Why White Men Get Shot by Cops
Yesterday, a man attacked a theater, dousing moviegoers with pepper spray and assaulting one person with a hatchet. The victim survived with a minor injury, but when SWAT officers showed up, the attacker went after them with a pistol and was killed. It turned out the gun, while it looked realistic, was an Airsoft model. It's a toy designed to shoot soft pellets at other people for fun.
The man has been identified as a white male with a history of serious mental illness. His name was Vincente David Montano. The first and last names are often Spanish (they can also be Italian), so he may be classifiable as Hispanic or Latino as well, depending on where his ancestors actually came from and how broadly the terms are defined.
This case may hold the key to a statistical pattern I've been pointing out for nearly a year now: Once one adjusts for murder rates, a good proxy for the most serious violent crime, whites are actually more likely to be shot by police than blacks are. The same is true if you instead adjust for rates of cop-killing. Blacks, while just 13 percent of the general population, are about 25-30 percent of those shot by police, half of murderers, and 40-45 percent of cop killers. The sociologist and former police officer Peter Moskos has noticed the same thing, as has my former National Review colleague David French.
"Suicide-by-cop," and mental illness more generally, might be a missing piece of the puzzle. Over the past decade and a half, the CDC's WONDER database puts overall suicide rates at 14.5 per 100,000 for non-Hispanic whites, 5.1 for Hispanics, and 5.2 for blacks; for serious mental illness, government survey data put rates at 4.2 percent for whites, 4.4 percent for Hispanics, and 3.4 percent for blacks. (The mental-illness survey tries to measure illness directly; it doesn't just ask about previous diagnoses.) To some degree, police-shooting numbers might reflect these disparities — with a skew toward whites and, for mental illness but not suicide, Hispanics — rather than rates of violent crime.
This is a hard theory to test more directly. We don't even know what percentage of police shootings are suicide-by-cop incidents. There are some obvious tells, such as when a suspect menaces cops with a fake gun, but in other cases suspects actually discharge weapons in officers' direction and researchers need to make an educated guess as to what the intent was. Yesterday's incident illustrates another difficulty as well — his final act was suicide, but he plainly also had the intention to harm others, seeing as he attacked someone with a hatchet. Estimates range from 10 percent to more than one-third of all police shootings' being suicides.
Good racial breakdowns of these incidents are hard to come by. Many suicide-by-cop studies focus on a single state or police department. And then there are efforts like the 2009 study that generated the "more than one-third" estimate cited above, which had a big and diverse sample ... that the authors conceded was "nonrandom." For what it's worth, in cases where the race was known, 18 percent of suicides-by-cop involved blacks, 46 percent whites, and 29 percent Hispanics. The figures for whites and Hispanics seem particularly implausible as national estimates.
The Washington Post has some helpful new numbers, though. Reporters are tracking every police killing in the country this year, and while they don't explicitly classify cases as suicide-by-cop, they do keep track of whether the person had a known history of mental illness. So far, about a third of whites killed by police did (93 of 287), compared with only about a sixth of blacks (23 of 143) and a fifth of Hispanics (19 of 91). The black-white gap in particular is substantial, though perhaps mental illness is more likely to go undiagnosed, or unnoted in police and media reports, among minorities.
The reporters are also tracking whether people were shot holding "toy" guns. (I've placed "toy" in quotation marks because they include BB and pellet guns; these are not lethal, usually, but unlike Airsoft guns they are not toys.) These cases aren't always suicide-by-cop — see Tamir Rice and John Crawford III, both black — and there aren't enough of them in the Post's data to say anything with confidence. But the early numbers are also consistent with the theory: 5 percent of whites (14 people), 3 percent of Hispanics (3 people), and 2 percent of blacks (3 people) were holding these guns.
I think the above amounts to a strong circumstantial case. And either way, this topic deserves a lot more study: If we're going to give police-shooting statistics more scrutiny — and we should — we need to find the correct baseline to compare them against. Racially, violent-crime rates skew one way, while suicide and mental-illness rates skew in the opposite direction. This makes it hard to tell whether police-shooting disparities reflect officers' bias or something else entirely.
Robert VerBruggen is editor of RealClearPolicy. Twitter: @RAVerBruggen