A new working paper published by the National Bureau of Economic Research (NBER) digs into the connection between school suspensions and adult incarceration. As Harvard Graduate School of Education reports, this paper—written by Andrew Bacher-Hicks and David Deming of Harvard University and Stephen Billings of the University of Colorado–Boulder—is “the first causal evidence that strict schools do indeed contribute to the so-called school to prison pipeline.”

The research focuses primarily on the Charlotte-Mecklenberg school district in North Carolina, where 23 percent of middle school students are suspended every year. A majority of the kids suspended are students of color, and boys make up the majority of those students.

The study’s authors found that young people who attend schools with high suspension rates are more likely to be arrested and incarcerated later in life, and less likely to attend a four-year college. From the Harvard article:

“One of the arguments in favor of suspensions is that if a student is removed from the classroom, they’re no longer causing disruptions, and so removing disruptive students could have positive benefits on those who remain in the classroom,” says Bacher-Hicks, a Ph.D. candidate in public policy at Harvard. “But we found for all students, there are large negative impacts on later-life outcomes, related to attending a school with a high suspension rate. That suggests there are not overwhelmingly positive benefits of removing disruptive peers from the classroom.”

The research ultimately shows that there are virtually no long-term benefits associated with suspension. “We tested a range of variables, and only one short-run benefit popped out for any subgroup: There was some evidence that test scores increased for non-minority males,” Bacher-Hicks says. “But even for that subgroup, we find negative long-run outcomes in terms of arrests and incarcerations.”

You can read the full study here.