19

The Economist says the following in an article on recent mass shootings in the USA which discusses the evidence that restrictions on gun ownership reduce deaths:

...a mass shooting leads to a 75% rise in measures easing gun control in states with Republican-controlled legislatures

Do Republican legislatures tend to relax gun laws after mass shootings?

matt_black
  • 56,186
  • 16
  • 175
  • 373
  • 2
    How long is it _"after"_ mass shooting? I suspect that in US it might basically mean all the time :-P – vartec Jun 20 '16 at 18:24

1 Answers1

32

Maybe - there is preliminary evidence that supports it.

The Economist article is referring to this very recent working paper:

They found that:

when looking at enacted laws, the impact of mass shootings depends on the party in power. A mass shooting increases the number of enacted laws that loosen gun restrictions by 75% in states with Republican-controlled legislatures. We find no significant effect of mass shootings on laws enacted when there is a Democrat-controlled legislature.

They looked up to one year after the mass shooting for any effect on the legislation changes within the same state as the mass shooting. They compared it to to states in which no mass shootings had occurred (after controlling for some potential confounding factors).

By assuming that the occurrences are "salient and plausibly random occurrences" and by adding controls for the confounding factors that are time varying (unemployment, divorce rates, legislature size and restrictions, political parties, etc.) they claim that their study demonstrates causality. This is reasonable only if their assumption is correct, and they can successfully identify all of the important confounding factors.

The findings are discussed in a more readable version in a Harvard Business School web article. [Please be aware of the affiliations here: this is basically a press release version.]

This has not yet passed peer-review nor garnered post-publication review/criticism. This may give a skeptic some pause about whether to accept the claim yet. It is preliminary evidence to support the claim, but further peer-review to identify flaws and biases is yet to be completed.

Note: We've seen in previous questions that the definitions of mass shooting vary and affect the conclusions. The definition of "mass shooting" they use is:

as an incident in which four or more people, other than the shooter(s), are unlawfully fatally shot in a single incident not related to gangs, drugs, organized crime, or domestic disputes.

Community
  • 1
Oddthinking
  • 140,378
  • 46
  • 548
  • 638
  • Any idea why they don't include incidents "related to gangs, drugs, organized crime, or domestic disputes" ? – Matthew Moisen Jun 18 '16 at 19:26
  • "increases" implies causation, not just correlation (Plus, Economist's "leads" in the quote in OP). Did the paper support the causation implied in the synopsis? – user5341 Jun 18 '16 at 20:29
  • 2
    @MatthewMoisen: I believe these exclusions are not uncommon. I speculate it is because people/the media have a more visceral reaction to random shootings than targeted killings, perhaps because they figure they aren't involved in such gangs etc, so it won't affect them. – Oddthinking Jun 19 '16 at 00:30
  • 1
    @user5341: Good point. Causation can't be safely inferred from this study. – Oddthinking Jun 19 '16 at 00:32
  • 8
    @MatthewMoisen Those exclusions tend to come from the groups that track mass shootings, as there is no central government source. It appears that the authors of this paper assembled their own list, but the same exclusions are used by others, such as [Mother Jones](http://www.motherjones.com/politics/2012/07/mass-shootings-map): "Crimes primarily related to gang activity, armed robbery, or domestic violence in homes are not included." and [Stanford](https://library.stanford.edu/projects/mass-shootings-america): "The shooting must not be identifiably gang, drug, or organized crime related." – Zach Lipton Jun 19 '16 at 01:27
  • 1
    @MatthewMoisen Excluding gang/drug activity is reasonable: When a gangster/drug dealer engages in a shootout with mass casualties, It's usually against other gangsters/drug dealers and not randomly-targeted. I suspect that the a-posteriori risk of being targeted in such a shooting, given that you're not part of a gang/don't deal drugs, is quite low. I don't know the rationale for excluding domestic disputes however. – Iwillnotexist Idonotexist Jun 19 '16 at 02:53
  • 1
    @user5341: actually, I have to retract that. A more careful reading suggests they claim causality. I'll edit the answer soon to explain their justification. – Oddthinking Jun 19 '16 at 06:56
  • 1
    @ZachLipton While they don't specifically use the term "mass shooting," the FBI does have a definition for "active shooter incidents," which is defined very similarly to the Mother Jones definition, except that the FBI includes them regardless of how many people were killed (even if no one was killed.) [Their definition](https://www.fbi.gov/about-us/office-of-partner-engagement/active-shooter-incidents/a-study-of-active-shooter-incidents-in-the-u.s.-2000-2013) is "one or more individuals engaged in killing or attempting to kill people in a populated area." – reirab Jun 19 '16 at 19:00
  • 1
    @ZachLipton In particular, the FBI **does** exclude "shootings that resulted from gang or drug violence." The reasoning behind the exclusions is that the typical causes and results of those situations are vastly different and, therefore, it's much more useful to study them separately when attempting to identify patterns and potential solutions. – reirab Jun 19 '16 at 19:04
  • @Oddthinking Can you please clarify whether the study is only concerned with mass shootings that happened within the state whose laws are being examined or if they're referring to any state passing a law in the wake of a mass shooting in the country? If it's the latter, it doesn't seem like they'd have much of a control data set, since "within one year of a mass shooting" is basically all the time when the entire U.S. is considered. – reirab Jun 19 '16 at 19:09