Effects of Automatic Criminal Record Expungements on Employment

Abstract

Several states have enacted broad automatic expungement laws, such as Clean Slate and expungement for cannabis-related offenses, that destroyed millions of criminal records. Advocates of the automatic expungement laws propose that removing criminal record information might help people with records to improve their economic outcomes by removing barriers to employment. However, this policy might have adverse effects on disadvantaged demographic groups. When risk-averse employers realize that there are many people in the labor market whose criminal background cannot be observed because of automatic expungement, they might hesitate to hire job applicants from demographic groups that are likely to include the majority of ex-offenders, particularly black people with no sort of college education. I test this hypothesis by exploiting the adoption and timing variation of the automatic expungement policies across states. I apply the difference-in-differences and event-study approaches as my identification strategies using individual-level monthly CPS (Current Population Survey) data. The results show that the automatic expungement laws decrease the probability of employment by 3.99 percentage points (-7.79%) for low-educated black people. The magnitude of the effect is higher when I restrict the sample to young black individuals with no high school diploma. Their probability of employment is reduced by 10.8 percentage points (-27.69%).

Keywords: Clean Slate, Expungement, Employment, Criminal Record, Statistical Discrimination, Race

JEL Classification: J08, J71, J78, K4