The Right to Carry Has Not Increased Crime: Improving an Old Debate Through Better Data on Permit Growth Over Time

Abstract

Over the last 30 years, a majority of US states adopted Right-to-Carry (RTC) laws at the same time that crime rates dramatically decreased. A large literature has examined whether RTC laws contributed to or slowed this decline in crime, with most studies concluding that they have no significant effect on crime. However, this research has been plagued by methodological challenges, many of which are exacerbated by the common approach of modeling the effect of RTC laws using a binary dummy variable to indicate a one-time change in policy. Recently, Donohue, Aneja and Weber (2019a) have employed a novel synthetic control approach which they suggest indicates that RTC laws significantly increase violent crime. However, we show that this analysis is highly sensitive to modeling choices, and Donohue et al. chose a specification that has been criticized by Kaul et al. (2017) as mistaken because it prevents covariates from exercising any influence on the development of predicted crime rates. Correcting this to properly incorporate covariates dramatically changes the estimated effect in many states; and comprehensive synthetic control analysis reveals no significant effect on crime. Given the methodological challenges inherent in binary approaches to modeling the effects of RTC laws, we gather data on the growth of carry permits in states over time, which allows us to investigate the phenomenon of interest - the actual ability to carry - in a manner that is theoretically more valid and econometrically more powerful. Employing two different methods for estimating missing data - modeling the growth of permits as a logistical growth process and imputing missing data using the Amelia II package - we find that the growth in carry permits has no effect on violent crime rates, homicide rates, firearm homicide rates, or non-firearm homicide rates. This study provides further, strong evidence that the dramatic growth in the ability to carry firearms for self-defense in recent decades has not exacerbated crime rates.