Four years ago, coverage of the heavily armed police response to protests in Ferguson, Missouri fueled a national debate about police militarization. Police claim militarized units enhance public and officer safety. Critics claim they target racial minorities and erode trust in police.  Jonathan Mummolo wanted to know who was right. His findings have just been published in the Proceedings of the National Academy of Sciences. PBS NewsHour called this work “arguably the nation’s first systematic analysis on the use and consequences of militarized force.”

Using the Freedom of Information Act, he collected data that include records of all Maryland SWAT deployments, about 8,200 of them, between fiscal 2010 and 2014, plus an original national panel measuring SWAT team acquisition (about 9,000 agencies), combined with survey experiments. Professor Mummolo finds:

• SWAT teams were mostly deployed for non-emergencies (such as the service of search warrants). Property damage is common. Use of deadly force is rare.
• SWAT teams were deployed more often in areas with a high percentage of African American residents, even after controlling for local crime rates and other area attributes.
• When an agency acquires a SWAT team, there are no detectable decreases in violent crimes or officer injuries/deaths (within agencies over time).
• Seeing militarized police in news reports—relative to images of traditionally equipped police—may harm police reputation. For example, the public is less likely to support police funding or to want police in their own neighborhood. There is also some evidence for inflated perceptions of crime.
• These effects do not appear to vary with the race of survey respondents (though white respondents have more positive views of police at baseline).

Mummolo concludes that the claimed benefits of militarized policing—improved public and officer safety—are not, on average, supported by the data. The often-discussed trade-off between civil liberties and public safety appears in this case to be a false choice. The important policy implication of this critical work is that, while SWAT teams were created to handle violent emergencies like active shooter situations, restricting their use to these emergencies rather than using them in routine police work may benefit both police and citizens, especially citizens of color.

Jonathan Mummolo is assistant professor of politics and public affairs at Princeton University, and a CSDP faculty associate.