How much would eliminating drug crimes decrease racial/ethnic gaps in criminal conviction?

Since the 1970s, criminal justice contact has become an increasingly common event in early adulthood, and disproportionately so for African American men. Policymakers often argue that reducing drug-related conviction rates is among the easiest ways to reduce racial/ethnic disparities in incarceration. These arguments are often backed by statistics that convey the number of drug offenders in contact with the criminal justice system at a given point in time. Unfortunately, we know little about the extent to which over-time conviction risk and associated racial/ethnic disparities may be affected by drug-related policy changes. Using a novel application of the single decrement life table to analyze data from the National Longitudinal Survey of Youth 1997 (NLSY97), I present a quantitative thought experiment to consider the extent to which the elimination of drug-related offenses would affect racial/ethnic disparities in men's cumulative probability of conviction by age 30. Consistent with prior research, results indicate that black men are at disproportionately higher risk of ever experiencing a drug-related conviction, and of experiencing a drug-related conviction at each conviction instance. More surprising, however, is the finding that while the removal of drug sentencing may significantly impact racial/ethnic disparities associated with conviction, only a relatively small proportion of those ever convicted would avoid conviction altogether in the absence of drug-related sentencing.