New Yorkers released under the state’s bail-reform law were re-arrested less often than comparable defendants who came through the system before reform, according to long-term studies from the Data Collaborative for Justice at John Jay College of Criminal Justice released in the fall of 2025 — findings that cut against the central claim of the law’s critics.

The research, drawing on New York State Division of Criminal Justice Services (DCJS) and Office of Court Administration data, tracked defendants for roughly 50 months and used quasi-experimental methods to compare similar cases on either side of the 2020 reforms.

What the New York City data showed

In New York City, defendants released under bail reform had significantly lower rates of re-arrest across every category the researchers measured. Over the 50-month follow-up, overall re-arrest was 57% for the reform group versus 66% for the comparison group. Felony re-arrest was 33% versus 40%, and violent-felony re-arrest was 20% versus 25%.

In each case, the people who would previously have been held on bail — but were released under the new rules — were re-arrested less frequently, not more, over a multi-year horizon.

The statewide picture

A companion DCJ analysis covering suburban and upstate regions, built on nearly 18 months of statewide court data, reached a consistent conclusion: New York’s initial bail-reform law did not increase overall recidivism, and in some places modestly reduced long-term reoffending. Statewide, overall re-arrest rates fell about 2%, including a statistically significant 2.7% reduction in felony re-arrest.

The pattern was not uniform across every defendant. Researchers found reform tended to reduce recidivism for people facing less serious charges and with limited or no recent criminal history, while tending to increase it for those facing more serious charges with recent records. Crucially, the short-term re-arrest bumps seen among higher-risk releasees were concentrated in the first months after release and did not persist over the two-year horizon — a finding that complicates simple “reform increased crime” narratives.

Why it matters in the 2026 landscape

Bail reform has been one of the most contested issues in New York politics since it took effect, blamed by opponents — and some prosecutors and police officials — for a perceived revolving door, and defended by reform advocates as a constitutional correction to a system that jailed people simply because they couldn’t pay. The Legislature has amended the law several times since 2020 in response to that pressure.

The DCJ findings arrive as the broader crime picture has improved markedly: NYPD reported its safest year ever for gun violence and a 16-year low in subway crime in 2025. Disentangling how much of that improvement is attributable to policing, the post-pandemic environment, or pretrial policy is precisely the kind of question the John Jay methodology was designed to probe by holding case characteristics constant.

The studies do not settle the policy debate on their own. They are observational analyses of administrative data, not randomized experiments, and the authors are careful to note the limits of subgroup findings. But they add substantial empirical weight to the conclusion that, at the system level, releasing more defendants pretrial in New York did not produce the surge in reoffending its critics predicted — and in New York City, was associated with measurably lower re-arrest.

Verification

Frequently Asked Questions

Who conducted the study?
The Data Collaborative for Justice (DCJ) at John Jay College of Criminal Justice, using New York State DCJS and Office of Court Administration data. The findings were released in reports in October and November 2025.
What were the New York City results?
Over a roughly 50-month follow-up, NYC defendants released under bail reform had lower rates of overall re-arrest (57% vs 66%), felony re-arrest (33% vs 40%) and violent-felony re-arrest (20% vs 25%) than comparable defendants from before reform.
Did bail reform increase crime?
The studies found it did not increase recidivism statewide and in some regions reduced long-term reoffending. Statewide, overall re-arrest fell about 2%, with a statistically significant 2.7% drop in felony re-arrest.
Were there exceptions?
Yes. Reform tended to reduce recidivism for people facing less serious charges with limited criminal history, but tended to increase it for those facing more serious charges with recent records. Short-term increases for higher-risk releasees faded within the first months.