Race and Suicide: Shifting Trends
On a long flight this weekend, I devoured Barry Latzer's new book The Rise and Fall of Violent Crime in America. I highly recommend it to crime-stats nerds: I'm not sure it gives a completely satisfying explanation of why crime rose and fell as dramatically as it did, but the book comes close, and it's full of interesting little tidbits.
Here's one that prompted me to do some additional research. In a sidebar, Latzer explores the theory that suicide and homicide are inversely related to each other — that groups with high homicide rates tend to have low suicide rates, and vice-versa. It's a pattern I've noticed myself, and apparently the observation goes back more than a century.
However, Latzer reports that subsequent research "thr[ew] cold water" on the idea. A 1960s study of young African Americans found high suicide rates, to the point that blacks age 20-35 (of both sexes) had twice the suicide rate that white men had. Given high homicide rates in this demographic, that would seem to go against the theory.
The study's author further claimed that high suicide rates for young blacks went back to at least the 1920s, and suggested that suicide and homicide, rather than being opposites, are driven by the same tendencies. Earlier research had gone wrong by ignoring the role of age.
Even taking the factual claims at face value, I'm not sure the argument works. It's nice to have data specifically on the young, but it doesn't change the broader relationship between suicide and homicide. If groups that commit more homicide (when they're young) also commit less suicide (when they're old), we still have a phenomenon here that's worth trying to understand.
But are the factual claims even true? I checked more current numbers from the CDC. What is true is that very young blacks — those 12 and under — may have higher suicide rates than non-Hispanic whites, though suicide is extremely rare at that age for everyone. Beyond that, a gap opens up in the opposite direction.
Maybe things just changed since the earlier studies. The oldest easily accessible CDC numbers I can find are from 1968 to 1978. (Unfortunately, they're less fine-grained, with age groups instead of specific ages and no way to filter out Hispanics.) Here are the results:
The gaps for the young are much smaller here, which is interesting. (One factor is the recent rise of mortality for whites, which remains poorly understood.) However, it still seems a stretch to imagine black young adults' having twice the suicide rate of comparable whites within a decade of the period these numbers come from. The country obviously changed a lot between the 1960s and the 1970s, but my guess is that the local data the earlier studies relied upon was flawed.
At any rate, these trends, and the suicide-homicide connection in general, deserve more study.
Robert VerBruggen is editor of RealClearPolicy. Twitter: @RAVerBruggen