“Large capacity” magazines: Diagnosis or symptom?

(from luckygunner.com)
(from luckygunner.com)

The January 14, 2020 issue of the Journal of the American Medical Association included a Quick Uptake news article, “Large-Capacity Magaine Bans Linked With Fewer Mass Shootings, Deaths” (first online December 18). It references a study in the American Journal of Public Health by Louis Klarevas, Andrew Conner and, one of our frequent “guests” here, David Hemenway: “The Effect of Large-Capacity Magazine Bans on High-Fatality Mass Shootings, 1990–2017”.

Essentially, their argument is that “large-capacity magazines” (at least a better phrase than “high-capacity” for those containing more than 10 rounds) are associated with a higher incidence of “high-fatality mass shootings” (those with more than 6 fatalities other than the shooter’s). Of the 69 HFMSs counted during the study’s 28 years, 44 involved LCMs and 16 didn’t, while in 9 cases magazine capacity couldn’t be determined. 49 occurred in states allowing LCMs compared to 20 where they were banned. More shooters, 81%, used LCMs in states without bans versus 55% in ban states. Likewise, LCM bans were associated with fewer fatalities in each incident (averaging about 7 per shooting in ban states versus about 12 in non-ban states).

That’s a lot of suggestive numbers. But do they really mean that “LCM bans have saved lives”? Let’s look closer.

In 1990, only New Jersey had instituted an LCM ban. From 1994 to 2004, they were outlawed nationally during the federal “assault weapon ban”. From 2004 to 2017, a total of only 8 jurisdictions (7 states and D.C.) had established bans. Which states are not described. On a very gross level, that’s barely 1/6 of the nation. It very much matters which sixth these are because of how different each state’s mix of urban and rural, economic factors, gun and other cultural aspects. Their covariables of population density, education level, unemployment and incarceration rates, and approximate gun ownership incidence don’t nearly address the distinctions that exist between states.

This, again, is why only comparing the same geographic area over time is a useful control; comparing one state against another is always comparing apples and oranges. And as always, discovering a difference before and after a change (e.g., LCM bans) is not sufficient analysis—observing a change in trend is necessary. This is a lot more work—for example, by Mark Hamill’s group that soundly found no correlation between “State Level Firearm Concealed-Carry Legislation and Rates of Homicide and Other Violent Crime”.  

Similarly, while the overall increase in incidence of HFMSs is noted, how is this changing in different localities? We can’t tell without know what those places are.

An even more fundamental problem is the definition chosen of a HFMS as one in which 6 fatalities (other than the shooter) occur. There is no clear explanation as to why that number seemed best. They even note that mass murders are “typically defined as ≥ 4 homicides”, not six. In addition, “there has been an average of 2.5 incidents per year”, certainly not enough to draw conclusions about common elements. One has to wonder whether different definitions produce different findings.  

For more (and better) information about the relevance of LCMs in shooting, check out Gun Facts. This is a site you should have on top of your gun bookmarks for its great collection of hard data about every aspect of the role and use of firearms in modern life.

Finally, the same elephant is in the room every single time that public health “experts” claim they’ve proven some danger due to the irrationality of allowing American citizens to choose their best defense. There is no consideration of the value of “LCMs” for defense, just as they will never recognize the value of firearms in lawful defense of innocent life. We know that American civilians probably perform somewhere north of 1 million (per the CDC) to over 2 million (per Kleck) defensive gun uses each year. By any standard, farm more lives are saved that way than are actually lost to firearm misuse.

So where is the data on the frequency of gang attacks on individuals, or in multiple assailant home invasions? That’s what we need to compare the questionable danger of LCMs in criminal mass shootings to their value in self-defense against more than a single attacker.

How does this “research” ….. thee? Let us count the ways:

  • Researcher bias
  • Seeing a firearm accessary as the primary variable (as the prime issue)
  • Likely cherry-picked definitions, and therefore data
  • Setting up straw-man arguments (as if the key to reducing mass shootings is the magazine)
  • Rationalizing that correlation = causation (“the theory   . . . makes sense”)
  • Ignoring data that is far more important that the study variables
  • Insignificant significance in the conclusions drawn (here, on negligible bases)

If this list sounds familiar, it should. Check out our paper, Reading “Gun Violence” Research Critically. You can, too.



Robert B Young, MD

— DRGO Editor Robert B. Young, MD is a psychiatrist practicing in Pittsford, NY, an associate clinical professor at the University of Rochester School of Medicine, and a Distinguished Life Fellow of the American Psychiatric Association.

All DRGO articles by Robert B. Young, MD