Police Violence- The View From 10000 Feet
A rigorous perspective for those struggling to find their place in the police violence debate
In debates on policing in America I am often torn. On the one hand, I see vivid incidents of innocent Americans shot dead by police. Each time one of these horrific moments occurs- my first thought is that policing in America is broken- it needs to be pulled out, root and stem, and replanted with a better alternative.
But, then I hear horror stories of street violence, of unsolved murders, of communities ripped apart- much of which could have been avoided if we had more, not fewer, police. And this gives me pause. If we reconstruct policing in America, we might be providing a cure worse than the disease.
For those who relate to this line of thought- I think the only solution is to reframe the argument on rigorous, neutral terms. Instead of going back and forth between these emotionally stirring examples we should ask the essential question:
How bad is American policing?
Does it merit reform or complete overhaul?
Let’s take the first part of that question: how bad is American policing? Every year Americans are killed by police. But, the scale matters. There are, no doubt, police killings in other first-world countries. So, how much worse is American policing in particular?
This answer then should push us towards a preferred solution for American policing. Why is this the case?
Let’s say there is a fire department in a major city. But, every year or so, a single fireman is caught murdering a household pet. Would we disband the department, leaving thousands vulnerable to devastating fires while we rebuilt that public service as a whole? Likely not.
But, let’s imagine a different fire department. For some reason it’s heavily populated by sadists. Now it’s not just a single pet a year. Every year, 1000 family pets are murdered by the city’s firemen. Yes, these murderous firemen are still a minority of the force. But, we wouldn’t simply reform the department. That large amount of cruelty can’t be borne, and we’d simply need to maintain a stiff upper lip and live with more fires while we reconstructed the department from scratch.
So, there is some amount of police killings that justify a total reconstruction of policing in America. And there is also some amount of killings that justify reform.
Where is the threshold?
I’m not going to provide an answer in this article. But, I hope to lead you to your own.
First, let’s set a benchmark. Most European and Oceanic OECD countries have a high quality of life and, just from reputation, a highly peaceful system of social order.
So, if America had the same number of police killings per capita as an average of these countries, then we should believe that America likewise has a peaceful system of social order. This would push us toward reform, not reconstruction.
It’s not so simple though. Police killings don’t exist in a vacuum. If a country has a higher rate of violent crime, then it’s likely that their police have to confront more violent situations more often. In this situation it might not be the police to blame for these killings, but the high crime rate.
So, its not police killings per capita that we should measure. Its police killings per capita- taking into account the average rate of violent crime.
If we compared a representative assortment of European and Oceanic OECD countries and looked at how they did on this measure, that would be our benchmark.
Now, here’s the tricky part. How much worse should police killings be in the U.S. to justify a reconstruction of the police.
What if American police killed 50% more people given the U.S. crime rate?
What if American police killed 25% more people given the U.S. crime rate?
What if American police killed 300% more people given the U.S. crime rate.
Policing in America stirs up strong emotions, but if you’re still following me, I’m asking you to try to neutrally engage in this exercise. What is your threshold? How bad does this metric have to get before you think we should reconstruct policing in America?
Got your answer? Now I’ll give you the stats.
I used statistics from the World Population Review, incorporating all of the European and Oceanic OECD, which had available data for police killings and homicides per capita (see footnotes for which countries were included- but I don’t think there’s obvious bias in this sample)1.
The weighted average of the police killings in these countries was 2 police killings per 10 million people. The weighted average for homicides in these countries was .92 per 100,000 people. So killings divided by homicides give a benchmark number of 2.19.
In the U.S. there are 28.4 police killings per 10 million people. But, the U.S. also has 5.35 homicides per 100,000 people. So killings divided by homicides give a number of 5.3.
Now there are many other perspectives to look at how bad police killings are in America. But, with this metric-
America has roughly 150% more police killings than European and Oceanic OECD countries.
Is that bad? If so, how bad? Think back to what you thought in the previous section. But, at least we’re now using some neutral rigor to see this problem in an at-large perspective.
To my mind, these numbers push towards reform- not reconstruction. If we had 150% worse pollution than Europe we wouldn’t burn down factories, we’d restrict them further. If we had 150% more drug overdoses we’d look at new laws, not at a ruthless reorganization of the DEA.
But, you can disagree with me on that. My threshold might be too low. But, at least we’re talking about this problem with the perspective it deserves.
Here are the countries I used for the comparative statistics: Australia, Canada, Finland, France, Germany, Japan, Netherlands, New Zealand, Norway, Poland, Portugal, Sweden, United Kingdom.
The stats are pulled from these links:
https://worldpopulationreview.com/country-rankings/police-killings-by-country
https://worldpopulationreview.com/country-rankings/murder-rate-by-country