A importance of a preliminary paper published by Doucette et al. (2022) studying the potential effects of permitless carry laws on officer involved shootings should not be understated: the rate at which states have adopted permitless carry laws is nothing short of astounding. To date, half of U.S. states now have permitless concealed carry laws which allow law-abiding adults to carry firearms concealed in public with no licensure, training, or other administrative requirements. Effectively, unless an adult has committed a felony, a citizen may carry a firearm without notifying any government entity in half of the United States.

Given how rapidly these laws have been passed, public health scholars are concerned with the effect these laws may have on different criminological outcomes. Indeed, past public health researchers have typically concluded that liberalized concealed carry laws increase rates of violent crime. Doucette et al. (2019) concluded that permitless carry laws are associated with 29% higher rates of workplace homicides. Siegel et al. (2017) likewise concluded that liberalized concealed carry laws increase homicide. Thus, Doucette et al. (2022) ask the important question: will permitless carry increase officer involved shootings, too?

Doucette et al. argue that officer involved shootings increase, on average, by 29.5% after the adoption of permitless carry laws using an augmented synthetic control model. This works out to approximately four additional officer involved shootings per year compared to what would have happened without the adoption of permitless carry. Doucette et al. argue that because most homicides of police officers involve firearms, the introduction of permitless carry may increase a police officer’s perceived risk when encountering civilians on the job.

Overall, I do not believe the authors’ results, as presented, offer a strong case that the association between permitless carry laws and officer involved shootings are causal. Further, I believe their theoretical explanation for this finding is lacking.

Doucette et al. present their key results in tabular form in Figure 5 of their report (p. 380). In the report, they list all ten states which were able to be tested in their sample: Idaho, Kansas, Kentucky, Maine, Mississippi, Missouri, New Hampshire, North Dakota, South Dakota, and West Virginia. In four states, the relationship between permitless carry and officer involved shootings was indeed positive. However, in six states—over half of the states they studied—the result was not significant.

Not only were over half of the states tested not significant, but in five of the states—half—the direction of the relationship was actually negative, albeit not significantly so. If over half of the states in the sample did not experience an increase in officer involved shootings, and in half of them the direction of the relationship was negative, it is extremely likely that the results are not causal. Indeed, Doucette et al. do not attempt to explain this puzzling result at all. Why is permitless carry associated with more shootings in Idaho, Mississippi, Missouri, and West Virginia, but not in Kansas, Kentucky, Maine, New Hampshire, North Dakota, or South Dakota? While none of the negative associations are significant, why are there fewer shootings in Kentucky, Maine, New Hampshire, South Dakota, and North Dakota? If there were a handful of outliers, it would not be a concern—but when over half of the sample does not have a significant association with the dependent variable, the problem ceases to be an issue of outliers. The results of the paper simply do not convincingly suggest the relationship between permitless carry and officer involved shootings is causal. The authors, in this case, should not rely on the average treatment effect to draw their conclusions.

The explanation preferred by Doucette et al. as to why this association may be causal is also lacking. The authors argue, “less restrictive CCW laws can increase an officer’s perceived risk of personal injury in an encounter which may in turn increase the likelihood that deadly force is used” (p. 381). However, there is no strong evidence for this claim. In fact, a comprehensive public survey of law enforcement attitudes towards guns, gun control, and gun violence has been published by the organization Police1. The vast majority, 91.3%, of officers polled stated that they believe civilians who have not been convicted of a felony should be allowed to carry weapons concealed. Over half, 54.7%, believe armed civilians are very important in reducing crime rates overall (PoliceOne, 2013). Of course, simply because police officers support carrying in public does not mean they support permitless carry laws—indeed, on other questions, large numbers of police supported firearms training. Still, these are not the attitudes which would be expressed by a population overly worried about encountering an armed ci cite should be tvilian.

To defend their explanation for why permitless carry may increase officer involved shootings, they point to other research which suggests higher levels of gun ownership increase officer involved shootings (Hemenway et al., 2019). If true, this finding may provide some evidence for Doucette et al.’s theoretical explanation. The cited work, however, was a cross-sectional study. Cross-sectional datasets are almost always plagued by issues of endogeneity. As Arellano (2003)explained in his work on panel data econometrics: “researchers have often been confronted with massive cross-sectional data sets from which precise correlations can be determined but . . . had no information of policy interest” (Arellano, 2003 p. 8). Cross-sectional datasets are simply not adequate for determining causality. Overall, the explanation for their results is simply not convincing, and the cited research to bolster their theoretical explanation does not offer strong evidence for their claims.

In fact, it is possible to tell a story where constitutional carry laws reduce officer involved shootings. If every adult, other than convicted felons, are allowed to carry firearms, the burden of proof would be on an officer to determine whether a civilian is carrying a firearm illegally. Because of this, it may reduce the number of instances a police officer even encounters an individual, as a citizen being armed is no longer cause for suspicion.

There is also evidence that permitless carry laws may reduce the risk to police officers. Some unrefereed work suggests per capita rates of police killings have fallen after the adoption of permitless carry laws, and rates of assault and robbery may have fallen after the adoption of constitutional carry laws as well (Moody, 2022). Other research suggests permitless carry laws have no impact on homicide (Hamill et al., 2019). Both of these findings suggest the risk to officers may actually be lower as a result of permitless carry laws—or, in the case of Hamill et al. 2018, have no impact on the risk at all. Either way, there is simply no strong evidence that these laws actually increase risk to police officers, and some evidence suggests it may reduce it.

Overall, Doucette et al. (2022) have done the important work in producing an intriguing preliminary study as to the results of permitless carry laws and officer involved shootings. Their statistical methods were well-chosen, and the results of their paper can help guide future scholars who wish to study the impact of permitless carry laws on officer involved shootings. However, the results presented in their report do not suggest the association between permitless carry laws and officer involved shootings is causal. The effect of permitless carry laws on officer involved shootings is unclear, and to the extent an effect exists, it is not consistent across states. Further, the theoretical explanation they presented was not convincing; there is some limited evidence to suggest permitless carry laws may reduce actual risk officers face while on the job. At worst, it likely has no effect.

WHAT TO READ NEXT

WHAT TO READ NEXT