Update: November 15, 2017
Using data from 2014 through mid November 2017 (killedbypolice.net for 2014 and the Washington Post thereafter) Oklahoma City Police kill an average of 6.3 per year; NYPD 0.57 a year. The rate in Oklahoma City is 11 times as high. The rate per officer is 27 times higher in Oklahoma City. That means a person is Oklahoma City is 11 times more likely to be killed by police and a police officer is 27 times more likely to kill.
I've said for a while that when it comes to police use of lethal-force, an exclusive laser-like focus on race is misguided. It's is a red herring. If one actually wants to reduce police-involved shootings -- as opposed to simply being outraged at the latest incident -- there are easier ways to do this than eliminating racism and racial disparity in America. There are low-hanging fruits to reduce the overall level at which cops shoot people.
There will be the next police-incident worthy of outrage. We can go from incident to incident, outrage to outrage, and pretend it's just about race. But it's not.
I'm not saying race doesn't play a factor. This is American. And indeed, blacks make up a greater percent of unarmed people killed by police. The disparity could be racial bias; it could be related to violence in segregated America; it could be something else. Honestly, we're never going to settle the debate, and I don't know if we need to. Police misconduct doesn't only happen to blacks. And the numbers of innocent unarmed people killed by police is simply not that large. Nor is it increasing.
Police have shot and killed 706 people this year. Forty-one were unarmed. Fifteen of those were black. (Keep in mind "unarmed" does not mean no threat, and conversely somebody could be armed and not be an imminent threat.) I get the argument that murder is worse at the hands of the state. I even agree with it. I understand police need to be held accountable. But at some point the numbers matter, at least to put things in perspective.
This is a country of 320 million people. There are 765,000 sworn police officers. There are 15,000 murders (and murderers). What's an acceptable level of police-involved shooting? What's the goal? And if you're not happy addressing that question, or if you think the only acceptable answer is zero, than you're not a productive part of the solution.
Look, I know some cops do shitty things. And others make honest mistakes. But there are more cops in America than residents of Baltimore. We can and should criticize individual incidents. But we don't harp on every crime in Baltimore -- and there are a lot -- to show how the whole city is filled with evil. (And I do wish we cared a bit more about victims like Michael "Chef Mike" Bates who was just shot and killed even after he complied with the three men who robbed him.)
Does a bomb in Chelsea mean we should ban Muslims from America? (No, is the answer.) There will be the next horrific crime and the next terrorist attack just as sure as there will be the next bad police-involved shooting. Instead we're seeing something close to a moral panic, with police as the Folk Devils, we need to reduce how often they happen.
There are probably a few dozen bad (as in criminally bad) police-involved shootings a year. That's a couple a month, keep in mind. And if they're all recorded, that's one every other week. But far more numerous are shootings which may be legally justifiable but did not have to happen. They're justifiable but not necessary. We're talking perhaps something in the rage of a few hundred a year. And the bulk of these happens west of the Mississippi (see a future post). The best way to reduce bad shootings is to reduce the overall level of police lethal force.
Twenty-five percent of those who are shot and killed by police are black. Since blacks are only 13 percent of the general population, some claim this represents an "epidemic" of police violence against African Americans. But using the overall population as the denominator for interactions with police makes no sense.
America is filled with racial disparities in poverty, violent crime, calls for police service, and those who felonious kill police officers. I mean, 96 percent of those killed by police are men, and men make up less than half the population. Is there an epidemic of misandric cops gunning for other men? I don't think so. It's more likely that men are more likely to pose lethal threats to police officers.
And this brings me back to Oklahoma, where Terence Crutcher was shot and killed by a police officer even though he wasn't an imminent threat. A while back I red-flagged Tulsa and Oklahoma because I couldn't help but notice: they sure do seem to be a hell of a lot of police-involved shootings in Oklahoma. And now we have more data than we did a year ago.
