New research by Tse-Chun Lin and Runtong Lin identifies recreational marijuana use as an emerging risk factor of domestic violence. Their study provides the first plausible causal evidence that recreational marijuana use causes an 13.2% increase in domestic violence rate in Colorado. Such an effect is particularly prominent in Denver city where more than half of marijuana retail stores in Colorado are established.

Domestic violence is a major public health problem. Over 30% of men and women in the United States have experienced physical violence by an intimate partner in their lifetime, according to a report by the Centers for Disease Control and Prevention (CDC) in 2018. 

With around 26 million users, marijuana is the most used illicit drug in the United States. Its use is more common than the use of all other illicit drugs combined, as reported by the Substance Abuse and Mental Health Services Administration in 2017. 

The existing psychopharmacological (the study of substances that influence mental states) evidence suggests an overall small yet positive association between marijuana use and interpersonal violence. But a positive association between marijuana use and interpersonal violence does not necessarily mean that the former causes the latter: For example, both of them could be driven by common factors such as economic distress. Understanding whether marijuana use indeed causes interpersonal violence is crucial for relevant policy discussions.

This study uses the recreational marijuana legalisation in Colorado to provide causal evidence beyond an association. The gist is that the implementation of this legalisation in Colorado plausibly increases marijuana use due to lower prices, more consistent quality and less stigma from committing a crime. For this reason, the legalisation event can be viewed as a positive shock to marijuana use, from which it is possible to observe the before-and-after change of domestic violence rate, presumably caused by higher marijuana use. 

To form an (even) more rigorous experiment, the researchers then compare the before-and-after change of domestic violence rate in Colorado to that in Kansas, which neighbors Colorado and does not allow recreational marijuana. 

Such comparisons of domestic violence rates 1) before and after recreational marijuana legalisation and 2) between Colorado and Kansas resembles a laboratory experiment, where some patients receive treatment and others do not. This research design (called ‘difference-in-differences’) helps to answer the core research question: Does marijuana use cause domestic violence? 

Using data that contains 135,164 domestic violence incidents documented by the National Incident Based Reporting System (NIBRS) from 2011 to 2016 in Colorado and Kansas, the study finds a 13.2% increase in the domestic violence rate in Colorado relative to Kansas after the former legalises recreational marijuana. The result provides strong evidence that recreational marijuana use increases domestic violence rate. 

The researchers also find that implementation does not increase the domestic violence rate when limiting the sample to young offenders under the age of 21. This (lack of) result makes sense since anyone under the age of 21 cannot legally access marijuana and thus should not be subject to its impact on domestic violence.

Moreover, since over 70% of marijuana retail stores in Colorado are established in Denver, the study finds a stronger impact of a 24.1% increase in domestic violence after Denver enacts recreational marijuana legalisation, compared with other cities in Kansas. 

Finally, after excluding incidents in which the offender is suspected of using alcohol, the result attenuates slightly in magnitude but remains statistically significant. This result mitigates the concern that the main finding is predominantly driven by the influence of alcohol abuse on violent behavior, rather than that of recreational marijuana use. 

Overall, these findings highlight domestic violence as a potential outcome of recreational marijuana use. In light of the advent of recreational marijuana legalisation in the United States and worldwide, the study contributes to the discussion of the negative externalities of an emerging policy practice.


Contact details

Runtong Lin is with Faculty of Business and Economics at The University of Hong Kong. E-mail: Please send correspondence to this author. 

Tse-chun Lin is with Faculty of Business and Economics at The University of Hong Kong. E-mail: