By Stephen Koppel & René Ropac

The current study isolates the effect of what is arguably the most consequential bail reform provision: eliminating the option to set bail or detain people for most misdemeanor and nonviolent felony charges. The study uses what is known as a “difference-in-differences” causal design—comparing the change in re-arrest rates from before to after initial bail reform implementation among charges seeing the elimination of bail versus charges remaining legally eligible for bail.

What Did We Find?

• Pretrial Recidivism: During the brief pretrial period (capped at 6 months for all cases), eliminating bail had no overall effect on recidivism. However, recidivism increased among a small high-risk group with a pending case.

• Two-Year Recidivism: Over a longer two-year follow-up—including the period both before and after a case disposition—results grew more favorable to bail reform. Charges seeing the elimination of bail had significantly lower felony re-arrest rates than charges still exposed to bail and detention. In addition, there was no longer evidence of a recidivism increase for the “high-risk” subgroup (or any other subgroup).

What's the Upshot?

Our latest study adds to a growing body of research analyzing the effects of New York’s bail reform on public safety. Short-term recidivism increases appear limited to a small high-risk subgroup, with the current study indicating that such increases were no longer present when extending the follow-up period to two years. 

Meanwhile, considering all five of DCJ’s recidivism studies, a pattern emerges that, overall, expanding pretrial release under bail reform reduced recidivism in New York City—especially over a long-term 50-month tracking period—while having no clear effect in suburban and upstate regions. Each prior DCJ study (two in New York City, one outside the City, and a statewide study released last month) reported these overall effects, while adding more nuanced results for key subgroups of interest.