« Strange firearms accident | Main | The Revolutionary origins of the Second Amendment »
Graph on correlation between "shall issue" and homicide rates
Over at No Lawyers, Only Guns and Money. Essentially, as "shall issue" rapidly spread, homicide rates rapidly fell. They draw, I think, the sound conclusion that "Violent crime is a complex issue, but national data is clear that there is no positive correlation between liberalized concealed carry laws and increased violent crime." Coincidence doesn't prove causation, but lack of coincidence certainly raises serious questions about lack of causation.
5 Comments | Leave a comment
Yes, and one could claim with equal justification, that small pox would have disappeared even more quickly due to some other factor, but the widespread use of the vaccine slowed what would have been a quicker and more complete decline. The real issue in either case would be to identify some plausible heretofore overlooked factor. I haven't heard a good one yet for either case.
I don't want to sound like I'm being argumentative, but I disagree. For vaccines, confounding factors have been rigorously controlled for. For guns, it's very hard to run a study without all sorts of differences between the control group and the sample group, and so it's very hard to control for all the other factors that can affect crime rate. This is why people like John Lott can write whole books on the topic and people like the Bradys can write whole books that totally disagree. This is why a cut-and-dried, "permits went up / crime went down" analysis is inadequate.
I always have a problem with studies that use homicide rates.
Homicide rates, especially handgun homicide rates, are especially sensitive to availability and adequacy of emergency medical care. I believe that Kopel has stated that only 1 in 6 intentionally inflicted gunshot wounds is fatal. If that's true, the handgun homicide rate could be halved if we changed that to 1 in 12 by better trauma response and care.
A better metric is the violent crime rate, which remains the same regardless of the level of available trauma care. In the various analyses of VCR versus handgun laws and permitting that I've done, there is a small but identifiable negative correlation between shall-issue CCW and violent crime: more guns, less crime.
1) The graph in question is not based on homicide rates. It states clearly that the UCC total violent crime rate is being used.
2) "permits went up / crime went down" analysis is inadequate" for what? The authors intent was very simple - demonstrate by means of descriptive statistics that increases in concealed carry availability did not translate into higher rates of violent crime (no wild west, blood in the streets scenario as predicted by some hoplophobes). On a cumulative basis that appears to be demonstrated clearly.
3) "The challenge of interpreting correlations like this is that we must consider the question, "How would the rates of homicide have been different without shall-issue," not, "Would the rates of homicide have gone up or stayed the same?" Not really. Read the rest of the post at No Lawyers Only Guns and Money for detail.
4) Utilitarian analysis such as this has its limits, and I as the author am acutely aware of them.
As much as I would like to agree with you, I just don't think it's that simple. The challenge of interpreting correlations like this is that we must consider the question, "How would the rates of homicide have been different without shall-issue," not, "Would the rates of homicide have gone up or stayed the same?"
In other words, perhaps shall-issue increased homicide rates by 10% every year, but some other factor pushed it down by 15% every year, for a net decrease of 5% per year. We might look at that and say, "Well, we can't say for sure that shall-issue made it better, but it sure doesn't seem to have made it worse!" And we would be totally wrong.