Wider gun ownership does not cause more gun-related crime, according to this new paper from the CEPR.
In a study of US counties, the authors found that there was a positive correlation between the prevalence of gun ownership and gun homicide rates. However, this is due to reverse causality, they claim; where crime rates are high, law-abiding people buy guns for self-protection. Controlling for this, they found, there's no link:
Our findings provide no support for the more guns, more homicide thesis. The appearance of such an effect in past research appears to be the product of methodological flaws, especially the failure to properly account for the possible effect of crime rates on gun ownership levels...Guns among criminals may increase homicide while guns among non-criminals decrease it, with the two opposite-sign effects canceling each other out. The most straightforward policy implication of such a combination of effects would be that gun control measures should focus on reducing gun prevalence among criminals while avoiding reducing it among non-criminals.
hum a few comments here:
1. What the eff is the CEPR doing having opinions about US counties? Are there really no more problems to be solved in the UK, where the taxpayers pay the bills?
2. The "endogeneity" issue here of people buying guns because they fear crime, and then commit crimes with them, might also be described as an "arms race". It is by no means obvious to me that it is the sort of thing that ought to be corrected for, assuming that one is looking for the answer to the question "do guns cause crime" rather than the correct estimate of the coefficient on guns in a crime equation.
3. It is usually considered better science to base the "policy implications" on the results, not on some hypothetical interpretation of the results. If Kleck et al want to support the conclusion that law enforcement ought to concentrate on criminal-held guns, then this research doesn't disaggregate and so it can't support that.
4. Of course this is also another bullet fired in Kleck's years-old feud with Lott, isn't it?
Posted by: dsquared | January 24, 2006 at 06:07 PM
Those area of the U.S. with high gun crime rates share two attributes: 1) Strict gun control laws. 2) They are cities run for years by Democrats.
Strike those areas from the U.S. numbers and gun crimes are comparable to Germany.
Posted by: Max | January 24, 2006 at 08:57 PM
Strike the balls from my Uncle Leslie and he'd be my Aunt Leslie.
Posted by: dsquared | January 25, 2006 at 09:03 AM
(also, are you removing German cities from the German statistics?)
Posted by: dsquared | January 25, 2006 at 09:04 AM
What do they use to account for the difference in homicide rates then? The U.S is approx 6-7 per 100,000, UK approx 1.5 per 100,000. the UK's is rising and the US is falling but surely there must be something to fill the gap between the numbers.
Posted by: CB | January 25, 2006 at 03:25 PM
What do they use to account for the difference in homicide rates then? The U.S is approx 6-7 per 100,000, UK approx 1.5 per 100,000. the UK's is rising and the US is falling but surely there must be something to fill the gap between the numbers.
Posted by: CB | January 25, 2006 at 03:26 PM
CB: this is exclusively a study on US counties and you really can't extrapolate from such a study to the UK which is a quite different society in terms of urbanisation, inequality, demographics and legal system. In general, I am of the unfashionable opinion that it is not possible to draw any conclusions at all about criminology from international comparisons.
Posted by: dsquared | January 26, 2006 at 07:19 AM