Dear Cecil: I keep seeing a stat saying Americans are eight times more likely to be killed by a police officer than by a terrorist. How does this data break down into black, white, Hispanic, and Asian-American? Eric Ward
I was hoping you’d ask.
To hear some in the media talk, the racial breakdown for Americans killed by cops is a deep mystery. While the FBI publishes annual statistics for “justifiable homicides” by law enforcement, the race of the victims isn’t publicly available. In the wake of the Ferguson killing, nobody seemed to know how you could find out.
Vox.com, showing more enterprise than most, learned the FBI compiled “Supplementary Homicide Reports” providing additional unpublished info, including race, about slain “felons” — the FBI’s grotesque term for all justifiable-homicide victims, suggesting anyone killed by a cop is automatically guilty.
Vox obtained the report for one year, 2012. “The FBI’s data shows that 32 percent of the felons killed by officers in 2012 were black,” Vox wrote, while pointing out that black Americans make up just 13 percent of the U.S. population.
What’s more, black people accounted for 42 percent of those “not attacking when killed; not killed with rifle or shotgun.” In other words, police used handguns to kill nonthreatening blacks far out of proportion to other races.
Vox noted that people interacting with the U.S. criminal justice system were disproportionately black, which surely is key to understanding what’s going on here. However, when readers clicked on a link presumably expanding on this notion, they arrived at a Vox video titled “The Racism of the US Criminal Justice System in 10 Charts.”
Let’s stop right there.
To start with the basics:
- While the FBI doesn’t publicize the racial breakdown of people killed by cops, the information is obtainable if you know where to look. It’s kept in the National Archive of Criminal Justice Data, a public website maintained through the University of Michigan. The FBI’s Supplementary Homicide Reports (SHRs) are available from 1976 to date.
- As Vox rightly notes, the SHRs aren’t entirely trustworthy. SHRs are voluntarily submitted to the FBI by local jurisdictions, and the completeness of the data has evolved. So comparisons over time must be viewed cautiously.
- One thing jumps out when you start browsing: the number of justifiable homicides by law enforcement officers (hereinafter JHBLEOs) has been surprisingly steady over the years, fluctuating between 300 and 462. There were spikes around 1980 and again in the early 1990s, possibly reflecting jumps in violent crime in those years. Then again, we seem to be in a mini-spike now (there were 426 JHBLEOs in 2012), even though violent crime has dropped.
- Another striking phenomenon is the massive drop in the percentage of black people among those killed by cops. From 1976 to 1980, exactly half of JHBLEO victims (967 of 1,934) were black. The trend since then has been down. For the most recent five years available, 2008-2012, it’s about 30 percent. I’ve posted charts showing the racial breakdown over time in numbers and in percentages.
- Since you asked, the number of Asian- and Native Americans killed is low, usually in the single digits per year. Hispanic JHBLEOs show up in the SHRs only from 2003 on, and fluctuate in the range of 15 to 19 percent. The Hispanic fraction of the U.S. population is 17 percent.
What do we conclude from all this? Black people inarguably are killed by cops in disproportionate numbers, and are more likely to get caught up in the criminal justice system. Is that direct evidence of racism?
Not necessarily. It may simply mean there’s more violent crime in black communities.
Black people account for a disproportionate share of arrests for violent crime — in 2012, 49 percent of murder arrests, 55 percent of robberies, 34 percent of aggravated assaults, and so on. Does that reflect unfair targeting by police? Not likely. According to a Justice Department study, 47 percent of murder victims between 1980 and 2008 were black, and 93 percent of black victims were killed by other blacks. Nobody can seriously claim those numbers were cooked.
Conclusion: there’s a lot of violent crime in black communities, and thus presumably a lot of police activity. It stands to reason that, the more times people with guns are sent into a community looking for other people with guns, the more violence will result. It’s not necessary to impute this to racism.
Look again at the trend. In 1976, black people accounted for 52 percent of murder arrests, 47 percent of murder victims, and 52 percent of JHBLEOs. In 2012, black people accounted for 49 percent of murder arrests, 49 percent of murder victims, but just 30 percent of those killed by cops.
Are those horrifyingly large numbers? You bet. Is all this violence unrelated to historical discrimination? Hardly. Are some cops racist? That’s certainly one explanation for Vox’s revelation that 42 percent of non-attacking people killed by police handgun fire are black. Is the fact that, overall, 30 percent of people killed by police are black in itself evidence of racist cops?
I’m not saying such evidence can’t be found. But this isn’t it.
Send questions to Cecil via email@example.com.