> “While we cannot entirely rule out the effects of under-reporting,” the authors wrote, “our results show that crime complaints decreased, rather than increase

> “While we cannot entirely rule out the effects of under-reporting,” the authors wrote, “our results show that crime complaints decreased, rather than increased, during a slowdown in proactive policing, contrary to deterrence theory.” Original research is here: https://www.nature.com/articles/s41562-017-0211-5 https://www.nature.com/articles/s41562-017-0211-5 Both authors are political scientists who are highly unlikely to have a command of econometrics. I am not sure if they took into account the panel nature of the data and in those circumstances where authors focus solely on either the cross-section or the time-series nature of the process, it is not uncommon for "results" to be nonsense.