Criminal Background Checks and Hiring Ex-Offenders

This player requires the latest version of Flash.

2009 NIJ Conference 2009

National Institute of Justice
Al Blumstein, Ph.D., Carnegie Mellon University
Kiminori Nakamura, Doctoral Student, Carnegie Mellon University

Jolene Hernon: I'm Jolene Hernon. I'm from the National Institute of Justice. I'm here today with Kiminori Nakamura and Al Blumstein from Carnegie Mellon. They're talking about the panel they were on yesterday about criminal background checks and hiring ex-offenders.

Al, why don't you tell us a little bit about why you decided to propose this study? What were you thinking?

Al Blumstein: I was on a task force established by BJS and The Search Group to look at the issue of the backgrounding of America. It was called the Task Force on Backgrounding of America—a recognition that more and more we are doing background checks on people. And so we came to realize that number one, background checking is becoming more and more ubiquitous. Eighty percent of large corporations do background checking, and, and also a large fraction of the population have criminal background, criminal history records, having been arrested or convicted of something. Some, some 40 years ago, we did a study that showed that over 50 percent of males in the United States would have an arrest record for non-traffic offense; that number has undoubtedly increased appreciably. We got 14 million arrests every year and so more and more people who are absolutely capable, risk, low risk, can hold a job quite admirably but are denied jobs because of that background check showing an arrest. So we wanted to know and, and lots of people have studied the issue of how significant a criminal history background is in whether an employer hires them. So, but nobody yet had an estimate of when after someone has been, has, has done something, has an arrest record, or a conviction record, at what point their risk becomes sufficiently low that its lower than the general population and close enough to people who've never had an arrest. That, that information is stale and should be ignored or should not be used as a discrimination against a larger and larger number of people who are in that position of having a criminal background. And so that impelled us to start looking at it and to try to find out when some of these transition points occur.

Hernon: Well, an employer's worried that if I hire you, you committed a crime in the past. You're likely to commit another crime.

Blumstein: And, and that is certainly a possibility with me, but it's also a possibility with someone who's never done a crime. And so there's always risk out in the society. The issue is when did the risk get low enough that it warrants hiring that individual.

Hernon: Well, Kiminori, can you tell us about whaddaya know about how frequently people commit new crimes? If they'd committed it once, how likely are they to commit another one? What do employers need to worry about?

Kiminori Nakamura: Well, many, like, recitative studies in the past have shown that most recidivism occurs in the first three to five years, ya know, tracking, for example, people just, who are just released from prisons. So, ya know, their risk of rearrest or reincarceration is the highest in the first couple years, but then after that, ya know, it declines pretty monotonically. So what we are interested in is those people who stay clean much longer, longer than three to five years. And, ya know, so we are interested in what their risk looks like and how low that risk is compared to appropriate benchmarks.

Hernon: You call this point "redemption," that which, the point ... Well, describe to me what redemption, what you mean by redemption.

Blumstein: Redemption is a concept that, that I got involved in, in the issue some years ago when, when someone argued—when we are computerizing criminal history records—someone argued that we shouldn't computerize the records because computers don't understand the Judeo-Christian concept of redemption, which means that you're, in effect, forgiven for some past misdeeds. A rejoinder to that was that paper records certainly didn't understand the concept of redemption but computers could be taught. And so we, we would find ways by looking at criminal history records to see when people's risks are sufficiently low that they warrant being hired. So, and the concept of redemption is, is an important one in terms of a society that recognizes that people, many people do misdeeds particularly when they're young, when they don't have the restraint, when they don't have the maturity. And as they get older, opportunities should be available to them to, to redeem themselves following the, the passage of time and the misdeed. Now this shouldn't get in the way, by the way, of a whole variety of efforts to try to find opportunities for employment for people who have been in prison who have recently committed crimes. There's a large number of people ... We have over 2 million people in, in prison or jail today. And it's important that we find opportunities for them to get jobs, because the jobs are important mechanisms for moving them into legitimate operation, legitimate work, legitimate society and, and as an inhibitor to recidivism during those first three to five years.

