Police Killings: Is the U.S. an Outlier?

Police Killings: Is the U.S. an Outlier?

For the first time in U.S. history, we have a good count of how many people were killed by the police. The Guardian estimates that number at 1,136 in 2015 (1,012 were shooting deaths). The Washington Post, using a narrower definition, counted 965.

Much attention has been paid to the demographic breakdown of those killed. But few have made a serious effort to put the numbers in international perspective. When international comparisons of fatal police-involved shootings (FPIS's) are made in the media, often differences in population or violent-crime rates are acknowledged but are not taken into account in whatever numerical comparison the article makes.

There is a simple way to make a fair comparison: Choose a violent crime that serves as a good proxy for the threat that police officers face, and divide the rate of FPIS's by the rate at which that crime is committed. This takes into account differences in both population size and violent-crime rates between countries.

It's not immediately clear which measure of violent crime is best. Three obvious candidates are homicide rates, firearm homicide rates, and police homicide rates. To my mind, firearm homicide rates are superior to homicide rates because they not only represent the amount of violence in a society, but also indicate how heavily armed the perpetrators of violence tend to be. We'd expect police to do more killing as criminals become better armed, even holding the overall homicide rate constant. Police homicide rates would also be a good measure to use, but good international data do not seem to be available. So firearm homicide rates it is.

Most of the homicide data I used come from the United Nations Office of Drugs and Crime (UNODC); I averaged the available rates for the years 2005-2012. But there is no single source of international FPIS data — many developed countries, including the United States, don't keep good records of these incidents. So, I gathered estimates of FPIS's from government and media sources. (For more information on sources and data, see this document. The numbers are available in this spreadsheet.) Inspired by a Guardian article, I was able to find reliable numbers for six developed countries: the U.S. (from The Guardian's own project), Australia, Canada, England and Wales, Finland, and Germany.

Before comparing FPIS's, it's informative to note the large gaps in population size and gun violence between the United States and other countries. For example, the U.S. population is four times the population of Germany, nine times the population of Canada, and 14 times the population of Australia. More surprising are the large gaps in firearm homicide rates. The U.S.'s rate is about 5.5 times Canada's, 12 times Germany's, 18 times Australia's, and 34 times England and Wales'. No sensible comparison of FPIS's can be made without taking into account these huge disparities.

With such large gaps in population and gun violence between countries it is hardly surprising that American cops kill more people. When we take these factors into account by calculating the number of FPIS's per firearm homicide, the data tell a much different story than typical media reports. Contrary to claims that the United States is "the outlier" in FPIS's internationally, after taking firearm homicides into account, its rate is roughly equal to that of Australia and Canada.

However, even the violence-equated rates do not absolve American police of criticism. FPIS's in the U.S. are still roughly two times as likely as in England and Wales, three times as likely as in Germany, and 10 times as likely as in Finland.

Reports that the United States' FPIS rates are astronomically higher than those of all other economically developed countries are misleading. At a minimum, population size and the threat of violence that police face must be taken into account to make a fair comparison. This said, law enforcement in the U.S., Australia, and Canada may have much to learn from other developed countries with much lower rates of fatal police violence.

Robert Schwartz writes about politics and public policy. Twitter: @R_Shwortz

Comment
Show commentsHide Comments

Related Articles