DOJ Report: The Cannabis Effect on Crime: Time-Series Analysis of Crime in Colorado and Washington State



In many ways the legalization of cannabis by ten states and the District of Columbia, as of March 2019, constitutes a grand ongoing experiment into how a major public policy initiative does or does not accomplish its expected outcomes. One of the principal expectations of the proponents of Initiative 502, the voter-initiated bill authorizing the recreational sale of marijuana in Washington, was that crime would decrease. Crimes generally were expected to decline in number, but particularly those crimes associated with the use of marijuana (e.g., possession, black market production, sales and distribution of cannabis, burglaries or thefts believed to be committed to secure funds to purchase marijuana). Some preliminary studies released shortly after legalization have intimated that crime rates have been going up rather dramatically in some of the states that have legalized recreational marijuana (Smart Approaches to Marijuana, 2018). In Washington State, early reports suggested that the number of marijuana-related offenses such as assault, theft, harassment, and vehicular offenses increased in Washington after the legalization (Northwest High Intensity Drug Trafficking Area [NHIDTA], 2016), but that “violent crime is down since Washington legalized marijuana” (Santos, 2017Santos, M. (2017). What actually happened to violent crime after Washington legalized marijuana. The News Tribune. Retrieved from [Google Scholar]). Or, paradoxically, the article by Malcolm Gladwell in The New Yorker, claiming (based on a book by Berenson, [2019Berenson, A. (2019). Tell your children: The truth about marijuana, mental illness, and violence. New York, NY: Simon & Schuster. [Google Scholar]]) that violent crime had increased in Washington state post legalization.

As Garland (2001Garland, D. (2001). The culture of control. Chicago, IL: The University of Chicago Press.[Crossref] , [Google Scholar]) has noted, there is a strong political demand for immediate answers to often quite complicated questions of public policy. In short, many politicians are inclined to make use of the earliest available data, and unfortunately too often what is available for public consumption at the outset of change in policy represents research employing limited pre/post analyses or misrepresentation of facts. Too often the results reported from such analyses fail to take into consideration the context of practice. For example, consider headlines associated with increasing citations for public marijuana consumption, in and around major cities. In many ways, these headlines are interpreted to suggest that marijuana users are increasingly consuming in public, a practice which was explicitly banned in Washington law. However, to some extent, these increases may in fact relate to property ordinances and rental agreements banning smoking, where violation is an automatic qualification for termination of the lease. Such policy conundrums create an environment where it is illegal to smoke in public and essentially illegal for marijuana users to smoke in their residence. Additionally, pressure from retail establishments and other members of the public can create pressure on police officers to issue citations.

In the absence of more rigorous and robust types of analyses, policy discussions and decisions in those states considering the liberalization of their own cannabis laws are prone to believe the misleading conclusions disseminated about likely outcomes. A variety of claims regarding the deleterious effects of legalization have already been made in a number of instances such as in Berenson’s widely cited book (2019) about the purported dangers of marijuana and Vestal (2019)’s column for the Spokesman Review. Some politicians have also linked the legalization of marijuana with increases in violence, often without the support of empirical data (Adams, 2018Adams, M. (2018, March 28). California officials say marijuana legalization causing more violent crime. Forbes. Retrieved July from [Google Scholar]). Advocacy groups, both for and against marijuana legalization might also contribute to this problem. For example, the group Smart Approaches to Marijuana (2018), frequently presents anecdotal or single-site evidence about potential increases in crime, without a robust analysis to support assertions.

Recognizing the importance for public policy making of more robust research designs in this area, this study uses a quasi-experimental, multi-group interrupted time-series design to determine if, and how, crime rates in Colorado and Washington State were influenced by the legalization of recreational marijuana in 2012 and the start of retail sales in 2014. The objective of the current study is to evaluate whether cannabis legalization would lead to changes in crime rates. This multi-group interrupted time-series study is more rigorous than the limited pre/post analysis frequently used to resolve political discussions because its quasi-experimental design has greater ability to assess causality than correlational studies (Cook & Campbell, 1979Cook, T. D., & Campbell, D. T. (1979). Quasi-experimentation: Design & analysis issues for field settings (Vol. 351). Boston: Houghton Mifflin. [Google Scholar]). As such, this research is timely in that these were the two earliest states to legalize the growing, processing, and commercial sale of cannabis for recreational use. Notably, we observed no statistically significant long-term effects of recreational cannabis laws or the initiation of retail sales on violent or property crime rates in either Colorado or Washington.

As the nationwide debate about legalization, the federal classification of cannabis under the Controlled Substances Act, and the consequences for crime – from legalization – continues, it is essential to center that discussion on studies employing contextualized and robust research designs with as few limitations as possible.

Read full report


Latest posts