Amanda Agan was reading a newspaper one morning this past fall when she came across a story that gave her pause: New Jersey would soon pass legislation making it illegal to ask about criminal backgrounds on a job application.
As a postdoctoral research associate in the Princeton University Department of Economics, Agan, 32, was familiar with the concept, but skeptical about its results.
"I had in the back of my mind that there's some literature that has shown that these kind of policies can cause racial discrimination on behalf of employers," Agan said. "I had to guess that this sort of policy might make it so that employers were going to use other characteristics, particularly race, to infer whether someone has a criminal history."
The report found that before ban-the-box policies were enacted, applicants with stereotypically white names, like Scott, Cody and Dylan, received close to 7 percent more callbacks than applications with stereotypically black names, such as Tyree, Terrell and Daquan. After BTB policies were enacted, however, applicants with white names received 45 percent more callbacks.
"We have a policy that may do some substantial good for a group of people for whom getting access to jobs is important, namely people with records who face various obstacles to getting employment," Agan said. "But it may come at the cost of another disadvantaged group in society."
The two sent out more than 15,000 fake job applications to employers in New York and New Jersey before and after BTB policies passed to find whether racial biases — intentional or otherwise — were a significant factor in hiring job applicants.
And to ensure each application had a fair chance, Agan and Starr used identical qualifications for each application, changing only the name and criminal background. They chose "racially distinctive names" by going through the New Jersey Department of Health birth certificate data from 1989 through 1996, which they then used to test the effects of recently enacted BTB policies.
"Ban the box laws have been advocated as strategies for reducing racial disparity in employment, which is one thing that makes our findings particularly striking and unfortunate, because it suggests that the laws have backfired in achieving that goal," Starr said.
According to the National Employment Law Project, a worker's rights advocacy group, there are now 24 states and more than 100 cities and counties that currently have BTB laws in place. Ten states have even extended the policies to the private sector as well.
NELP staff attorney and Ford Foundation fellow Beth Avery said her group advocates for BTB policies because they "could give people who are struggling to find work because of an arrest record the chance to work."
She said Agan and Starr's study, like other recent ones on BTB, are being conducted too soon after the policies go into effect, which doesn't give employers time to adequately "adjust their hiring policies."
But Starr, 40, pointed out that while the intentions of backing BTB policies was to "open doors to people who have criminal records," it may actually hurt minority populations who don't have a criminal past.
"The intuition was fairly simple: opening doors for people with records is going to reduce an obstacle to employment that disproportionately affects people of color, so therefore it should be a tool for reducing racial disparities in employment," Starr said. "But most people of color don't have criminal records. People of color appear to lose the ability to signal to employers that their records are clean," Starr said.
Research from the University of South Carolina shows, however, that arrest rates are close among different races and ethnicities, reporting that 50 percent of black men — and 40 percent of white men — in the United States are arrested before they turn 23 years old.
USC researchers also found that a criminal background can "impede employment, reduce access to housing, thwart admission to and financing for higher education and affect civic and volunteer activities such as voting or adoption."
Agan and Starr anticipated their study would show some racial biases in hiring processes based on previous research in the field, but both admitted to being surprised by the magnitude of the racial disparity they found among applications that received callbacks.
"We expected that because there are widespread, group-based generalizations that associate — sometimes I think in an exaggerated way — African-Americans, especially young black men, with having high rates of criminal records," Starr said. "But I have to say I was a little bit surprised with the magnitude of the effect we found."
She and Agan found that employers who don't ask about criminal records "may make assumptions about criminality based on the applicant's race," and that BTB policies could be detrimental to minorities and people with criminal backgrounds.
"For decades, researchers have been saying that when you take away individualized information that decision-makers want access to, they will often then rely instead on group-based generalizations," Starr said.
And NELP research shows that there does seem to be bias among employers when looking at applicants with a criminal record, which increases dramatically when the applicant is a person of color. In a study released in June, NELP found that while 17 percent of white applicants with a record were called back after applying for a job, only 5 percent of African-Americans with a criminal record heard back from the employer.
"Employers are willing to, in order to avoid interviewing someone that they will never be forced to hire anyway, employers seem to be engaging in racial discrimination that is so clearly illegal," Starr said. "Unfortunately, racial discrimination laws have become rather difficult to enforce."
She maintains that the policies could have unintended consequences that disproportionately affect specific populations, particularly black men, whose unemployment rate is nearly double that of the national rate, according to the Bureau of Labor Statistics.
"The gains it makes for people with records come at the expense of people with color who don't have criminal records," Starr said."You don't get to tell an employer that you don't have a record until maybe it's too late to get your foot in the door."
Avery recently co-authored a paper disputing Agan and Starr's report, as well as other recent studies criticizing BTB, that said the policies are working, "both by increasing employment opportunities for people with records and by changing employer attitudes toward hiring people with records."
"Employers are essentially throwing applications in the trash if the person has indicated that they have a record," Avery said. "What ban the box policies are trying to do is give the applicant an opportunity to show that they're more than just their records."
And while Agan said she and Starr were hesitant to offer any alternatives or solutions to BTB policies because "there simply aren't any easy ones," she did say the idea that better enforcement of existing policies could end some of the "neative consequences" that they found in their research.
Avery echoed that idea in her response to the study, writing that instead of "rolling back the clock" on BTB policies, more should be done to prevent employers from discriminating based on race by better enforcing Title VII laws that prevents discrimination based on "race, color, religion, sex and national origin."
"Step one would be addressing how to enforce those laws better," Avery said. "We need to address race discrimination in hiring through enforcement of our existing policies and through additional policies that can address that discrimination and address the underlying bias. That's how we're going to move forward."