I agree with Radley Balko that - other things equal - less intrusive policing is better.
And of course we can all agree that it’s wonderful that Dallas, like other cities, now has an historically low murder rate.
But it’s also worth noting that - according to the news story Balko’s post links to - “historically low” for Dallas means 9 homicides per 100,000 population. The comparable figure for New York City is 4 per 100,000.
Does that prove that more intrusive policing (or a better social safety net, or higher alcohol taxes, or the availability of real bagels) reduces homicide? Of course not. But critics of NYPD tactics - including the undersigned - also need to acknowledge the truly spectacular drop in crime that coincided with the COMPSTAT era.
Footnote These figures are also unlikely to be cited by the “more guns, less crime” crowd. But it’s worth noting that NYC has not only an unusually low homicide rate, but an unusually low ratio of gun killings to total killings.
New York has notoriously ferocious gun laws, along with aggressive policing which - among its other effects - makes it much more likely that carrying an unlicensed firearm will lead to arrest. One of the striking findings from the crackdown on turnstile-jumping was the relatively high percentage of the arrestees who were illegally armed. But the even more striking finding was how much that percentage decreased after the crackdown started.
Can you say “deterrence”? I thought you could.
Well, if crime reduction is the only concern…
Certainly aggressive policing will act as a deterrence, but such tactics tend to hit a lot of false positives (citizen harassment, arrests for minor crimes not associated to violence, etc.).
I'm with Madrocketscientist - you've got to weight the false positives issue there as well, especially since this is ostensibly also about "quality of life" policing and constantly stopping people (particularly non-white folks) is a major hit on quality of life.
Plus, that's a pretty small absolute difference even if the relative difference is big.
A short course in homicide rates would be useful; if anyone knows of good reference, please post. What I think I know, derived in large part from memories of Pinker's "Better Angels," http://www.amazon.com/Better-Angels-Our-Nature-Vi… is as follows. The four per 100,000 that you cite for NYC is hugely improved in a couple of decades, pretty good in the sweep of human history, and similar to or slightly better than the U.S. national rate; much of (largely gun-free) Western Europe is at one or two, pretty much the best ever known; and nine is definitely nothing to brag about. Note Balko's hat tip to Kevin Drum and the lead theory.