Statistical Discrimination in the Criminal Justice System: The Case for Fines Instead of Jail

Abstract

We develop a model of statistical discrimination in criminal trials. Agents carry publicly observable labels of no economic significance (race, etc.) and choose to commit crimes if their privately observed utility from doing so is high enough. A crime generates noisy evidence, and defendants are convicted when the realized amount of evidence is sufficiently strong. Agents may also be convicted even when no crime has occurred. Convicted offenders are penalized either by incarceration or by monetary fines. In the case of prison sentences, discriminatory equilibria can exist in which members of one group face a prior prejudice in trials and are convicted with less evidence than members of the other group: If income is inversely related to prejudice, prison sentences have a lesser deterrence effect on the disadvantaged group whose members will thus commit more crimes, justifying the initial prejudice. Applying this argument to lifetime income, it is even possible that all individuals earn the same wage, but racial differences persist in the crime rate because the expected duration of employment is less for persons who face a higher prejudice (because they will be jailed more often). Such discriminatory equilibria cannot exist with monetary fines instead of prison sentences, as they represent a stronger penalty for persons with lower earnings. Our findings have implications for potential reforms of the American criminal justice system.

Keywords: Statistical discrimination, criminal justice, prejudice, deterrence, fines, imprisonment

JEL Classification: D72, D78