Should we have unarmed police?

Apropos of the recent protests regarding Americans killed by police… Now that the crime rate has fallen so much in the U.S., why continue to arm the typical police officer? It is true that we are a nation of gun nuts, but it is still a minority of Americans engaged in criminal activity who carry guns, right? Why should every police officer bring a gun onto the scene? That would seem to invite a huge escalation of the violence, either with the officer afraid that the suspect is going to grab the gun or that the suspect might choose to shoot him or her before the gun can be pulled out, etc. The British seem to manage with the first line of law enforcement being unarmed with deadly force. Is it crazy to think that it could work here? (The Economist did a comparison of shootings by police in Britain versus the U.S. in an August 15, 2014 story.)

Related: My October 2014 posting about armed police approaching a stalled-out car.

26 thoughts on “Should we have unarmed police?

  1. Our overall gun homicide rate is 30 times the rate in England and Wales, and the rate among African Americans is 4 times that. It seems to me that really limits the applicability of their experience to our situation. With all of the concern about police shootings lately it is worth noting that the risk of being killed by the police is still much lower than the homicide risk for police officers. A national use of investigation system (focused on root cause analysis rather than finding culpability) could help reduce police killing further.

  2. “With all of the concern about police shootings lately it is worth noting that the risk of being killed by the police is still much lower than the homicide risk for police officers.”

    The Economist article that I cited says that “In 2013, 30 cops were shot and killed—just a fraction of the 9,000 or so murders using guns that happen each year.” That’s 0.33 percent of the murders. http://www.bls.gov/ooh/protective-service/police-and-detectives.htm says that there are 780,000 police and detectives in the U.S. The total number of employed Americans is about 147 million. So police are 0.53 percent of the workforce. (It is a little tough to get an exact comparison because we don’t know how many of the 9000 murders cited by the Economist were of employed versus unemployed individuals.)

    Conclusion: A police officer is not substantially more likely to be killed than an average American worker, consistent with the fact that American police officer is not a statistically dangerous job.

    The Economist says that “In 2012, according to data compiled by the FBI, 410 Americans were ‘justifiably’ killed by police—409 with guns. That figure may well be an underestimate. Not only is it limited to the number of people who were shot while committing a crime, but also, amazingly, reporting the data is voluntary.” So the police aren’t doing the majority of the killing in the U.S., but it is not a statistically insignificant risk. Given that police officers are only about 0.24 percent of the overall population, if they are shooting and killing roughly 1 in 20 of the Americans who are shot and killed, the risk of being shot and killed by a police officer is about 20X the risk of being shot and killed by an average American.

  3. Simple answer to a complex problem, no. Unarmed police will not work in US. And given the gun culture and history of the US it’s politically unthinkable to even broach the question.

    However, police must be held accountable for their actions. One solution may be to require each state to establish a special prosecutors’ office to investigate and prosecute police for suspected crimes committed while acting as a police officer. The necessarily incestuous relationship between the normal public prosecutor’s office and the police is a conflict of interest on its face.

    In the end, it’s the tax payers that pay wrongful death lawsuit awards. Eric Garner’s family (Staten Island) plans to file a wrongful death lawsuit against NYPD for $75 million. I think they will win.

    Civil police is a very difficult job to do right. And too often the quality and character of police personnel are not up to the job. The salary, benefits, and very generous retirement packages offered to police today should allow recruitment of a better brand of law enforcement people. One can hope.

  4. The British are unique in not arming their police. Yet other countries like France or Germany that arm their policemen do not have anywhere near the incidence of police shooting civilians as we do (by 2-3 orders of magnitude). Or even Canada, to look at a more directly comparable country.

    The prevalence of guns in the US means disarming the police is not really viable, but the total impunity of policemen and lack of accountability when they shoot someone means they have little incentive to moderate their trigger-happy ways. I’m not a knee-jerk police-hater BTW, my grandfathers were a judge and a deputy chief of police respectively.

  5. 1. Note that Eric Garner was choked to death so firearms per se are not the problem. Problem in Garner’s case was greedy nanny state – what he was doing should not have been a crime in the 1st place. At most a ticketable offense, if that.

    2. US is still more violent than UK (although they are catching up) so it is still not a fair comparison. Different cultures and traditions.

    3. US has a 2nd Amendment so citizens have a Constitutional right to bear arms. Police would want to be at least as well armed as the citizens.

  6. My answer would be yes, absolutely. And the other thing we should have is an absolute requirement of continuous audio and video recording of every police officer while on duty. It would be easily within the capabilities of current technology, far less expensive than the frequent factual disputes over police conduct, and a long-overdue deterrent against deliberate police misconduct.

