Below, Jennifer Doleac shares five key insights from her new book, The Science of Second Chances: A Revolution in Criminal Justice.
Jennifer is an economist and expert on criminal justice policy. After nearly two decades in academia, where she was an economics professor studying crime and criminal behavior, she now leads the Criminal Justice Program at Arnold Ventures, a philanthropy focused on evidence-based policy.
What’s the big idea?
What if the best way to reduce crime isn’t harsher punishment, but smarter incentives? By applying the tools of economics—experiments, cost-benefit analysis, and behavioral insight—we can design policies that make communities safer at far lower cost.
Listen to the audio version of this Book Bite—read by Jennifer herself—in the Next Big Idea App, or buy the book.
1. Many public safety issues can benefit from the expertise of economists.
Most people think that economics is just about money and GDP. In fact, there are at least three ways that economists can be helpful in the fight against crime:
- Incentives: The criminal justice system is full of misaligned incentives. If we want people to change their behavior, in any context, we need to change the incentives they’re responding to. In the public safety context, this applies to would-be offenders as well as everyone from police and judges to victims and witnesses.
- Evidence: Of all social scientists, we are the ones most obsessed with distinguishing correlation from causation. We use randomized experiments and find clever natural experiments that separate people and places into treatment and control groups as if we were in a lab. The powerful empirical toolkit we’ve developed allows us to determine whether a program or policy will work—and by how much it would move the needle. There are lots of well-meaning programs, but we need much more careful testing to determine which truly improve people’s lives and how to make them even more effective.
- Tradeoffs: Because we are so good at quantifying the costs and benefits of various policies, economists can help policymakers and practitioners allocate their scarce resources to make communities safer. The reality is that no one can do everything—community leaders have limited money, staff, and bandwidth. Economists are pros at thinking through tough tradeoffs to get the biggest bang for our collective buck.
2. Small, carefully targeted changes to incentives can have big benefits.
We don’t need to wait for big structural changes in order to dramatically improve people’s lives. There are myriad examples of interventions that work, and that both sides of the political spectrum can get behind. These solutions reduce crime and the harms of the criminal justice system, and typically cost less than the options we’re used to hearing about on cable news.
One persistent problem facing courts across the country is that many people charged with a crime do not show up in court for their hearings. Almost a quarter of felony defendants who are released pretrial fail to appear for their court dates. Now, people already face a strong incentive to appear for these hearings: they face a new warrant for their arrest if they don’t show up, which will increase the consequences they’ll eventually face. If this incentive isn’t effective, it might seem that the only remaining option is to put people in jail while they await trial. But pretrial detention is expensive and disrupts people’s lives in ways that can increase crime rather than reduce it. For instance, they might lose their job or housing if they’re locked up for several days.
But what if missing court hearings is sometimes a mistake, either because people didn’t realize they had a hearing at all, or they didn’t understand the consequences of missing it? All the incentives in the world won’t matter if people don’t know about them.
A cool experiment in New York City showed that sending text message reminders about court hearings and making information about the court date and consequences for missing the hearing much clearer led to a 13 percent reduction in failures to appear. That’s a big impact for changes that were close to free. This intervention didn’t completely solve the problem, but it made a big dent. And it’s way cheaper than keeping someone in jail. There are many other clever changes we could make throughout the system that would make everyone better off.
3. Chances of getting caught deter crime more than punishment.
Crime is largely a young person’s game. Young people don’t think very far ahead, so the threat of a long sentence doesn’t do much to change their behavior in the present. That is, long sentences don’t have the deterrent effect that we think they will. What does change behavior is consistent, immediate consequences for bad behavior—even if the consequences are more moderate.
In the U.S., the probability of arrest for violent crime is less than 50 percent, and for property crime it’s much lower. Investing our scarce public safety dollars in police and investigations to solve more crimes, more quickly, would have outsized benefits for communities and for those who might get into trouble. One tool that police use to solve crime is a DNA database. They collect DNA profiles of people who were convicted or arrested in the past. They can then compare those profiles to DNA from crime scenes to identify suspects. This increases the likelihood that people will be caught if they reoffend.
“Long sentences don’t have the deterrent effect that we think they will.”
