The disproportionate killing of innocent and unarmed Black men and women by police forces across the United States continues today despite more than five decades of steadily integrating many of those same police forces with Black male and female officers. The centuries-long legacy of racism informing policing methods across the nation—costing the lives of untold numbers of innocent African American civilians—obviously still lingers.

But can affirmative action litigation intended to increase the presence of Black officers in the ranks of local police departments reduce the number of non-White civilian deaths at the hands of police over time? This is the question we set out to answer in our new working paper, “The Impact of Affirmative Action Litigation on Policy Killings of Civilians.” Our research examines whether the nationwide wave of affirmative action litigation in the wake of the enactment of the Civil Rights Act of 1968 reduced the police killing of non-White civilians.

Our findings are telling. Counties across the country that experienced affirmative action litigation saw fewer non-White individuals killed by police. This includes counties where litigation was not successful in leading to an affirmative action mandate. Between 1960 and 2010, we estimate roughly 70 fewer non-White deaths, and police departments facing affirmative action litigation reduced non-White deaths overall by roughly 40 percent.

The diversification of police forces, however, is only a partial solution to the harms caused by racism within U.S. police forces. Other factors, such as a legal system that continues to criminalize, to a greater degree, those living in underrepresented communities, will also influence police interactions within those communities. Above all, though, our research highlights the vital role of federal interventions in addressing police behavior and the use of lethal force in the United States.

Read the full article about police force diversification by Robynn Cox, Jamein Cunningham and Alberto Ortega at Equitable Growth.