Guns are bad, ok? If you see a gun, shoot it quickly!
Perhaps, perhaps. Back in 1997 an identity-challenged and yet statistically-smart chap in America worked out that more guns had lead to less crime. Specifically, he looked at states that had introduced “shall-issue” laws for concealed carry (CC) handguns and found that the laws had lead to less crime.
This started a big academic debate, which is well summerised at the beginning of the latest contribution by Moody & Marvell (The Debate on Shall-issue Laws, Econ Journal Watch, Vol 5, No 3, Sept 2008, pp269-293). This is a quick summary of the story so far…
A year after the 1997 article by Lott & Mustard (L&M) showed “more guns, less crime”, two other academics by the name of Black & Nagin (B&N) came out with a different conclusion. By doing state-by-state analysis (instead of aggregate analysis) they found that the outcomes were mixed, with no significant effect in either direction. Lott responded quickly, showing that the B&N averages missed the trends. Specifically, L&M suggested that crime may follow an inverted-V with crime increasing before the CC laws and decreasing afterwards, so that an average of before and after might show no difference.
In 2000 Lott updated his study and in 2001 various other studies came out supporting the “more guns, less crime” hypothesis in the Journal of Law and Economics. Lott updated his study yet again in 2002, which again supported the original idea. At this stage, Lott was winning the debate.
Enter… Ayres & Donohue (A&D), who in 2003 did another study which adjusted for when the laws were introduced. And they criticised L&M for not controlling for the crime-spike caused by crack-cocaine. L&M had modelled the impact of crack-cocaine and found it to be insignificant… but given the A&D criticism, this became an issue that needed to be explicitly addressed in future studies.
A&D looked at the five years after the introduction of the laws in each State, and found different outcomes in different States. They concluded that there were more States were the “shall-issue” laws created a bad outcome, with an “overall small increase in crime”.
Moody & Marvell (M&M) have done two things to update the A&D study. First, they are using more recent data, which is always nice. But more importantly, they are extending their time-frame beyond 5 years. This is important. A&D found that the CC laws created a negative effect for the first few years and then a positive effect each year after that. So if you study the laws over 5 years it gave a negative result. But if you studied the laws over 6 years you get a positive result. If you put the impacts into dollars and then discount the future impacts at 5% per year, after 7 years you get a positive result. Generally, policy cost-benefit analysis is done over 20 or 30 years. M&M were conservative and chose only 10 years, and found a benefit from the CC laws.
M&M also look at the academic score-card. They mention significant papers on the issue, of which 14 say “more guns, less crime”, 8 say “dunno” and 3 say “more guns, more crime”. All of the “more crime” studies are in non-refereed journals. A meta-study done by the National Research Council (NRC) concluded that they could make no conclusion, though some of their own reserach supported the “less crime” scenario.
Personally, I feel most comfortable with NRC conclusion that we don’t have enough information to make strong conclusions in either direction. On balance, the evidence is for “more guns, less crime” but as will all new areas of study, it is reasonable to be modest about how much we really know.
As a person who believes that freedom is a good thing, I think that the burden of proof is on the person who wants to take away freedom. Given the lack of evidence that “guns are bad” then the only reasonable conclusion is to not introduce excessively burdonsome gun laws. Of course, this will run against the polite opinion of the gentrified chattering classes. I can only offer my apologies. I wish I could tell you what you want to hear, but unfortunately that’s just not where the facts lead.