The effect of price and retail availability on the use of illegal and legal non-medical cannabis in Canada and the United States

Abstract

Background – Cannabis policy has liberalized in North America over recent decades. In October 2018, Canada legalized non-medical cannabis. The regulatory framework includes some similarities with jurisdiction in the United States (US) that also legalized non-medical cannabis, while setting international precedents in other policy areas. An objective of legalization in Canada is to reduce the illegal market by transitioning consumers to the regulated market. Cannabis prices and retail availability have a strong influence on consumption patterns, including purchase source; however, there is little evidence examining the association between price, retail availability and purchasing in the legal and illegal markets following legalization in Canada and the US.

Objectives - The current thesis examined four primary research questions: 1) To what extent are purchase source and quantity purchased associated with the price of dried flower in Canada pre-legalization? 2) To what extent is retail availability associated with purchasing dried flower from legal sources in Canada post-legalization? 3) To what extent is price associated with legal purchases of dried flower in Canada post-legalization? and 4) To what extent is price and the time since legal retail stores opened associated with legal purchases in US states who have a non-medical retail market?

Methods - Data come from Waves 1-3 of the International Cannabis Policy Study, a repeat cross-sectional survey conducted in Canada and the US. Wave 1 was conducted in August-September 2018, prior to legalization in Canada, with Wave 2 and Wave 3 conducted post-legalization in September-October 2019 and 2020, respectively. Respondents were recruited through online commercial panels, were aged 16-65, and had purchased dried flower in the past 12 months (Canada: n2018=1,227, n2019=2,506, n2020=2,417; US: n2019=5,643, n2020=4,267). In Study 1, linear regression models examined the association between price, quantity purchased and purchase source in Canada pre-legalization. In Study 2, binary logistic regression models examined the association between retail availability and likelihood of purchasing dried flower legally in Canada post-legalization. In Study 3, binary logistic regression models examined the association between price and legality of source in Canada post-legalization. In Study 4, binary logistic regression models examined legal purchasing of dried flower in US states that had a legal non-medical retail market. All analyses used weighted data.

Results Study 1 - Overall, the mean self-reported unit price (price-per-gram) of dried flower among Canadian cannabis consumers was $9.56 CAD in 2018. The unit price of dried flower significantly decreased as quantity purchased increased. For example, the mean price of dried flower purchased in smaller quantities (<3.5g) ($12.81/gram) was more than double the price purchased in larger quantities (>28g) ($5.60/gram). The estimated quantity discount elasticity was -0.21 (95% CI: -0.25, -0.18). The most common purchase sources used were family member/friends (53.0%) and illegal dealers (51.7%). Price-per-gram varied across sources; however, variation was largely accounted for by consumers purchasing different quantities from different sources.

Study 2 - Overall, 47.7% of past-year Canadian consumers reported last purchasing dried flower legally in 2019, with variation across provinces (range=40.5%-81.2%). Likelihood of purchasing dried flower legally was greater among those who lived closer to a legal retail store based on Euclidean distance (<3 km vs. 10+ km: AOR=1.56, 95% CI: 1.20,2.02), and who had shorter self-reported travel time to a retail store (<5 mins vs. >15 mins: AOR=2.24, 95% CI: 1.56,3.21).

Study 3 - The proportion of consumers last purchasing dried flower from legal sources increased from 2019 to 2020 (45.7% vs. 58.1%), and in the past 12-months, the average amount of dried flower consumers reported purchasing from legal sources increased from 55.7% in 2019 to 67.5% in 2020. The mean unit price (price-per-gram) of dried flower was $10.64 in 2019 and $10.41 in 2020, and when quantity was accounted for the unit price of dried flower was $7.09 in 2019 and $6.83 in 2020 (CAD). The mean price of legal dried flower decreased in 2020 ($12.63 vs. $11.16; p<0.001), but remained more expensive than illegal dried flower in both years ($12.63 vs. $9.04 in 2019; p<0.001, $11.16 vs. $9.41 in 2020; p<0.001).

Study 4 – The most common purchase source in ‘illegal’ and ‘medical’ states was dealers (2019: 48.8% and 43.5%; 2020: 50.6% and 39.3%, respectively) compared to stores/dispensaries in ‘recreational’ states (2019: 58.7%; 2020: 57.9%). The mean unit price (price-per-gram) of dried flower in US ‘recreational’ states increased in 2020 ($11.44 vs $12.24 USD; p=0.044), and when quantity was accounted for the unit price was $7.34 in 2019 and $7.95 in 2020. Consumers in ‘recreational’ states reported purchasing 74.0% and 77.0% of their dried flower from legal retail sources in 2019 and 2020. The odds of purchasing from legal sources was greater with each additional year with a legal retail recreational market (AOR=1.68, 95%CI: 1.58, 1.77), with few other sociodemographic differences.

Conclusions –The legalization of non-medical cannabis in Canada and some US states represent a unique opportunity to evaluate a novel policy with public health implications. The current thesis demonstrated that consumers show relatively quick uptake of the legal market in the first few years of legalization, and that price and retail availability contribute to consumers purchasing legally; however, it may take several years until retail markets and supply chains establish. At least in the immediate post-legalization period, retail availability may be relatively more important than price in encouraging consumers to purchase from the legal market. With the Canadian government undergoing an evaluation of the Cannabis Act, beginning October 2021, the current evidence may contribute to the legislative review.