Every day in the United States, approximately 500,000 people sit in jail cells not because they have been convicted of any crime, but because they cannot afford bail. They are legally innocent. Many of them posed no identifiable flight risk and no threat to public safety. They are in jail because a judge set a dollar amount and they did not have it.
The cash bail system was designed under the assumption that money creates an incentive to appear in court. That logic has always been weak. A wealthy person who forfeits bail money experiences far less financial pain than a poor person who forfeits the same amount. The actual effect of cash bail is not to make defendants appear in court. It is to sort defendants by income, keeping poor people in jail before trial while releasing people with means.
The consequences of pretrial detention are severe and well-documented. People who are detained before trial are more likely to plead guilty, even when they did not commit the charged offense. This effect has been documented in multiple jurisdictions and is not explained away by selection bias in researchers' most sophisticated analyses. The mechanism is clear: sitting in a jail cell while your job disappears, your apartment lease lapses, and your children face childcare crises creates overwhelming pressure to accept a plea offer, regardless of its merits.
A study of federal courts published in the Journal of Criminal Law and Criminology found that detention before trial increased the probability of conviction by 25 percentage points. A separate analysis in Pennsylvania found that detained defendants received sentences 42 percent longer than those who were released pretrial for comparable offenses. These are not marginal differences. They are the difference between employment and unemployment, between family stability and family dissolution, between a life with a criminal record and one without.
New Jersey's 2017 bail reform provides the clearest natural experiment we have. The state eliminated cash bail for most offenses, replacing it with a risk-assessment tool that recommended release or detention based on factors correlated with appearance and public safety. In the first two years after reform, the jail population fell by 44 percent. Failure-to-appear rates did not increase. Crime rates did not increase. The predicted crisis did not materialize.
What did change was the composition of who was detained. Pretrial detention became increasingly limited to people who had demonstrated specific risk factors: serious violent charges, prior failures to appear, active supervision violations. The arbitrary filter of bank account balance was removed.
Critics of bail reform, particularly after a high-profile crime involving someone released pretrial, often argue that the risk-assessment tools are imperfect and that any released defendant who reoffends represents a failure of the system. This framing is not honest. The standard for evaluation cannot be zero risk. The alternative, cash bail, also produces crimes: it is just that those crimes are committed by people who were given a different form of preferential treatment. The relevant question is which system produces better aggregate outcomes across the full population. On that question, the evidence favors reform.