  7. Fazal,

    Police in France, Germany, and Canada are killed at orders of magnitude lower rates than in the US. Its not that surprising they don’t need to kill as often. I’m also guessing (no data) that citizens in those countries tend to comply with police instructions. A common factor in the events currently in the news is that police were managing non-compliant individuals which obviously increases risk. I’m not saying there aren’t problems with policing in the US, but what is driving the numbers is the overall level of violence in our society. Of course the police must own their own actions, but we can’t put it all on the police.

    Izzie (1),

    Perhaps NYC doesn’t have the rules exactly right, but it is still better to have the greedy nanny state manage this kind of problem than to have the local merchants hire Tony Soprano to convince people like Eric Garner to move their business elsewhere. That would be even worse for the Eric Garner’s of the world.

  8. Jack,

    Additional advantages of audio and video recording by police are that it moderates citizen behavior lowering risk and discourages frivolous police abuse complaints reducing noise in the complaint analysis system.

  9. No doubt there is a problem. I think there are two relatively simple changes that could be made to reduce the number of people killed by police guns:

    1) Give the police another tool they can trust. I don’t have the statistics, but it seems that a lot of people killed by police guns are threatening police with a weapon other than a gun. It makes sense to respond to a gun threat with a gun, but using a gun to neutralize an attack by to a knife, screwdriver, fiats, or car feels old school. There must be better technologies to deal with these other threats.

    2) Police (and armed civilians) are taught to use guns (deadly force) to respond to threats of death or great bodily injury. I think “great bodily injury” is not defined well enough. Of course, when presented with a real crisis, an officer should not be making a sophisticated analysis “,is this great bodily injury, or not?” However, if that term was better understood during training, cops would be less inclined to always turn to their gun when they felt threatened. In an interview, Darren Wilaon said something like “I was thinking ‘am is legally justified in shooting this giuy?’l This clearly should not be the question an officer should be asking. This suggests that the officer’s preference was to shoot the guy. I think the officer should be asking “What is the appropriate tool to neutralize this threat”?

  10. philg,

    According the Bureau of Labor Statistics ( http://www.bls.gov/news.release/cfoi.nr0.htm ) the rate of fatal injuries for U.S. workers in 2012 was 3.4 per 100,000. According to the FBI ( http://www.fbi.gov/news/pressrel/press-releases/fbi-releases-2012-statistics-on-law-enforcement-officers-killed-and-assaulted ) 95 police officers were killed in the line of duty in 2012. Using your number for the total number police and detectives gives us a fatality rate of 12.2 per 100,000.

    According to the BLS (http://www.bls.gov/news.release/cfoi.t01.htm) 381 American workers were killed by guns in in 2012. Using your number for the total number of American workers this gives a gun fatality rate of 0.26 per 100,000.

    According to the FBI (same source) 44 police officers were killed by guns in the line of duty, giving a gun fatality rate of 5.64 per 100,000.

    Conclusion: The occupational fatality rate for police officers is 3.6 times the rate for an average american worker and the risk of being killed by a gun is 21 times higher for police officers than for the average American worker.

  11. philg,

    Even using the larger estimate of 1,600 police shootings per year I have seen in the media gives a police shooting fatality rate of 0.51 per 100,000, which is an order of magnitude lower than the gun fatality rate for police officers I posted previously.

  12. Anonymous: Thanks for that BLS link. It is interesting that the number of people who kill themselves at work (out of despair at being parked in a cubicle?) is close to the number who are shot and killed by someone else (and the gap narrowed between 2012 and 2013).

  13. (But I am not sure it is reasonable to infer that an unarmed police officer is more likely to be shot and killed than an armed one. If the police officer is armed there is always a gun close to him or her that a criminal can grab and then shoot with. Also, a criminal might reasonably try to shoot an armed police officer before that officer can pull out the gun and shoot the criminal.

    http://www.houstonchronicle.com/local/investigations/item/Bulletproof-Part-1-Unarmed-and-Dangerous-24419.php says “Officers shot unarmed civilians who ‘reached’ or ‘grabbed’ for their waistlines …”

    http://www.bbc.com/news/magazine-19641398 discusses the pros and cons of this issue from a British perspective: “There’s a general recognition that if the police are walking around with guns it changes things,” says Richard Garside, director of the Centre for Crime and Justice Studies. Arming the force would, say opponents, undermine the principle of policing by consent – the notion that the force owes its primary duty to the public, rather than to the state, as in other countries.)

  14. I should clarify that while being a police officer is among America’s more dangerous occupations, it is much safer than the most dangerous occupations (e.g. logger, roofer, taxi driver).

    philg: My point was not that armed officers are less likely to be shot than unarmed officers (I agree that can’t necessarily be inferred from first principles). My point was that police in America are working in a significantly different environment than British police. I am not sure it is reasonable to infer that American police could safely and effectively police America without guns just because British police are able to do so in Britain.

