Did a Connecticut Gun Law Work?
(See bottom for an important update.)
A new study says the state's 1995 gun-control measures (including requiring a permit to purchase a handgun and raising the handgun-purchase age to 21) cut gun homicides by 40 percent over the next ten years. There was no reduction in non-gun homicides. Apparently the researchers estimated what Connecticut's crime rates would have been without the law by looking at similar states. Unfortunately I don't have time to give the study the thorough treatment it deserves (I haven't even gotten my hands on the full paper, which is paywalled), but I was able to perform a quick reality check on its central claim.
First of all, forget "gun homicides." They can be a useful statistic sometimes, but at the end of the day we want to see a reduction in the total number of dead bodies. If gun homicides fall but non-gun homicides rise — because killers choose other weapons instead of refraining from killing, for example — we have not accomplished anything. Furthermore, gun homicides are two-thirds of homicides nationwide and about two-thirds of murders in Connecticut as well (even long after the 1995 law), so a decline of 40 percent is big enough that it should show up in the overall murder rate unless non-gun killings rose.
This is what we get when we look at the murder rate in Connecticut as a proportion of the nationwide rate. I've included two bordering states as well for reference:
We can always have fun trying to tell a narrative about a graph: Look, there really is a drop in 1995, and the early 2000s look pretty good! No, actually, that was just falling back to the previous level! And what about the late 2000s? But with such a startling alleged reduction in gun homicides I was hoping to see a more obvious effect in the murder rate.
[Update: John Lott has a lot more data on Connecticut's crime rates.]
Robert VerBruggen is editor of RealClearPolicy. Twitter: @RAVerBruggen
Another update: I told myself I wasn't going to get sucked into spending a lot of time on this, but the Washington Post has some more key details. First of all, the "synthetic Connecticut" they used to estimate what crime would have been like without the law is "mostly Rhode Island." One glance at my chart above shows the problem with that.
The Post also has this chart from the paper, showing gun homicides in Connecticut, in "all control states" (the 39 states without permit-to-purchase), and in the synthetic Connecticut. Make of it what you will; to me, it looks like Synthetic Connecticut is being driven by the weird crime spike in Rhode Island, while Real Connecticut doesn't look all that different from the bigger control group (though the authors claim there's a bigger drop).
Okay, one last update: The Post writes that "Synthetic Connecticut" is "mostly Rhode Island, with some Maryland, and traces of California, Nevada and New Hampshire."
Here are the overall murder rates for those states (with Connecticut itself included too), divided by the national rate:
These states may have had similar crime trends to Connecticut's before 1995, but it's hard to say why; they're geographically spread out, they're politically different, and they have a wide range of murder rates. And while Rhode Island had an odd spike right after Connecticut's law went into effect, Maryland (the next-highest contributor to Synthetic Connecticut) had a rise in crime as Rhode Island cooled back off. The idea behind the study appears to be that, if not for Connecticut's gun law, the state would have shared both of these crime waves.