do gun factories kill people?

Prison Law Blog, a blog I enjoy very much, today posts the following graph to support the proposition that guns “cause” homicide, in the sense that the more guns are produced, the more homicides will follow.

The problem I see with that proposition, at least as supported by this graph, is that the sharp rises in gun production seem to follow, rather than lead, increases in homicide. This tends to suggest to me that increasing homicide rates drive demand for guns, which in turn drives production.

And how do we read the fact that in the period at the end of the chart — about 2004 onward — gun production has swung upward sharply, while homicide has steadily declined? Of course, it might be possible that increased gun production is leading the coming wave of homicides, and we will soon see the result of all those extra guns in circulation. But it’s also possible that other social forces have been driving gun purchases upward recently even in the face of a decline in homicide (and crime generally). (Increased political polarization during the Bush years and economic insecurity among the 99% would be my guess for likely candidates.)

(The same post also links to a this UN report, which apparently contains a neat chart showing the correlation between gun availability and homicide rates by country. I admit I haven’t looked at the report yet, but of course the same causality problem raises its head — how do we know countries with higher homicide rates don’t have incentives to liberalize gun laws? Or both variables might be tied to other cultural factors not shown. I’m not being flip — it’s just really, really hard to nail down the direction of causality arrows on these issues.)

This entry was posted in Uncategorized. Bookmark the permalink.

5 Responses to do gun factories kill people?

  1. peterwagner says:

    Causation is a tricky thing, that’s true. But when we have a correlation, what role should the establishing of norms have? What I find so frightening about Stand Your Ground is not its direct effect on crime or safety, but the larger message that it sends about the acceptability of using weapons. (And that is, unfortunately, why the NRA is so enamored of the laws.

  2. Eric says:

    Hrm. I haven’t adequately thought this through, but I do feel obligated to point out a technicality: the chart and the related articles aren’t correlating homicides to gun availability, they’re correlating gun homicides to gun availability. Which may or may not change anything. Do gun homicides displace knife or lead pipe homicides, for instance? Or do gun homicides stay high (or increase) while the general drop in homicide rates reflects decreased usage of wrenches and ropes?

    I might have to admit that I don’t know that the additional information would have a huge influence upon my thinking about gun control and the Second Amendment: I’m inclined to think that gun ownership for sport and substinence is very justifiable while gun ownership for other purposes declines in value (there may be exceptional exceptions: owning an authentically historic firearm for purposes of historic re-enactments, for instance, might be justifiable even if it’s arguably frivolous); it also seems to me that self-defense is a lousy reason to own a firearm and that it’s generally hard to justify the acceptance of handguns we have in this country (say, as opposed to rifles). And regardless of what kind of gun control/freedom might be rationally justifiable, I think the Second Amendment is an outmoded artifact of our Constitution’s origins: i.e. it’s part-and-parcel of the Constitution’s quaint and obsolete “standing armies are evil and the only protections we need are the Atlantic Ocean and our state militias” approach to national defense combined with an even more clearly obsolete ideas about revolutions (surely the Federalism-States’ Rights-The People balance was sorted out sometime between Washington crushing the Whiskey Rebellion and Lincoln crushing the Southern Betrayal, and even if it wasn’t, good luck against those drones with that pawnshop pistol, right?).

    I suspect that what the PPI might be doing is trying to make a bad argument sound good via lies, damn lies and statistics. I mean, supposing one came up with a chart showing gun ownership spiking while gun homicides dropped to zero? Would that really be dispositive over whether a background check and waiting period to deter unmedicated paranoid schizophrenics from buying unregistered firearms at flea markets was/wasn’t a good idea? It seems at best like you’d need different numbers (I dunno, maybe the specific number of UPSes purchasing unregistered firearms for instance, or some kind of data to show whether or not background checks had any effect on gun violence), at worst it seems like the numbers might be less-relevant altogether as keeping someone from buying a gun today is not the same thing as keeping someone from buying a gun.

    For that matter (and I’m trying to wrap this up, honest!) it occurs to me that a focus on gun homicides also might be insufficient in and of itself without a look at other factors/variables/stats/whatever. I mean: suppose for instance I showed you one chart that showed that gun sales increased while gun homicides went down but then followed that up with a chart showing that gun sales increased while gun-related violence increased, or that gun sales increased while gun injuries increased. Declines in homicides might be the result of improved ER care in such scenarios, or there might be some other variable involved, right? One might reasonably conclude that an increase in (non-fatal) armed robberies in which a firearm was used was a bigger gun-access problem than a plateaued number of gun deaths. Of course, that’s all thought experiment stuff as I am too lazy right now to go looking for such numbers to challenge or confirm my beliefs, and would much rather take that time to make a cocktail. But I figured I’d put it out there, anyway.

    • peterwagner says:

      To be clear, I’m the author of the original article. There is a correlation, and that’s important.

      But to draw more complicated conclusions, you need more than a hundred words and one graph. In fact, I’d recommend The Crime Drop in America which inspired the graph. The point of the long articles in that book is a good one: Crime is down, but why is it down, and can we deduce anything useful before it starts to go up again?

      A lot of what the scientists in The Crime Drop do is disaggregate the trends so we can start to draw more meaningful conclusions. Their gun chapter is quite eye-opening, especially about the role of cheap handguns; but the chapter that looks at the effect of incarceration on crime rates might be up your alley in terms of the questions asked.

      Enjoy the cocktail!

    • Gavin says:

      “Of course, that’s all thought experiment stuff as I am too lazy right now to go looking for such numbers to challenge or confirm my beliefs, and would much rather take that time to make a cocktail.”

      The very problem in America today… One spent a lot of time writing and correcting a very long rant before posting it, yet had no time to look for actual data concerning the subject?!! (note the 3 minutes, lol 😉

  3. Pingback: cowboys, cops, and race | The Handsome Camel

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s