New York Times publishes column from the world’s least reliable gun researcher

John Lott is back to his old tricks.

CREDIT: Screenshot
CREDIT: Screenshot

John Lott, a discredited researcher with a history of pushing false and misleading information about guns, scored space in Tuesday’s New York Times. The article, which argues that background checks are flawed and ineffective, continues his streak of misleading and deeply flawed analyses.

It’s unclear why, exactly, the New York Times believed Lott is a credible source of gun research. Lott’s trademark assertion that more guns means less crime was formally and thoroughly repudiated by a panel convened by the National Research Council. In one case, a researcher has suggested that Lott fabricated an entire survey on defensive gun use. Lott has also been caught publishing studies with severe statistical errors.

Faced with mounting criticism, Lott created a fake online persona, Mary Rosh, that would defend his work online.

In recent years, Lott falsely claimed one of his papers was published in a peer-reviewed journal, pushed blatantly false statistics about the relative rate of gun crimes in the U.S. and Europe and failed to report statistically significant results of his research that didn’t match his preferred ideological position.

Nevertheless, the New York Times gave Lott another chance. Its readers were not well-served.

This time, Lott attempts to derail the dwindling momentum in support of bipartisan legislation introduced by Senators Chris Murphy and John Cornyn in response to the massacre at the First Baptist Church in Sutherland Springs, Texas in November that would help ensure that all records of individuals who are prohibited from gun possession are provided to the National Instant Criminal Background Check System (NICS). Lott grudgingly concedes that doing so is a “good goal” but quickly pivots to what he views as a much bigger problem with the background check system: that it erroneously prevents too many people from buying guns. As evidence for this claim, Lott takes two tangentially related data points to draw the definitive conclusion that the system makes too many “false positives.” Lott cites data from the FBI showing that, from 2006 to 2010, roughly 377,000 people failed a background check. He then says that, because only 460 of these people were charged by federal prosecutors with illegally attempting to purchase a gun when they were prohibited from doing so and only 209 were actually convicted of this crime, the remaining 376,000 denials must have been in error.


Lott’s analysis on this point is deeply misleading. There are a myriad other reasons why federal prosecutors might choose not to prosecute cases where a person who was prohibited from gun possession broke the law by attempting to do so. As Robert Verbruggen of the National Review explained back in November, these cases are often declined by prosecutors because of the high burden of proving that an individual knowingly lied on ATF Form 4473, as opposed to having made an honest mistake. These cases are also time intensive and often viewed by prosecutors as being a low priority.

Indeed, the study cited by Lott for these low prosecution numbers makes this very point. In 2010, only 480 background check denial cases were not referred to federal prosecutors because the individual was subsequently found to have been denied in error because they were not, in fact, prohibited from buying a gun. Far more often, according to this study, cases were not referred for prosecution because they were deemed to have “no prosecutive merit” (1,661 cases) or “federal or state guidelines not met” (1,092 cases).

Lott also fails to account for prosecutions of these cases that occur at the state level, saying only that there are a “similarly small number” of them without any supporting evidence or data. In fact, states have become increasing engaged in following up on these cases and, when appropriate, prosecuting them as violations of state law. For example, since 2013, Pennsylvania state police have investigated every background check denial case in the state, leading to nearly 1,000 arrests for this crime. Virginia state police have a similar policy and arrested 1,265 denied purchasers in 2015.

A better way of quantifying the frequency of “false positives” would be to look at the number of successful appeals of background check denials. In 2016, roughly 30,000 people appealed their denial to the FBI— only 1,834 of these appeals were successful. To be sure, there is a backlog in the appeals system, meaning that there may be more successful appeals in the queue. But even so, this is a far cry from the tens of thousands of people Lott suggests are erroneously denied a gun purchase each year because of inaccuracies in the background check system. Additionally, a September 2016 audit of NICS conducted by the DOJ Office of the Inspector General found that the system has a 99.8% accuracy rate when it comes to denials.

Lott also reverts back to his usual safe space of arguing that the best way to reduce gun violence is to increase the number of civilians who carry loaded concealed guns around with them—a ready army of “good guys with guns.” This myth has been debunked repeatedly, including in a new study by researchers at Boston University that found that states with weak concealed carry licensing standards had markedly higher gun homicide rates that states with stronger standards.


Chelsea Parsons is the vice president of Gun Violence Prevention at American Progress. ThinkProgress is an editorially independent publication of the Center for American Progress Action Fund.