  15. To further contextualize the gun risk for police officers, their risk of being shot and killed on the job is roughly twice their (outrageously high) risk of being killed by a gun because they are Americans. In other words, becoming a police officer roughly triples one’s chance of being killed by a gun. I wanted to add this because 3x gives a different impression than the 21x I cited earlier (both numbers are correct, they just measure different parts of the same picture).

    I also wanted to expand on the reasoning behind my call for a National Use of Force Investigation System. Part of the problem with how we as a country are approaching this issue is that we want to categorize each police killing as either a “crime” or as “justified”. The reality is there is a big space between “this officer committed a crime” and “there is nothing wrong with this picture”, and that space provides a lot of room for improvement.

    Finally, I don’t want to give the impression that I dismiss Phil’s original suggestion of disarming our front line police. It is an interesting idea worthy of serious consideration.

  16. philg:

    >the risk of being shot and killed by a police officer is about 20X the risk of >being shot and killed by an average American.

    The risk of being shot and killed by a police officer is lower than the risk of being shot and killed by an average American (the total number of Americans killed by police is lower than the total number killed by average Americans and the denominator is the same for both rates). I think what you mean is that police shoot and kill people at a much higher rate than average Americans do (by my estimation that number is more like 50X -100X than 20X).

  17. Phil,

    Have you considered the relatively low murder rate for cops could stem from the fact that since cops are armed, a perp may hesitate in trying to attack a cop whom they know is armed?
    I’d bet the murder rate for U.S. cops would increase markedly if they went unarmed.

  18. Mark,

    It is true there are some tactical scenarios where an armed cop is safer than an unarmed cop, but those may very well be balanced by different scenarios (some described in the original posting) where the cop and/or the citizen are safer if the cop is unarmed.

    One problem the posting doesn’t address is that disarming police makes it much more difficult for them to safely apprehend an armed criminal intent on escape. The British seem to have an effective system for dealing with this scenario, but it is unclear whether their system would be cost effective here because we have so many more armed criminals.

    This in turn creates an incentive for criminals to carry guns when they might not otherwise do so because a gun makes it much easier for them to evade arrest by an unarmed police officer. This additional incentive is much more problematic in America than in Britain because it is so much easier for criminals to (legally or illegally) obtain and carry a gun in America than in Britain.

  19. Anonymous:
    Chicago has tough gun laws. They also have a tremendously high murder rate.
    I would appreciate hearing a logical answer to this seemingly paradoxical fact.

  20. Folks: I’m not sure that it would be that tough for a criminal in Britain to get a gun, especially a shotgun. See http://www.telegraph.co.uk/sport/9446474/UK-gun-owners.html ; I think it is more that a lot of routine work, either as a criminal or as a police officer, doesn’t require a gun.

    Here’s an example from our forthcoming book. A University of Pennsylvania grad was married for four years. Then she decided that she wanted the freedom to spend $235,000 per year in tax-free child support, plus a free house, all child expenses paid, and taxable alimony. She launched Kosow v. Shuman in 2010. Like most of the other Massachusetts custody and child support plaintiffs that we learned about, she alleged that the defendant was somehow abusing her and the child. This necessitates police and/or Department of Children and Families involvement.

    On June 8, 2010 she called the Weston Police Department to the house [Weston is one of the wealthiest suburbs of Boston] “to request an emergency restraining order against her husband … on the basis of verbal argument and alleged past sexual abuse.” It turned out that Grandma and Grandpa were actually at the house at the time of the alleged abuse. From the police report: “We then asked [the plaintiff] if she had any where to go and she said no and when asked she said she did not have any money. … About 2230 hrs [the plaintiff] came home and started yelling at [the defendant] and calling him a rapist, pedophile, and molester. [The defendant] stated that he did not say anything and both of his parents agreed to that. … I think informed [the defendant] and his parents that [the plaintiff] stated that she did not have any place to go and [the defendant] quickly stated that he had just bought her a million dollar house … I then asked informed him that [the plaintiff] said that she did not have any money and all 3 laughed out loud all at the same time. I then asked what was so funny and [the defendant] told me he had just given her $25,000. … Sgt. Kasprzak arrived to the scene with a[n] abuse prevention order and had [the plaintiff] fill out the affidavit. After it was filled out Sgt. Kasprzak called the on duty Judge Jeffery Locke. The Judge was made aware of the fact that [the parents] have a divorce hearing on Friday 6/11/2010 in Cambridge, .., no physical abuse occurred tonight, the 2yr old was not in any danger, [the defendant’s] parents were staying in [the] house. … At this time the emergency 209A was denied. While Sgt. Kasprzak explained this to [the plaintiff] she was lying on the couch with her feet up, doing something on her phone, she then went upstairs.”

