Another Iffy Gun Study
Over the past few years, Daniel Webster — probably the nation's most prominent gun researcher from the "public health" school — has coauthored a series of studies suggesting that laws requiring a permit to purchase a handgun (even from private sellers as opposed to gun dealers) substantially reduce gun violence. I've published a parallel series of blog posts conducting basic reality checks on the data, and I guess there's no reason to stop now.
The latest study looks at suicide in Connecticut and Missouri. Connecticut enacted permit-to-purchase in 1995; Missouri repealed its law in 2007. To figure out what would have happened without these changes, the researchers use a "synthetic state" approach. For each state being studied, they look at other states, see which ones had similar suicide trends before the law changed, combine those states into a "synthetic" version of the state, and then see what happened in the synthetic state after the law went into effect.
This approach is interesting, but it's far from foolproof, as evidenced by Webster's previous study of homicide in Connecticut. The "Synthetic Connecticut" in that study ended up being mainly just Rhode Island — and Rhode Island had an odd murder spike right after Connecticut's law passed. The finding that Connecticut's law worked boiled down to a patently silly claim that, without the law, Connecticut's rate would have spiked just like Rhode Island's.
At any rate, here are the main findings on Connecticut from the new study:
As Webster more or less conceded in an interview with the Washington Post ("This leaves us wondering ..."), the "synthetic" approach once again completely and utterly failed in this state. Synthetic Connecticut had more gun suicides and more non-gun suicides after the law went into place, so unless gun control can reduce non-gun suicides, what we're seeing is bad methodology, not the good effects of gun control.
The results from Missouri make a little more sense on a superficial level, though it's hard to take them too seriously given the above:
I'd be remiss if I didn't offer a simple chart giving the basic numbers on suicide rates in these states, relative to national trends. Here are the CDC's age-adjusted suicide rates, divided by the national rate. If you look hard and squint just right, you can see support for both theories (Connecticut suicides falling after 1995 and Missouri suicides rising after 2007), I guess.
(Minor data note: The CDC's WONDER system changed how it coded causes of death in 1999, but this doesn't seem to have affected overall suicide tallies much.)
Gun laws may well reduce suicide rates, but this study is not particularly good evidence for that notion. And the "synthetic state" method quite clearly needs a lot of work before it will be ready for prime time.
Robert VerBruggen is editor of RealClearPolicy. Twitter: @RAVerBruggen