We're not seeing an epidemic of police killing black people in particular in Oklahoma. The Sooner State is pretty white (72 percent, 8.6 percent Native American, and 7.4 percent black). The racial disparity in Oklahoma is pretty much in line with the rest of the nation. Since 2014, nationwide, the average annual rate of being shot and killed by police is 3.2 per million. It's higher for blacks (6.93) and lower for whites (2.37). That's a 3:1 ratio.
What we see is that more white people get killed by cops in Oklahoma than all people killed by cops in majority minority New York City. Simply put, police in Oklahoma are shooting a lot of people and the NYPD isn't. In Oklahoma, cops shoot and kill 28 people per year. In New York City, which has more than twice as many people as the entire state of Oklahoma, police kill about 5 people a year. What gives?
People in the state of Oklahoma are 12 times as likely as New Yorkers to be killed by police.
People in Oklahoma City are 20 times [11 times, see update, above] as likely as people in New York City to be shot and killed by police! New York City has about 2.5 times more police officers per capita. That means an officer in Oklahoma City is about 50 times more likely than an NYPD officer to shoot and kill somebody. [27 times, see update above]
These differences are huge! Shocking! Unbelievable!
And yet nobody seems to notice or care. [See all the states in this post.]
I assume most of the police-involved shooting even in Oklahoma are legally justifiable. I'm not saying these cops are committing crimes, but I am saying a large percentage of these shootings aren't necessary. They don't need to happen. I mean, it's likely cops in Oklahoma will always shoot more people than cops in New York City. Sometimes police have no choice but to shoot somebody. And Oklahoma isn't New York. But it doesn't have to be 12 or 20 times more. I can't conceive of how a per-capita disparity this large could be justified or explained away by any variables except police training.
So I look at the Terence Crutcher being shot, and I think: maybe that really is how police in Tulsa roll. I don't know. And I wonder what it is about NYPD training and policy that so reduces use of lethal force. Whatever it is, and I'm sure it's a combination of things, it shouldn't be that hard for somebody to copy best practices. Instead of asking what individual police officers are doing wrong (though we can ask that, too), why don't we figure out what the NYPD is doing right? We have models that work. The solution involves some combination of better hiring standards, better policy, better training, and more accountability.
Just reducing Oklahoma's use of lethal force to the national average would save 14 lives a year. That seems doable. And good. It's good for the people not to get shot. And it's good for social and racial justice. And it's also good for police officers who get to go home without killing somebody. Cops don't want to shoot people. You think Officer Betty Shelby wouldn't like to go back in time and not shoot?
And let me mention I'm only picking on Oklahoma because of the recent Tulsa shooting. Oklahoma isn't even the worst state when it comes to high levels of police-involved shootings. Currently, in 2016, it doesn't even crack the top five.
[I did some brief computations on crime (some 2015 UCR data is already out!) because police violence is best predicted by public violence. In 2014 and 2015, Oklahoma has an annual murder rate of 5.4 per 100,000. This is 30 percent higher than New York City's 4.1. Aggravated assaults and total violent crime, however, are 35 percent higher in New York City. So it seems that Oklahoma does have a violent murder problem separate from any crime problem. But nothing here would even get close to accounting for twelve- and twenty-fold differences in police use of lethal force.]
Notes: Annual rate is based on the sum total of Jan 1, 2014 to Sep 20, 2016, multiplied by 0.367.
2014 data: http://www.killedbypolice.net/
Oklahoma crime stats: https://www.ok.gov/osbi/documents/Crime%20in%20Oklahoma%2C%202015.pdf
Crime stats: http://www.criminaljustice.ny.gov/crimnet/ojsa/indexcrimes/Regions.pdf
Race data is from the Washington Post, so it starts in 2015. Annual rate is the sum from 2015 to Sep 20, 2016, multiplied by 0.58. National rates based on 318.9 million with a white population of 200 million and a black population 36 million. Feel free to double check my math. Corrections and comments always welcome.