One of the longest-running arguments in the gun world is the issue of risks versus benefits of guns. The argument erupted in the late 1980s when the gun industry shifted its marketing from hunting and sport shooting to using a gun for self-defense. Chickens then came home to roost first with a survey published by Gary Kleck in 1995 and then a book published by John Lott in 1998. The Kleck survey claimed that as many as 2 million violent crimes were prevented each year because the alleged victims were able to defend themselves with a gun. Lott took this idea one step further, claiming that in jurisdictions which began to issue concealed-carry licenses, homicides went down. An entire academic cottage industry now exists which argues both for and against, an argument amplified by a lengthy and very detailed paper published by the National Bureau of Economic Research (NBER.) You can download the paper from my website here.
The lead scholar, John Donohue, has been chasing John Lott for at least 15 years, and has published (according to his Stanford University bibliography) at least 15 articles, op-eds and other comments about what he considers to be the methodological flaws and mistaken conclusions in Lott’s work. The NBER paper is the latest salvo coming from Donahue’s arsenal, and while my friend John Lott seems to be busy hopping from one alt-right radio show to the next, I suspect he’ll sooner or later post a response to what Donahue and his colleagues have said.
What they have said is that they extended Lott’s time-period from 2000 to 2014, and compared changes in violent crime rates between states which passed ‘right-to carry’ (RTC) laws and state that did not. In the 9 states which never adopted RTC laws, violent crime declined by more than 40%, but in the 37 states which adopted RTC laws during the same period, violent crime declined by less than 10%. In other words, people living in ‘shall-issue’ concealed-carry states are far less safe from violent crime than people that live in states where CCW is given out only with cause.
The NBER paper contains additional data which clearly undercuts Gun-nut Nation’s belief that, in and of themselves, right-to-carry laws reduce crime. But I have a much bigger issue with Lott’s ‘more guns = less crime’ approach that has nothing to do with statistics or data at all. Ultimately, Lott’s approach rests upon an assumption about the behavior of a certain class of human beings – criminals – that has absolutely no basis in truth or fact. And the assumption is that criminals who are thinking of committing a face-to-face crime (murder, assault) will decide instead to commit an anonymous crime if they believe that the victim whom they are thinking of attacking might be carrying a gun. He states this specifically on Page 6 of the 3rd edition of his book: “the criminals in states with high civilian gun ownership were the most worried about encountering armed victims.”
The idea that the unplanned, spontaneous and impulsive behavior resulting in homicide will be influenced or changed by some sort of rational, objective and planned decision simply flies in the face of reality and what we know to be the circumstances in which homicide and other violent crimes occur. Lott’s hardly an expert on homicide but Lester Adelson certainly is: “With its peculiar lethality, a gun converts a spat into a slaying and a quarrel into a killing.”
The response to Lott by scholars like Donahue may clarify both the validity of relevant data and how it is used, but no matter how sophisticated the statistical method brought to bear, regression analysis can ‘associate’ different trends, the exact causal connection between, say, gun homicides and issuance or non-issuance of concealed-carry licenses remains vague at best.
Legitimate scholars like John Donahue are motivated by the hopes that their research will provide evidence which can be used to fashion workable public policies to deal with the injuries caused by guns. I have a policy suggestion that doesn’t need any scholarly validation at all: get rid of the damn things. That’s all you have to do.