Hernon: You're looking at four different kinds of crimes: robbery, assault?

Blumstein: We're looking at a whole variety of crimes. We focused specifically, initially on, on robbery, aggravated assault, and burglary partly, because there's enough people, large enough number of people who've done those, and they're all serious crimes that we care about.

Nakamura: But we're also looking at other violent offenses, as well as property offenses. And we're more and more interested in the drug offenses as well, because they're, ya know ...

Blumstein: And sex offences because the country is very obsessed with sex offenders, and it's important that we understand the different kinds of sex offenders.

Hernon: Right. So what are you finding? Can you summarize the—

Blumstein: We haven't done those yet.

Hernon: Not on sex offenders but you do have some numbers on robbery?

Blumstein: Yes.

Hernon: And assault?

Blumstein: Kiminori?

Nakamura: So the, we, we call this, the redemption times T star and the T of the star. But those redemption times are longer for those people who started or who were arrested at early ages as well as in terms of the crime types. Those who were arrested for robbery tend to have large numbers for redemption times, so they need to stay clean longer to achieve that redemption.

Hernon: I, I heard the number seven batted around.

Blumstein: Let, let, let me put it, a somewhat different take on that. That someone who did robbery as his first offense is someone whose clearly of higher risk than someone who might have done a burglary as his first offense, so that the redemption time for that will end up being longer. The, the time he's gotta wait before his risk is low enough, based on some standardized criterion, the time he's gotta wait is gonna be longer and similarly, someone who started at an earlier age is more risky. And, and so someone who started earlier should wait a bit longer.

Hernon: Somebody described this earlier —and it didn't come up in panel yesterday—but somebody described this to me as, sort of, actuarial tables that are done with insurance companies. If you're 25 and a male, you have a different payment, premium for your car than if you're a 18-year-old male. Do you think that's a related—

Blumstein: Both are related to the risk associated with you're driving concerns. And so it's a risk measurement. And so actuaries deal with different kinds of risks. We're dealing with risks that employers care about, which is that someone might commit a crime, hurt the reputation of the company and, particularly, expose the lawyer to—that the employer—to liability concerns for somebody who, who, who did the crime to one of the employer's clients. So that one, one of the proposals ... We've made a number of proposals for how to implement the results of this research, and one of them would be for a legislation that shows them individuals who have been clean for a reasonable amount of time, which we estimate should be protected by statute from vulnerability to someone suing them for due diligence.

Blumstein: It was particularly striking to hear the story of one of the people whose mother is in dementia and her brother moved into her apartment in public housing to take care of her. And it turns out that there's a rule on public housing that anyone who has ever been convicted of a drug crime may not live in public housing. And the brother was convicted 25 years ago of a drug crime, has been clean since then and was told that he can't move into public housing. Now, he might have been convicted of an armed robbery or even a murder—that wouldn't get in the way—so he passed a lot of knee jerk legislation that is intended to deal with the problem of the moment that, that impinged on the appropriate care for this mother. And so we've got to find ways in society generally, where so often redemption doesn't ever occur in life, where people who've ever done something in the past are prohibited forever from getting certain kinds of employment, or getting certain kinds of college loans, or certain kinds of housing. And we've got to rethink that in terms of the opportunity for redemption, particularly in the context of background checking becoming so ubiquitous —all over the place—with, with more and more employers and more and more job opportunities and so many more people getting involved with the criminal justice system.

Hernon: Maybe this is the place to talk a little bit about the false negatives and the decisions that are made based on Willie Horton ...

Blumstein: There's always the possibility that no risk is zero. The risk of someone who's never been arrested is never zero and the risk of someone who's done something in the past is never zero. It's always possible. And so as a society we've got to avoid some of the, again, knee jerk responses to an event that, that was always possible, but, but, can inflame the public, gets headline news, and, and leads to very rapid response in the political environment that leads to more and more people going to prison, less and less attention to the need for redemption of people who've done something bad in the past whose risk is not zero but whose risk is sufficiently low that, that risk should be accommodated.