Expanding DNA databases to include people earlier in their criminal careers has been shown to reduce recidivism by over 40 percent. That’s a huge public safety benefit, and puts those offenders on a much better path. Research shows they’re more likely to be enrolled in school or working because they’re in the database—knowing they’re more likely to get caught if they reoffend pushed them to make different, better choices. Increasing the probability of getting caught not only brings justice to victims and their families, but it also prevents crime from happening in the first place.
4. Arrest, incarceration, and probation should be intervention points.
Using our criminal justice system just to keep high-risk people off the streets is fine, but it isn’t nearly as cost-effective as it could be. Why not take advantage of the time people are under government supervision to help them build better lives—and avoid crime—in the future? There are some interventions we know are effective, but, frankly, we’ve been slow to invest in this area, and we need many more.
There is strong evidence that increasing access to mental health care for high-risk groups (especially young men) reduces criminal behavior and reoffending. This includes everything from cognitive behavioral therapy to medications for bipolar disorder and schizophrenia. A clever study used a natural experiment in Sweden to show that being incarcerated in that country actually improved mental health outcomes, because people were more likely to get treatment behind bars.
Now, Swedish prisons offer especially good health care. But that country also offers great health care outside of prison. The takeaway from this study, to me, is that there are people who would benefit from mental health care who will not engage in it until or unless they are behind bars. That country is smart to use imprisonment as a key intervention point to get people the help they need to avoid crime going forward.
We see some examples of this in the U.S. Jail administrators in Johnson County, Kansas, piloted a program for the 25 percent of people released from their jail who screened as having untreated mental illness. They had outreach workers contact those people and offer to make them an appointment at a local mental health clinic. That’s it—no additional follow-up, just a warm hand-off to people who could help. But that light-touch intervention, at just the right time, had big benefits, reducing the likelihood of another jail booking in the following year by 17 percent. That’s a huge impact for such an apparently tiny change.
5. Many good policy ideas don’t work, and some even backfire.
Our goal should be to fail fast and try something else, rather than pretend we won’t fail at all. This is a familiar concept in the tech world, but it is an uncommon approach in public policy and especially in the criminal justice policy space.
In the policy world, it’s much more common for advocates and lawmakers to claim they have the solution and then to focus campaigns on building momentum to implement that solution everywhere, as quickly as possible. An example of how this has gone very poorly is in efforts to help people with criminal records find jobs. We know that many employers are reluctant to hire people with records, which makes it more difficult for those job seekers to build stable lives and avoid crime. How can we encourage more employers to give these jobseekers a chance?
“Our goal should be to fail fast and try something else, rather than pretend we won’t fail at all.”
One appealing approach has been to delay when employers are allowed to ask about a job applicant’s criminal record. Traditionally, there is a question on job applications, something like “Have you ever been convicted of a crime?” If you check the box “yes” many employers would simply discard your application. Ban the Box policies sought to change this. They barred employers from asking about a criminal record until late in the hiring process, typically after a conditional offer had been made. By then, the thinking went, employers would have already considered the applicant’s abilities and qualifications, perhaps getting to know them a bit as a person, and so might be more forgiving of a criminal record. Unfortunately, it hasn’t worked out that way.
Research has shown that Ban the Box policies have little to no impact on employment for those with criminal records. Even worse, they have made it more difficult for young Black men without criminal records to find jobs. The problem with these policies is that they did not change employers’ underlying concerns about a criminal record. But now that they couldn’t ask up front, they tried to guess based on the remaining information they could still see. In practice, this meant they discriminated based on race instead. Removing information broadened discrimination, rather than reducing it. We’ve known this for a while now, but these policies are still in effect and are still being proposed in cities and states across the U.S., as well as in other countries.
A better approach would have been to try Ban the Box in some places and other strategies in others. This would have caught these problems early and allowed us to direct our energy and resources toward promising alternatives. Instead, we wasted a lot of time and harmed many people in the process. Going forward, I’m excited about interventions that directly address employers’ concerns, like insurance that shifts the risk of hiring someone with a record from the individual employer to the community, as an incentive to give more job applicants a chance. But we’ll want to do this as a careful pilot, to make sure it works before we scale it.
We need to try new ideas, but we need to do so strategically. Adopting a “fail fast” mentality ensures we approach tough policy challenges with humility and will speed up the rate at which we find effective, scalable solutions that change people’s lives for the better.
Enjoy our full library of Book Bites—read by the authors!—in the Next Big Idea App:
