Responding to GVPedia’s False Claims about Constitutional Carry Laws

Feb 16, 2022 | Constitutional Carry, Original Research

Professor Carl Moody and Dr. John Lott have put together a response that Devin Hughes’ GVPedia has put together attacking research that Professor Moody did on the impact of Constitutional Carry laws and other writing by Lott. A copy of Hughes’ latest claims is available here, though we have copied portions of those claims before our responses below. We have previously responded to many of Hughes’ other claims that have involved misquotes or statements out of context here.

Response to GVPedia by Professor Carl Moody, College of William & Mary

Claim: GVPedia has analyzed CDC data from states with Permitless Carry laws that have at least three years of data available. Our analysis found that States that pass a Permitless Carry law suffer from a 22% increase in gun homicide for the three years after the lawʼs passage, more than doubling the 10% increase for the country overall in the same time period.

Response: GVPedia (hereafter GVP) reports a simple analysis of firearm homicide rates before and after the passage of constitutional carry laws. My previous work on this issue is available here.

Firearm homicide is not a good measure of the effect of Constitutional Carry laws because murders can be committed with other weapons as well as by hands and fists. Also, gun homicide data include justifiable homicides, including homicides by police in the line of duty. Justifiable homicides are benefits, not costs, and they might understandably increase when more citizens are allowed to defend themselves with guns. The FBI murder rate has neither of these problems and is a better test of the effect of constitutional carry laws. Nevertheless, my analysis finds that Constitutional Carry laws do not increase firearm homicide. GVP did not challenge my findings showing a significant reduction of murder, no increase in violent crime, and no increase in deaths of police officers.

The GVP analysis does not indicate the use of any control variables, thus confounding the effect of the Constitutional Carry law with changes in police presence, incarceration rates, income levels, the unemployment rate, poverty levels, demographics such as the number of people in various age groups, and alcohol consumption. My analysis includes all those control variables and more. For example, my analysis includes the effects of other laws in force during the sample period, such as stand your ground laws, three-strikes laws, and several others. In addition, I include individual state trends to control for factors such as the increased use of surveillance cameras, cell phones and smartphones with cameras, improvements in trauma care that can save lives and turn potential murders into assaults, improvements in law enforcement such as DNA analysis, crime scene investigation, etc. Finally, I correct for bias associated with the failure to control for state effects that are constant during the years of the sample, such as climate, history, culture, attitudes toward crime, attitudes toward law enforcement, etc.

The GVP analysis does not report any statistical tests indicating whether the difference between the percent change in the Constitutional Carry states is significantly different from the percent change in the national average. The GVP analysis uses only three years before and after the passage of the CC laws. This is only a short-run analysis. In addition, the GVP analysis does not include the effect of the law in the year of passage. My analysis uses five years before and after the law to better assess the effect of the law. I have emailed GVP asking if they would share their data and programs with me but have not received any response.

I re-analyzed the data using the GVP technique of looking at only three years before and after the law and found a significant decline in firearm homicides after the passage of a Constitutional Carry law. (Stata programs and log files are available on request.)

Claimed Disinformation: “GVPedia and its colleagues were unable to independently replicate Dr. Moody’s analysis.” GVP p.3.

Fact: GVP did not attempt to contact either Dr. Lott or myself in order to determine what methodology I employed. How can GVP claim to replicate my analysis without knowing what I did? If GVP had contacted me, I would have sent them the Stata programs and log files as well as the data I used to do the analysis. With this information, they would have easily replicated my analysis.

The files are available here.

Claimed Disinformation: “When GVPedia uses the same methodology as Dr. Moody to analyze publicly available data….” (p.3). 

Fact. They don’t know what methodology I used. How can they claim to have used it? If they had used my methodology on the same data, they would have replicated my results. 

Claimed Disinformation: “There are sixty-five national level academic studies, the plurality of which show that weakening laws that prohibit Permitless Carry, often referred to as concealed carry laws, increase crime.” (p.2)

Fact: I don’t know what GVP means by “national level” academic studies. However, I do know that, since the seminal article was published in 1997 by Lott and Mustard, there have been 21 refereed articles published finding that concealed carry laws significantly reduce crime, 13 find that concealed carry laws have no significant effect on crime, and 9 find that concealed carry laws significantly increase crime. Therefore, using the best academic evidence, 34 peer-reviewed studies find that concealed carry laws either significantly reduce or do not increase crime while only 9 find that they significantly increase crime. 

Increased firearm carrying reduces firearm homicide by increasing the risk to potential attackers that the intended victims will effectively defend themselves. 

Responses put together by John Lott

Claimed Disinformation: Most studies show that weakening concealed carry laws reduces violent crime. 

Response: Hughes’ GVPedia claims we “misclassified at least two studies” in the list of academic studies that show the benefits of concealed carry, and we did so by including studies that didn’t study concealed carry. But he doesn’t even mention the two studies that he claims have been misspecified, so it is hard to respond. But people shouldn’t take our word for it —a link to all the studies we cited is available so that people could check them out. A more detailed discussion of these papers is available in the third edition of Lott’s book from the University of Chicago Press, and in the Amicus brief that the Crime Prevention Research Center submitted in the recent case heard by the U.S. Supreme Court on New York’s concealed handgun law.[1] As to the research I cited being “error-filled,” these were all peer-reviewed studies in academic journals. Our gun control advocate critic, Devin Hughes, has never published anything in an academic journal. He merely is excluding studies that he disagrees with. Dr. John Lott has published over 100 peer-reviewed academic studies and held academic positions at Wharton, University of Chicago, Stanford, and Yale, and was the senior advisor for research and statistics in the U.S. Department of Justice and the chief economist at the United States Sentencing Commission. Carl Moody is a professor of economics at the College of William and Mary.

Claimed Disinformation: Concealed carry permit-holders are “incredibly law-abiding,” meaning they couldnʼt be responsible for an increase in violent crime. 

Response: The discussion here cites two newspaper articles, but no one argues that permit holders never commit crime, just that they do it at extremely low rates. Yet, neither of the articles actually puts the numbers as a percent of those who have permits. Take the New York Times piece, which provides the most detail. For the sake of argument, let’s assume that all their numbers are correct. There are about 200 violent crimes committed over five years, an average of 40 per year. With about 550,000 concealed handgun permit holders in the state at that time, that implies a violent crime rate of 7.3 per 100,000 people.[2] By comparison, the average violent crime rate for 2006 to 2010 was 435.1 per 100,000 people, a rate that is 60 times higher.

Unfortunately, the New York Times never provides information on how many of these violent crimes involve guns, let alone a concealed handgun. Generally, only about 8% of violent crime involves guns, and the share of permit holders who committed a violent crime that used a gun to commit the crime is also similarly low. That is important because if permit holders having guns increased violent crime, giving them permits would only increase violent crimes involving guns.

But ignore this last point and assume that none of the violent crimes committed by permit holders would have occurred if they didn’t have permits (clearly a massive overestimate). These 40 violent crimes per year are a tiny percentage of the average of 40,026 violent crimes per year in North Carolina. Indeed, they are just 0.099% of 40,026.

So, can an increase in violent crime by permit holders cause a noticeable increase in a state’s violent crime? No. Indeed, even this obvious overestimate of the change would be so small it would not be possible to measure it.

Claimed Disinformation: Permitless Carry helps people protect themselves with a firearm, which already occurs 2.5 million times each year. Such defensive gun uses are 4-5 times more common than gun crimes. 

“Empirical evidence from the Gun Violence Archive reveals approximately 2,000 verified defensive gun uses annually, not 2.5 million.”

