Today the Department of Education and Justice Department released new discipline guidelines intended to reduce racial disparities in punishment in the nation’s schools. The move stems from a combination of factors: African-American students are disciplined more harshly, on average, by public schools; and suspensions and expulsions are associated with negative long-term educational outcomes for the disciplined students. The guidelines are technically voluntary, but as the Associated Press reports, “the federal government is telling school districts around the country that they should adhere to the principles of fairness and equity in student discipline or face strong action if they don’t.” Unfortunately, this federal pressure may end up hurting black students far more than it helps them.

The problem is that while expelling disruptive students may be associated with negative educational outcomes for the disruptor, it is associated with positive educational outcomes for the rest of his classmates. That is the finding of a uniquely sophisticated study conducted by Joshua Kinsler and published last year in the prestigious International Economic Review (a draft is available here). Kinsler found that cutting out-of-school suspensions in schools with many disruptive students lowers overall student achievement.

In that and earlier work, Kinsler also discovered that the disparity in punishments handed out to students of different races is almost entirely explained by the schools the students attend, and not by racism. Black students, Kinsler found, are more likely to attend schools that have harsh discipline policies, and hence are more likely to receive harsh discipline. But, within a given school, the punishments accorded to white and black students are generally the same. Majority black schools with severe discipline policies apply those policies in the same way to their white students, and majority white schools with more lenient policies also apply those policies in the same way to their black students (see Kinsler’s 2011 study in the Economics of Education Review, a draft of which can be found here).

There are much better approaches to school discipline than those practiced in most public schools today, but until such time as those policies become widely adopted, simply pressuring districts to mete out less severe punishments seems likely to drive down the academic achievement of the very students it is meant to help.

What are those better discipline policies and how can we encourage their widespread adoption? I offered some suggestions in my Senate testimony on the subject a little over a year ago.