Provo, Utah
LAST week the Supreme Court ordered California to reduce its prison population after finding that the state’s penal system was so overcrowded that it constituted cruel and unusual punishment. What the court didn’t do, however, was provide any guidance about how to do it, giving rise to fears of violent convicts being set free and increasing crime rates.
Rather than seek major criminal justice reforms to reduce the prisoner numbers, including scrapping California’s harsh “three strikes” sentencing laws, Gov. Jerry Brown has proposed simply moving the surplus state prisoners to county jails. This does nothing to reduce California’s disproportionately high incarceration rates and could just transfer the overcrowding to local jails.
Fortunately, there is a more lasting solution to overcrowding, one that gets to the heart of exploding inmate populations nationwide: reform the rules governing pretrial detention, in part by using formulas to help judges better determine which defendants are unlikely to commit crimes while on bail. Doing so not only would make the system more fair, but also would significantly reduce the number of people who are unnecessarily jailed and even reduce crime rates.
Every year America spends close to $66 billion to keep people behind bars. But almost 500,000 of the 2.3 million prisoners aren’t convicts; rather, they are accused individuals awaiting trial.
While some defendants are able to pay their bail and go free, most cannot, because many judges, lacking firm insight into what types of prisoners are too dangerous to release, set high bail amounts knowing the accused can’t afford them. Though some of these defendants will eventually be found not guilty and go free, keeping them incarcerated before their trials creates a burden on the prison system.
What’s more, detention begets more detention. Defendants detained before trial are more likely to be convicted if they go to trial, more likely to receive prison sentences rather than probation when sentenced, and, given their weak bargaining power with prosecutors while locked up, are more likely to have longer sentences.
A few jurisdictions, however, have begun to think outside the prison cell. In line with recommendations endorsed by the American Bar Association, Miami-Dade County cut costs associated with detention by supervising defendants outside jail at a total cost of around $400 per defendant per year, compared with $20,000 for incarcerated defendants. In Iowa, alternatives to pretrial detention saved the state’s Southern District $1.7 million in 2009.
These and other jurisdictions have also cut costs using technology, like G.P.S. trackers and ankle bracelets, that allow defendants to remain at home — with supervision — while awaiting trial.
True, while these solutions may make sense from a budgetary standpoint, critics worry that increasing pretrial releases will present a threat to public safety, especially since judges are typically left to make bail decisions based simply on a gut feeling.
The risk of release can be largely reduced by arming judges with more data to inform their decisions. Frank McIntyre, an economist, and I recently examined data from over 100,000 felony defendants over a 15-year period, and we found very clear trends regarding which defendants are more likely to commit crimes while free on bail.
For example, judges often detain too many older defendants (people over 30), defendants with clean records and defendants charged with fraud or public order offenses — in other words, people who are less likely to commit crimes while out on bail. On the other hand, judges release too many young defendants with extensive records, people who are more likely to break the law while awaiting trial.
This data could be used to create a set of guidelines that would give judges a better sense of which defendants to release. Of course, judges must use individual discretion and carefully consider local data with pretrial detention decisions. However, our models indicate that such guidelines could safely lead to the release of up to 25 percent more defendants — and a significant reduction in prison costs and crime rates.
Given eye-popping local, state and federal deficits, it’s unlikely that California will be the only state to face the tough choices involved in reducing its prison population. With the right data on pretrial defendants, though, judges can help make that task a lot easier.

Shima Baradaran is an associate professor of law at Brigham Young University and the chairwoman of the American Bar Association Pretrial Release Task Force.