Response: Eighteen national surveys find an average of 2 million defensive gun uses per year.[3] By contrast, the National Crime Victimization Survey has regularly shown about 400,000 to 500,000 crimes with firearms each year. Thus, defensive gun uses are about 4 to 5 times more frequent than crimes with guns. As to the claim that there is no peer-reviewed research that makes this point, see for example Lott’s book from the University of Chicago Press.[4]

Ninety-five percent of the Gun Violence Archive relies on counting defensive gun uses reported in news articles, but that is a dramatic undercount because the vast majority of successful self-defense cases don’t make the news. For example, in a report we did earlier this year that looked at defensive gun uses from Jan. 1 to Aug. 10 of this year, we found 774 defensive gun uses, fully 85 percent involving people shot: 43% resulting in death and 42% percent in wounding.[5] Less than 4% of cases involved no shots fired. By contrast, survey data indicate that in 95% of cases, when people use guns defensively, they merely show the gun to make the criminal back off.[6] Such defensive gun uses rarely make the news, though a few do. Understandably, someone getting killed is more newsworthy than a woman brandishing a gun and the criminal running away without committing a crime. Still, from a policy perspective, we care about both cases.

Claimed Disinformation: Defensive gun use is the most effective way to prevent injury and benefits populations that are especially vulnerable to violent crime 

Response: Lott’s research shows that letting blacks, women, in high crime urban areas, and older people produce the greatest reductions in violent crime.[7]

There is a long list of other research that shows that guns are the safest course of action when you are confronted by a criminal: for example, Gary Kleck and Miriam A. Delone, “Victim Resistance and Offender Weapon Effects in Robbery,” Journal of Quantitative Criminology 9 (1993): 55–81; Lawrence Southwick, “Self- Defense with Guns,” Journal of Criminal Justice 28 (2000): 351–370; and Jongyeon Tark and Gary Kleck, “Resisting Crime,” Criminology 42 (2004): 861–909. 

Surveys of academics show that criminologists and economists are very skeptical of the benefits of gun control.[8]


From John Lott

Claim: Some point to the rejections of permit applications based on NICS checks as a measure of how important it is to have permits. This point has been raised in Indiana and other states.

Response: The problem is that the numbers on background checks are a mess because about 99% of the denials are false positives.

The problem with the NICS background check system is that it usually uses roughly phonetically similar names and similar birthdays in doing most checks. Here is an earlier piece that I had at the NY Times in 2018 and a later piece that I had last year in the WSJ.

In any case, the bottom line is what happens to crime rates, and the earlier discussion here by Carl Moody makes it clear that things improve when Constitutional Carry laws are passed.

[1] John R. Lott, Jr., More Guns, Less Crime, University of Chicago Press, 3rd edition, 2010. Crime Prevention Research Center, “Crime Prevention Research Center Amicus Brief On New York State Rifle & Pistol Association V. Bruen,” Social Science Research Network, August 2, 2021 (

[2] For permit estimates see Crime Prevention Research Center, Concealed Carry Permit Holders Across the United States,” July 9, 2014 ( For all the annual reports see

[3] John R. Lott, Jr., More Guns, Less Crime, University of Chicago Press, 3rd edition, 2010. William English, “2021 National Firearms Survey,” Georgetown McDonough School of Business Research Paper, July 16, 2021 (

[4] John R. Lott, Jr., More Guns, Less Crime, University of Chicago Press, 3rd edition, 2010.

[5] John R. Lott, Jr., “There Are Far More Defensive Gun Uses Than Murders. Here’s Why You Rarely Hear of Them,” Real Clear Investigations, September 22, 2021 (

[6] John R. Lott, Jr., The Bias Against Guns, Regnery Press: Washington, D.C., 2003.

[7] John R. Lott, Jr., More Guns, Less Crime, University of Chicago Press, 3rd edition, 2010, pp. 4, 21, 64-75 181-4.

[8] Arthur Z. Berg, John R. Lott, Jr., and Gary Mauser, “Expert Views on Gun Laws.” Regulation, Winter 2019-2020, pp. 40-47 (