By Felix Owusu

The Suffolk County District Attorney’s Office (SCDAO) has taken steps to limit its use of criminal sanctions for individuals charged with nonviolent offenses, including identifying 15 common charges that arraigning Assistant District Attorneys (line ADAs) should presumptively decline to prosecute (DTP) or divert when possible. These changes, driven by policies introduced by District Attorney (DA) Rachael Rollins after her inauguration in January 2019, reflect mounting evidence that relying on policing, criminal adjudication, and incarceration to address non-violent offenses is costly, exacerbates racial disparities, and is often ineffective at improving public safety. Others are concerned, however, that increasing leniency for even minor offenses will embolden people who commit crimes and lead to more serious misconduct. Below I attempt to analyze the impact of these policy changes on case adjudication as well as future offending behavior for those whose cases were impacted. I primarily rely on administrative data from the SCDAO’s internal case management database as well as criminal records from the Department of Criminal Justice Information Services (DCJIS). Using an event study design, I find that after DA Rollins' inauguration, prosecution rates declined by roughly 5 percentage points on average for cases consisting of charges included in the declination and diversion policy (DTP list cases) and nearly 10 percentage points for cases involving nonviolent misdemeanors more generally. These average decreases mask substantial heterogeneity by offense category and defendant race. Prosecution rates declined substantially less for cases involving Black defendants, and decreases in prosecution rates were concentrated among a subset of DTP list offenses like driving with a suspended license. Consistent with the scope of the policy, prosecution rates for cases involving felonies or violent offenses were not similarly impacted. I also explore the impact of the declination and diversion policy on reoffending using a difference-in-differences design to account for unobserved factors that could impact recidivism throughout the policy's implementation. Consistent with past research, I find that the introduction of the declination and diversion policy was associated with small (although statistically indistinguishable from zero) decreases in overall reoffending and violent reoffending.  

Boston:The Rappaport Institute for Greater Boston at Harvard’s Kennedy School of Government, 2022.