Over the past decade, laws that limit the ability of employers to ask about a job applicant’s criminal record have become popular across the United States. More than half of US states have now implemented some form of this rule. These “ban the box” policies, named after the tick-box on many job applications that ask about criminal records, are promoted by social justice organizations as a way to help formerly incarcerated people get jobs and, as a consequence, reduce their odds of returning to prison.
Unfortunately, the evidence suggests that ban the box policies don’t work. In fact, an emerging body of research suggests that they may have harmful unintended consequences. Not only do the rules appear to intensify racial discrimination in hiring, they don’t seem to be helping job applicants with criminal records.
Jennifer Doleac, an economist at the University of Virginia who studies crime and discrimination, thinks the problem with ban the box policies is that they are a quick fix to a complex problem. She believes taking away information from employers is exactly the wrong idea. Rather, policies intended to help the formerly incarcerated find jobs should offer employers more information about why a person with a criminal record is worth hiring.
Quartz’s conversation with Doleac has been edited and condensed.
Quartz: What was the original motivation for “ban the box” policies?
Doleac: Ban the box was motivated by this basic problem that recidivism rates, the rate at which people who are released from prison end up returning, are extremely high in the US. One reason for that is that people coming out of prison have very low employment rates. This underemployment is, in part, because having a criminal record makes it difficult to get a job.
Various studies demonstrate that if an employer receives two identical job applications, but one application notes that the person has a criminal record, that person with a record is much less likely to get a call back for an interview.
The thought behind ban the box is that if employers are discriminating against people with criminal records, we just won’t allow employers to know that information until the end of the hiring process. It varies from place to place, but often the rule is you cannot do a background check until after a conditional offer is made. Employers can still rescind the offer based on what the background check reveals.
The idea is that if people who have records can get in the same room as the employer, they can establish a rapport, and explain the context of their record. Advocates think making a personal connection before that employer finds out about their crime could help people overcome whatever biases may have been acting in the first place.
Who supports these rules?
Mostly liberal, social justice advocacy organizations. The first one that comes to mind is the National Employment Law Project, but the criminal justice reform movement has wide bipartisan support, and ban the box has been no different. For instance, Koch Industries, which is generally not thought of as being a liberal organization, has been a strong proponent.
Why might an economist think banning the box is a bad idea?
Basically, the concern of economists is that if employers don’t want to hire people with criminal records, and you tell them they aren’t allowed to know this information anymore, they aren’t simply going to throw up their hands and just pick people at random. Instead, they are going to try and guess who has a criminal record and avoid wasting their time with those people. This leads to what economists call statistical discrimination—discrimination based on averages among a group.
In the United States, race is highly correlated with whether you have a criminal record—black men in the US are more than five times more likely to be incarcerated than white men. And so the fact that employers prefer people with clean records ends up having a disparate impact on black men—this was a major motivation for the policy in the first place. But when you hide criminal record information from employers, it might make them more likely to statistically discriminate based on race.
What does your research, and the other research out there, conclude about the effects?
There have been three recent papers, with very different methodologies, that questioned the usefulness of ban the box.
I have a study with Ben Hansen at the University of Oregon where we used the timing of implementation of ban the box laws across the country as a natural experiment for understanding its effects on young black men and young Hispanic men. We found that ban the box reduced employment for young black men by 5%.
Another big study, by Amanda Agan and Sonja Starr, was a field experiment where they sent job applications for fake applicants, with and without records, to employers in New York and New Jersey just before and after ban the box laws went into effect in those places. They found that employers started calling all black men back less for interviews, basically assuming that the majority of black men had records. At the same time, white men were called back at a rate a little higher than before.
So what happened to men with criminal records? If they were better off, some would argue this still might be worth it.
Another recent study from the Boston Federal Reserve Bank examined the impact of ban the box in Massachusetts, and found that employment actually fell for people with criminal records. This is consistent with the economic theory that job search is basically a matching exercise where employers and employees are trying to find good matches, and if you take information away it’s just harder to find those matches. And so if people are spending lots of time interviewing for jobs that they have no chance of getting because the employer is going to turn them down at the end, then they are just wasting their time.
So how confident are you that this is a bad idea?
I am personally very confident that it doesn’t work. But maybe, if you don’t believe the Boston Fed’s study is applicable in all places, and you believe that helping a few people with criminal records get jobs is worth putting a lot of people out of work, then this policy could still seem like a good tradeoff to you.
But it’s important to remember that it’s not ban the box or nothing. It’s a matter of ban the box or something else. The implication of this research is not that we shouldn’t try to help people with criminal records, it’s that there are better ways to do it.
You have suggested some other options. What are they and what’s the evidence that they work?
I am a big proponent of court-issued certificates of employability programs. With these programs, which have been popping up across the US, if you have a criminal record you can go to court to argue that you have been rehabilitated and deserve one of these certificates. You can bring character witnesses and present evidence like training programs you completed. If the judge decides in your favor, he can grant you an employability certificate.
This is useful for employers in two ways. One, it gives the employer better information about whether you are going to be reliable and productive—employers might reasonably think someone who was incarcerated might not be those things. Two, it could be effective if an employer is worried about a negligent hiring suit in the event that an employee committed a crime while on the job. Having this document might protect them in court.
Researchers at the University of South Carolina have done a couple of studies where they sent out fake job applications to employers and they randomized whether the person had a one-year felony conviction, a one-year felony conviction and one of these certificates, or no criminal history at all.
They found that the people with no criminal history, and the ones with a criminal history and one of these certificates were called back at equal rates. So basically the certificate wiped out the effect of the criminal record, at least in terms of getting your foot in the door.
What employers care about is getting more information, and if these certificates are a better signal about your productivity and your reliability than your criminal record, then they are going to use that instead.
Are there other alternatives to ban the box you think are promising?
I think a lot of the conversation about helping people with criminal records is to look for quick fixes, and ban the box is appealing because it’s free. People seem to think that perhaps we can solve this major societal problem by just hiding some information from employers. It turns out that it’s not that easy.
For decades now we have been putting people in prison for many years with almost no effort to provide education of job training or counseling, or really anything that people that are incarcerated need. And then we are surprised that they don’t do well when they get out. Even more frustratingly, then we blame employers for not wanting to hire them when in many cases, but not all, they wouldn’t make good employees. And so it seems like this is an area where we are just gonna have to buckle down as a society and admit we are going to have to make some major investments in things like prison education, therapy, and job training programs.