a program in Kentucky that reduced the use of money bail for low-level offenses by over 60%. The reform meaningfully increased pretrial release from jail due to lower financial requirements. The program increased failure to appear in court by 3.3 percentage points and had no detectable effect on pretrial rearrest, with the data ruling out even modest sized increases. Taken together, my results imply that one instance of pretrial misconduct needs to be at least 18 times as costly as one day in detention for money bail to justify its social costs. In Chapter 2, I study algorithmic recommendations in the criminal justice system. Algorithmic recommendations recommend actions to human decision makers based on algorithmic predictions. These recommendations are prevalent in high-stakes decisions but are rarely studied separately from the underlying prediction technology. In this essay, I study the effects of algorithmic recommendations on judge bail decisions using a policy change that added recommendations to a preexisting risk scoring system. First, I provide evidence that recommendations can change the costs of errors to judges. Judges become more lenient when lenient recommendations are present and do not offset this behavior by being more harsh in their absence. Second, I provide evidence that algorithmic recommendations do not uniformly change the cost of errors to judges. Judges adhere to lenient recommendations more frequently for white defendants than for Black defendants with identical underlying risk scores. In effect, discretionary responses to algorithmic recommendations can have unintended effects, such as widening racial gaps. In Chapter 3 (with Jeremy Cook, James Feigenbaum, Laura Kincaide, Jason Long, and Nathan Nunn), I study the 1921 Tulsa Race Massacre, which resulted in the looting, burning, and leveling of 35 square blocks of a once-thriving Black neighborhood. Not only did this lead to severe economic loss, but the massacre also sent a warning to Black individuals across the country that similar events were possible in their communities. I examine the economic consequences of the massacre for Black populations in Tulsa and across the United States. I find that for the Black population of Tulsa, in the two decades that followed, the massacre led to declines in home ownership and occupational status. Outside of Tulsa, I find that the massacre also reduced home ownership. These effects were strongest in communities that were more exposed to newspaper coverage of the massacre or communities that, like Tulsa, had high levels of racial segregation. Examining effects after 1940, I find that the direct negative effects of the massacre on the home ownership of Black Tulsans, as well as the spillover effects working through newspaper coverage, persist and actually widen in the second half of the 20th Century.