The Arizona Shootings, Gun Violence Research and the Facts vs. The New York Times

You know you’re doing something right when two New York Times columnists, Gail Collins and Nicholas Kristof, attack your research on the same day. Kristof followed up on criticism leveled earlier in the week by others writing for The Times. I must have said something substantial enough to warrant this attention. Let us check the facts. Gail Collins worries that law-abiding citizens carrying concealed handguns can't be trusted:

"One can only hope that Saturday's horrible attack in Tucson encourages more citizens to carry concealed handguns," wrote John Lott, Jr. . . . on Wednesday. As a model, he pointed to Joseph Zamudio, . . . Lott's theory was that Zamudio was able to lend a hand "because his legally carried 9 mm semiautomatice offered him protection." He neglected to mention that while Zamudio never fired at the gunman, he almost drew on an innocent man by mistake.

No, it was not my mere speculation. Zamudio himself told Fox News' “Fox & Friends” that he thought that carrying a gun made him willing to run towards the shots while almost everyone else was running for cover. As to the possibility of shooting an innocent man, that possibility didn't occur because Zamudio used caution and observed from the gun's slide that it was out of bullets. While gun control advocates like to view his composure as extraordinary, it actually happens again and again.

Of all the multiple victim shootings around the country in public schools, the Appalachian Law School, on city streets, churches, or in malls that have been stopped law-abiding citizens with concealed handguns, none, not a single one has resulted in innocent bystanders being shot. Indeed, rarely do the citizens with the concealed handguns actually pull the trigger, simply brandishing the gun stops the attack. Permit holders do not endanger others.

Take Arizona, since that is where all the focus is. As of December 1, 2007, there were 99,370 active permits. During 2007, 33 permits were revoked for any reason — a 0.03% rate — cases that did not involve using the gun to harm others. And this is true in state after state. Between October 1, 1987 and December 31, 2010, Florida issued permits to 1.9 million people. 168 permit holders had their permits revoked for any firearms related violation, a rate of 0.009%.

Some writers -- Mr. Kristof as well as earlier pieces by John Donohue and James Alan Fox from Tuesday's New York Times website -- argue that the academic literature disagrees with the my claim that concealed handguns deter crime.

Kristof wrote: "One scholar, John Lott Jr., published a book suggesting that more guns lead to less crime, but many studies have now debunked that finding (although it’s also true that a boom in concealed weapons didn’t lead to the bloodbath that liberals had forecast)." And John Donohue claimed: "while some early studies by John Lott and others suggested that state policies providing greater freedom to carry guns would reduce crime, empirical evidence refutes this view."

When my original findings were published in the mid-nineties, there was a widely held belief that right-to-carry laws could result in blood baths and see increased deaths. Ample empirical evidence has shown this not to be the case. The debate is rather over whether there is no effect, a small benefit, or a large benefit.

Despite the above assertions, most of the academic research, including recent research on right-to-carry research by economists and criminologists do indeed show that the laws reduce violent crime (see the table here from the third edition of "More Guns, Less Crime"). Among peer reviewed studies in academic journals, 16 studies examining national data find that right-to-carry laws reduced violent crime, 10 claimed that they found no discernible effect, and zero studies found a bad effect from the law. Five other non-refereed studies were more divided, with three finding drops in crime, one claiming to find no effect, and two saying that there were either no effect or possibly small increases in crime. But even "no discernible effect" is usually not the same as "debunking" or "refuting" a hypothesis. Rather it often means that the evidence is not sufficient to draw a conclusion.

Both Kristof and Donohue link to a paper by Ayres and Donohue that supposedly "debunks" my research. That is hardly true. Even if we were to fully accept the wording of the research paper, which is questionable, they claim to have found a small temporary initial rise in crime, followed by a long sustained drop. Yet, even the initial increase is a result of a mistake in how they set up their regressions and even their own more precise estimates of the yearly changes in crime never show that increase.

James Fox points to research on multiple victim public shootings: "criminologists Grant Duwe, Tomislav Kovandzic, and Carlisle Moody [that] found the effect of various right to carry laws on the incidence of public mass shootings to be negligible." Alas, unlike my own work with William Landes, the research restrict their sample to attacks where four or more people were killed. These cases are sufficiently rare that one cannot expect to find anything from such a small sample. Landes and I found significant results when we examined more common cases where three or more people killed or four or more people killed or injured.

It is flattering to have the New York Times spend so much space discussing my work this week (and even CNN got into the discussion), but it would be even better if The Times could accurately report on the debate.

John R. Lott Jr. is a FOXNews.com contributor. He is an economist and the author of "More Guns, Less Crime (University of Chicago Press, 2010), the third edition of which was published in May.".