Electoral Incentives, Race, and Judicial Decision Making - Evidence From One Million Criminal Cases

Abstract

The majority of states in the U.S. select judges via popular partisan elections. Do such elections induce judges to modify their sentencing, and if they do, are the effects concentrated among certain classes of defendants? Leveraging rich case-level data on all criminal proceedings between 1998 and 2020 from a populous county, I find that up- coming elections cause judges to sentence more harshly, with incarceration length tem- porarily increasing by up to 11.8%. Surprisingly, the majority of the impact is driven by the treatment of white defendants. I further show that type of election and judge party affiliation matter, suggesting the importance of the specific incentive structure faced by a judge. I also investigate and rule out alternate channels, such as manipu- lation of case timing and changes in prosecutor behavior. Overall, I show that judge elections cause substantial and arbitrary inequities between individual defendants, and that the results suggest personal biases likely contribute to the large baseline differences we observe between defendants of different races.