New field research from University of Michigan Law School professor Sonja Starr and Princeton post-doc Amanda Agan shows that hiding job applicants’ criminal records until the final stages of the hiring process—known as “banning the box”—doesn’t work so well.
Last November, President Obama, largely in the name of racial justice,
ordered a bouquet of criminal justice reforms. Chief among them was Ban the Box, which is intended to ease ex-cons’ transitions back to free society by nixing the “check here if you have a criminal record” box, a common feature of many job applications. Obama’s BTB, citing imbalance in the hirings of ex-cons, bans that box from job applications for federal employees.
The NAACP had long rallied for the cause, and since 2012, the federal Equal Employment Opportunity Commission has officially guided employers to ignore criminal records in the name of achieving racial parity. The policies went live in ten new states within just a few months of Obama’s executive order. According to the pro-BTB National Employment Law Program, by April 2016, 24 states had BTB policies, and in nine they were also extended to private employers. Even Koch Industries got in line.
Most private employers that issue form applications—rather than requesting a resume and cover letter—are looking for low-skilled, minimum-wage retail or food service workers. These are ideal positions for low-skill ex-cons and, as BTB’s logic goes, not ones a pesky application question should preclude.
Some legal scholars, however, have acknowledged that the absence of the box could encourage broader-scale racial profiling. Two recent takes on the policies’ unintended consequences, the Agan-Starr study and another from Harvard Kennedy School professor Daniel Shoag and Stan Veuger of the American Enterprise Institute, show that hiring markets constrained by BTB indeed find other ways to discriminate.
Shoag and Veuger conclude that Ban the Box policies “help low-skilled African American workers while probably harming the labor market opportunities [for] female workers” because women are less likely to have been convicted of crimes. Agan and Starr’s field experiment drew more damning conclusions: When BTB goes into effect, employers who used to ask about applicants’ criminal records turn to racial profiling to weed out candidates they deem likely ex-cons.
A new study from economist and UVA policy professor Jennifer Doleac finds similar unintended consequences of BTB: “[B]lack and Hispanic men without college degrees are significantly less likely to be employed after ‘ban the box’ than before.” She cites the same “statistical discrimination” Agan and Starr’s study highlights. In other words, when employers can’t see whether or not a candidate has a record, they’ll hedge their bets and assume the worst—based on race.
The NAACP did not respond to a request for comment on the study. And an EEOC official, claiming its unconfirmed accuracy, would not comment directly and instead offered the organization’s official position: that criminal records’ adverse effect on the success of African-American and Hispanic job applicants depends entirely on an employer’s just or unjust judgment.
Maurice Emsellem at the National Employment Law Project echoed EEOC’s reservations about the Agan-Starr study and objected to the scholars’ overly rational tone. But he won’t ignore the implications of their conclusion that discrimination finds a way. Of the Agan-Starr study, Emsellem said, “They signal the need to double down on enforcement of several civil rights laws and build on the progress that’s been made enforcing the EEOC criminal record guidance.
Agan and Starr based their findings on years of data. They (and a small army of graduate students) sent in 15,000 job applications to businesses in New York and New Jersey seeking low-skilled entry-level workers, before and after BTB took effect in those states. Then they tracked which applications received callbacks according to the applicants’ race: “The race gap in callbacks grows dramatically at the BTB-affected companies after the policy goes into effect. Before BTB, white applicants to BTB-affected employers received about 7% more callbacks than similar black applicants, but BTB increases this gap to 45%.”
They cite dual causes, forms of statistical discrimination—
As applicants did not reveal their race explicitly, in the absence of a checked-or-unchecked box, employers assume an applicant’s undesirable criminal past based on factors—e.g., first names—which “signal race.”
Through their research, providing “reason to question the idea that BTB will reduce racial disparity in employment,” Agan and Starr show the ways that Ban the Box backfires—and law-abiding black citizens from poor neighborhoods or disadvantaged backgrounds suffer the unintended consequences: Hiding a candidate’s criminal record, or lack thereof, can hurt his chances of breaking free from the cycle of poverty.
This post has been updated to clarify that Amanda Agan is a post-doctoral research associate at Princeton University and to reflect the publication of the Doleac study.
