Second Chance, My Ass

Ex-offenders need jobs to stay out of jail. But easy access to criminal records, a gift of the Internet age, means that employers won’t hire them.

What does a criminal record tell us about a person? Does it forever identify him as a threat to society? This is not an idle question: between a quarter and a third of American adults—around sixty-five million people—have some type of criminal record. About twenty million of them, or 12 percent, have a felony conviction. Among African Americans, the rate of felony convictions rises to an astonishing 25 percent.

The Eternal Criminal Record

by James B. Jacobs
Harvard University Press, 416 pp.

The most debilitating consequence of having a record is employment discrimination. Three out of four employers run criminal background checks on all applicants, and some 90 percent use them for at least some hires. In studies, an overwhelming proportion of employers say they are unwilling to hire someone with a felony conviction. Many refuse to hire anyone with a record, period. The Equal Employment Opportunity Commission has declared that it’s illegal for companies to categorically exclude ex-offenders, but its rules leave enough wiggle room to make them difficult to enforce.

The single most important predictor of future crimes is whether or not an ex-offender finds work once he or she leaves prison. This creates what economists call a collective action problem: hiring ex-offenders reduces recidivism, which is in everyone’s best interest; but to any individual employer, the perceived risk of hiring someone with a record outweighs the marginal benefit to society.

The current landscape of criminal records and background checks developed under little scrutiny. It just sort of happened, according to the New York University law professor James B. Jacobs in his recent book, The Eternal Criminal Record. Jacobs tries to catalog and comment on every aspect of criminal record keeping in America. It’s an impossible task for one volume, and the result is an overbroad and sometimes meandering book, written in an academic’s flat prose. Still, for the patient reader, there is probably no better single source of information on the ways in which criminal records are created, stored, shared, and used.

In the United States, criminal court records have traditionally been available to anyone who wants to see them. For most of the country’s history, however, the files lived in what the Supreme Court called “practical obscurity.” To find someone’s record, you had to figure out which county he’d been prosecuted in, physically go to that courthouse, and hope the clerk could find the file. It was inconvenient and expensive.

All that changed with the birth of the Internet. In 1996, for instance, only 50 percent of employers used background checks for some hires. But as more and more information was centralized by the state and made available online—spurred by the Brady Handgun Violence Prevention Act of 1993, which required states to create statewide databases—the commercial background check industry exploded. These companies can now compile data—most often by buying records online directly from the courts—and in turn sell it easily and cheaply. Single-state searches cost as little as $10, while a comprehensive nationwide search may cost only $60. Checking applicants’ criminal histories has become perfunctory.

As Jacobs points out, this is a distinctly American story. In most European democracies, where the right to privacy and the goal of rehabilitation are taken much more seriously, court records are presumptively not available to the public. Your criminal history is your own business. In Spain, for instance, “only judges, prosecutors, certain police agencies, and the record-subject” can access records from the National Conviction Register. (People with unexpunged convictions, however, are disqualified from public-sector jobs.) In the United States, by contrast, privacy and rehabilitation take a backseat to transparency. The Supreme Court has recognized a First Amendment right to open court records, endorsing the view that access to those records is critical to holding government accountable.

It is deeply ironic that a legal custom based on the idea of preventing governmental abuse has had the practical effect of making criminal punishments more draconian. Not surprisingly, racial minorities are impacted the most. In a landmark field experiment in 2003, the sociologist Devah Pager found that having a drug conviction had a significantly bigger impact on employer responses to black applicants than to otherwise identical white applicants. A 2011 study found that black ex-inmates’ wages grow 21 percent more slowly after prison than whites’.

There is little question that former prisoners are more likely to have problems such as poor employment histories, alcohol or drug abuse, and the like. Amazingly, however, there’s no proof that simply having a criminal record makes someone a worse employee than someone who doesn’t. Pager also surveyed New York City employers on their feelings about hiring ex-offenders. Most said they wouldn’t hire people with convictions because of fears of theft, violence, or general untrustworthiness. Yet the employers who had hired ex-offenders within the previous year overwhelmingly reported positive or neutral experiences. Many believed that ex-offenders made better employees, because they have much more at stake, and are hugely motivated to keep their jobs.

