The Effect of Prison Sentence Length on the Subsequent Employment and Earnings of Criminal Defendants

This paper examines the employment and earnings of people convicted of committing serious crimes, focusing on the effects of serving any time in prison and of the length of time served on long-term labor market outcomes. Regression analyses control directly for some of the most important factors that determine sentences (such as criminal history and offense type) and labor market outcomes (such as education, experience, demographic characteristics, and earnings history). An instrumental variables approach identifies the causal effect by essentially comparing otherwise similar groups of criminal defendants whose labor market outcomes differ because of the systematically difference sentencing decisions of the judge to whom their case was randomly assigned. In order to implement these methods, new data were created by linking information about criminal defendants in California federal district court felony cases from 1983-94 with quarterly earnings data collected through the California Unemployment Insurance system from 1987-97. The results show that incarceration has surprisingly little effect on employment in comparison to those who are not incarcerated, with employment rates only 0-3% lower after 5 to 8 years for those who served prison time. Further, employment rates for those with longer sentences rebound just as quickly to pre-conviction levels as do those with similar characteristics but shorter sentences. Negative earnings effects are more pronounced and are concentrated among white-collar criminals, who earn 10-30% less after 5 to 8 years than those who were convicted at the same time but were not incarcerated. Violent drug offenders have very low earnings in the legitimate sector overall, but these earnings appear to increase over the long-term after release from prison and do not vary with the length of time served.