    —————– end of excerpt

    So the cops just had to go to a child support plaintiff’s house in a rich neighborhood before the 11 o’clock news came on. Did they really need to be wearing guns for that? Suppose that during this activity a prankster 14-year-old had jumped out of the bushes. Because they have guns in their belts they end up shooting the 14-year-old.

  21. Actually I think that for police, domestic violence calls can be quite dangerous. Emotions run high and sometimes one or both of the parties end up attacking the police officer. I live in what is a very nice suburb overall, where most of the residents are middle class or above professionals like the Shumans and would not dream of attacking a police officer, but there is one section of the township that is literally on “the wrong side of the tracks” where the homes are much more modest and the occupants more blue collar. Should the police take their guns when calling on homes in that section but leave them off when they visit the upscale neighborhoods?

  22. Izzie: What I referenced wasn’t a “domestic violence call”. It was just a “check-the-box” procedure for a Massachusetts child support plaintiff (since it was a matter of public record when Ms. Kosow became a child support plaintiff, the police could have known what they were likely to find before leaving the station). Police officers in Massachusetts have to show up at traffic accident scenes and create reports. They show up to homes that have been burglarized, many hours after the theft, to take down a report. They also are hired to direct traffic when utilities are digging up the streets. If, without a gun, they cannot be safe while carrying out these activities, why are the rest of us safe? Why don’t we all have to carry guns all the time?

  23. Mark,

    I don’t know anything about Chicago gun laws or Chicago crime rates. One possible explanation for the paradox you present is that because guns are easily concealed and transported, gun laws imposed at the regional level are not particularly effective. I don’t understand what your question has to do with the rest of this thread.

    philg,

    >Did they really need to be wearing guns for that?

    My vote would be “no”. However, even with crime rates plummeting the FBI reports that there about 300,000 murders, robberies, and assaults committed using guns in 2013. On average, each of the 780,000 police officers in the U.S. has a nearly 40% risk of being called on to respond to such an incident (per year). If these crimes were committed here at the same rate as they are in Britain those numbers would be 10,000 and 1.3% respectively. With Britain’s numbers we could disarm 97% of our police officers and still maintain a secondary armed force whose members would face a 40% risk of being called to respond to a serious gun crime. Unfortunately, we don’t have Britain’s numbers.

    I get that police in Weston respond to gun crimes much less than that average, but they also kill people a lot less than average.

  24. Police go from activity to activity – one minute they may be sitting at a road construction site eating donuts (I’ve never actually seen them direct traffic – they just sit in their cars) and the next minute they may be called to a robbery in progress. It’s not inconceivable that in some relatively safe suburban areas they could be convinced to keep their guns in a locked box in their patrol cars, but this is not the American tradition and good luck getting the police unions to buy into this.

    Although there have been a few highly publicized cases recently where the police have shot someone without proper justification, this is really not a big societal problem. Out of the 30,000 gun deaths each year in the US, around 60% are suicides, 1/3 are homicides, 4% are accidents and around (its hard to find good stats) 2% are homicides by police, a major % of which are clearly justified by any standard – e.g. they are being shot at and are shooting back (legally almost 100% are deemed justified). So how many civilian lives would be saved by taking the guns away from police (and how many more police would be killed if the were disarmed)? You would disarm 780,000 police officers in order to prevent a handful of unjustified homicides.

    People in Weston are in very little danger to begin with. When was the last time the police in Weston shot anyone, let alone unjustifiably? The average US police officer will shoot and kill someone once every 1,300 years (780,000 officers kill ~600 victims annually). Probably the greater danger for you is that you will be hit by a police car (maybe we should take the cars away also?). Most of the questionable shootings are of young racial minorities in poor areas, where the police feel like they are in constant danger. Teens in these areas are already in a lot of danger themselves, but mostly from their own neighbors, not from the cops. Your hypothetical teen prankster jumping out of the bushes (to surprise a cop – is he nuts?) is just that – he doesn’t exist.

    Whenever a major frenzy is stirred up in the US (or anywhere) over something that is not as big a problem as it is made out to be, you have to wonder whether you are being manipulated by the press and by people with a political agenda and why? It’s easy to see this in other lands – when Russian TV suddenly aired tons of stories about how ethnic Russians were being oppressed in Ukraine, it was easy for us to see that this was just politically orchestrated propaganda. But when US media is suddenly full of stories about police shootings and fraternity rapes, we don’t have the same skepticism.

  25. “Chicago has tough gun laws. They also have a tremendously high murder rate.”

    If there is a causation here, I suspect that the high murder rate brought on the gun laws.

    And this brings up what I think is an interesting part of the gun control debate. People in dense urban areas are probably in fact safer with strict gun laws. And people in remote rural areas may well be safer with easier access to guns. So the seemingly contradictory views may be rational for both groups.

Comments are closed.