The data involving lithium has generally involved localized effects. You're talking about the general trends. And freakonomics didn't by any means solve the qu

The data involving lithium has generally involved localized effects. You're talking about the general trends. And freakonomics didn't by any means solve the question of the general trends. Here's a decent vox article that explores 16 competing theories that each have plausibility. Some of which have already been mentioned in the comments here (such as the end of leaded gasoline) https://www.vox.com/2015/2/13/8032231/crime-drop https://www.vox.com/2015/2/13/8032231/crime-drop