Criminal history reports are notoriously inaccurate—according to a U.S. Department of Justice report, 5.5 percent of background checks erroneously report a record where none exists. But even a perfectly “accurate” system would suffer from the problem that most illegal behavior is never detected, let alone punished; criminal records are only a rough proxy. The past three presidents have admitted to using drugs that millions of Americans have felony convictions for possessing. The enormous discretion wielded by police and prosecutors determines whether a certain act leads to an arrest, indictment, or conviction.

The good news is that momentum for reform seems to be building. Sixteen states and dozens of counties and cities, including Chicago and San Francisco, have recently passed versions of “Ban the Box” laws, which generally prohibit employers (sometimes just public, sometimes public and private) from asking about criminal records until after the job interview. (The name refers to the box applicants are routinely asked to check if they have any convictions.) Some of the country’s largest employers, including Target, Walmart, Home Depot, and, as of April, even Koch Industries, have adopted the policy for their own hires.

Jacobs makes a few commonsense suggestions for reform: Congress should do away with some of the laws barring certain ex-offenders from government contracting jobs. More nonviolent crimes should be decriminalized to spare people records in the first place. Existing regulations on background check companies should be better enforced. And so on.

But he defends the basic premise that employers should have access to applicants’ criminal records. His argument boils down to a claim about individual fairness: since ex-offenders “are responsible for their tainted biography,” it’s not unfair to discriminate against them; meanwhile, since employers bear the risk of a poorly performing employee, it would be unfair to prevent them from getting relevant information.

It’s “not irrational” for businesses to want more information about applicants, he writes, so let’s not deprive them of the ability to find it. Perhaps it would be rational to want to know about someone’s physical and mental health records, too. But are we better off making that information public? We take for granted that the answer is no. Given the societal cost of pushing ex-offenders out of the labor market—which, according to the left-leaning Center for Economic and Policy Research, costs the U.S. economy between $57 billion and $65 billion in output—why shouldn’t the same logic apply to criminal records?

It’s possible to imagine a much saner approach than the one we have— although, given the lobbying power of the multibillion-dollar background check industry, it won’t be easy to bring about. As Jacobs writes, “computerization did not cause fundamental rethinking of public access,” but it should have. For starters, states should stop making records available for bulk purchase online.

The current default, that private employers have access to online criminal records, should be reversed. The First Amendment protects access to physical court files, but digital criminal records should be presumptively private unless there’s a basis for believing that allowing people with certain types of records to hold certain jobs would pose a serious risk for society. Jacobs proposes empowering a federal commission to make recommendations about what kinds of convictions should be legally disqualifying for what kinds of jobs. Better yet would be a commission that would determine, as a matter of regulation, which industries should be given access to what kind of criminal history information. Some, but not all, private employees have a significant impact on public safety or welfare. It would make sense for the transportation industry—think truckers, pilots, train conductors—to be able to do background checks in order to find out whether someone has been convicted of drunk driving, because substance abuse by someone in that position has a clear potential to put many people in danger. On the other hand, whether the cashier at Best Buy has a theft conviction may make a difference to Best Buy, but there’s no reason to think it’s something that will affect society at large.

It should be illegal to ask about or report arrests that didn’t lead to conviction, because the stigma attached even to an unprosecuted arrest lets the government circumvent the principle that a defendant is innocent until proven guilty. There should also be tighter restrictions on how far back criminal history reports can go. Research suggests that even some types of violent felons are no more likely than the average person to reoffend after three or four years out of trouble. Yet convictions remain a potentially permanent barrier to employment.

Jacobs, citing the work of fellow academics, argues that restricting access to criminal records would disproportionately punish members of highly incarcerated minorities who don’t have criminal records, because employers will use race as a proxy for criminality. One study even suggests that this effect might outweigh the employment gains among minority members who do have criminal records. That’s an important concern, but a more thorough 2008 study by Keith Finlay, an economist at Tulane, found that while increased access to criminal records since 1997 has hurt ex-offenders’ employment, it has had no statistically significant affect on employment rates among non-offenders.

The policies surrounding criminal records are mostly the product of inertia. They also embody the assumption that everyone who commits a crime is somehow forever bad, dishonest, or dangerous. For everyone’s sake, governments and employers would do well to revisit that assumption. As long as ex-offenders are systematically kept out of the workforce, it may be a self-fulfilling prophecy.

Gilad Edelman

Gilad Edelman is editor of the Washington Monthly.