ANN ARBOR—”Ban the Box” legislation seeks to open doors to employment for people with criminal records by barring employers from asking about records on employment applications. More than 20 states and over 100 municipalities have passed such laws in recent years, some of which govern private employers.

But a major new study released today by researchers at the University of Michigan and Princeton University points to a serious unintended consequence of these laws: While they may indeed improve the prospects of people with records, this gain comes at the cost of encouraging a substantial increase in racial discrimination by employers.

“This consequence is clearly unintended—in fact, Ban the Box is often presented as a strategy for increasing black men’s access to employment,” said Sonja Starr, professor of law at the U-M Law School. “Unfortunately, we think our results strongly suggest that when it comes to this goal, it has backfired.”

Starr and co-author Amanda Agan, an economist at Princeton, conducted a large-scale field experiment. Over the course of a year, the authors (and their large team of U-M student research assistants) sent nearly 15,000 fictitious online job applications to entry-level, low-skill positions in New York City and New Jersey, varying the applicants’ race and criminal record. The applications were sent in black/white pairs in two waves before and after each jurisdiction’s adoption of Ban the Box.

Agan and Starr first found support for the basic premise of Ban the Box: when companies ask about them, criminal records are a substantial obstacle to employment. Applicants without records received 63 percent more callbacks than similar applicants with records. Agan and Starr found high rates of compliance with Ban the Box, meaning that the law effectively removes this obstacle—at least at the earliest stages of the hiring process, which is what Ban the Box seeks to affect.

But the law has a serious downside: It appears to dramatically increase racial discrimination by affected employers. When employers asked about criminal records, Agan and Starr found that white applicants had a relatively slight advantage: they received about 7 percent more callbacks than equally qualified black applicants. After Ban the Box was adopted, this gap ballooned to 45 percent. This change was not seen among employers whose applications were unaffected by the law because they never asked about records in the first place.

The researchers theorize that the reason relates to a phenomenon known as “statistical discrimination.” If employers don’t have information about criminal records, they are more likely to rely on their assumptions—including race-based assumptions. Specifically, employers may assume that black applicants have criminal records (even when they don’t), and that white applicants do not.

“Our results don’t necessarily definitively argue against Ban the Box,” Starr said. “It clearly has benefits for people with records, and policymakers might decide that those benefits are important enough to justify the law. But our results are very worrisome in terms of the effects for black male applicants, especially those without criminal records.”

“When you take criminal record information away, some employers seem to simply assume that black men are likely to have criminal pasts,” Agan said. “So black men without conviction records, who won’t be able to reveal that fact to employers, may be the ones who bear the costs of Ban the Box. This is especially troubling because black male unemployment levels are already more than twice the national average.”

More information:

• Study: Ban the Box, Criminal Records, and Statistical Discrimination: A Field Experiment
• Sonja Starr
• Amanda Agan