The second study is by Stephen Rushin, a legal scholar, and Griffin Edwards, an economist. In a paper soon to be published, they examine what happens to local crime rates when police departments come to the attention of the federal government for systematically violating civil rights, or, in more extreme cases, become subject to federal oversight and reform. Because such oversight is a form of criticism, it offers a test case of the thesis that increased scrutiny of the police leads to more crime.

Since 1994, when legislation was passed making federal oversight possible, 61 police departments around the country have been investigated (out of 18,000 agencies). Thirty of these investigations didn’t go anywhere. The rest led to settlements or consent decrees in which the departments agreed to change how they did business: how officers used force, how citizen complaints were handled and so on.

The researchers found that in communities whose agencies were only investigated, there was no change in crime rates. By itself, scrutiny of police departments doesn’t seem to move the needle when it comes to crime — contrary to the Ferguson effect thesis. On the other hand, when departments were required to alter their policies, signifying stronger public concern, crime was higher (temporarily) and officers spoke of working in a low-morale environment.