When countries legalise cannabis, do people drink more alcohol than before, or less? 

That’s a question you might discuss, perhaps over drinks watching a match, or not at all, if you’re participating in Dry January. It’s also an increasingly relevant question, as more countries legalise cannabis for medical purposes, and some consider allowing recreational use too. 

Research in the United States regarding alcohol-cannabis relationships has so far produced conflicting results. Some studies found that alcohol use decreased on average in states that allowed medical and/or recreational cannabis use. This would suggest cannabis use tends to replace alcohol use. Conversely, other studies found that alcohol use increased in those states, implying that cannabis consumption tends to encourage drinking. And some studies found no alcohol-cannabis connection at all.

A complication for all this research, however, is that cannabis remains illegal under U.S. federal law. That’s true even where state laws allow medical or recreational use. That confusing and contradictory situation likely discourages some Americans from using cannabis. Or, from admitting they use it. And that could have affected the U.S. alcohol-cannabis research results. In Canada, however, cannabis is fully legal.

Cannabis in Canada

Canada has allowed legal medical cannabis use since 2001. But usage began increasing substantially only after regulatory changes in 2015. 

Under Canada’s current medical rules, any physician can authorise cannabis treatments for their patients. Each authorisation specifies a length of treatment (up to one year) and defines a maximum daily dosage. After being authorised, most patients register with a licensed cannabis producer. This allows them to buy legal cannabis products online. (Canadian pharmacies do not sell medical cannabis.) Or, some patients choose to instead register with the government, and grow their own cannabis plants

Physicians can also divide a patient’s daily dosage between two authorisation forms. This lets the patient register with two producers and buy half their products from each. It also means that registration numbers somewhat exceed patient numbers.

As of March 2022, Canada had 283,302 medical cannabis registrations. That’s about seven registrations for every thousand people. By comparison, the U.K. has only about 17,000 patients with medical cannabis prescriptions. That’s about 0.25 patients per thousand people. Meanwhile, Israel has around 11, and Germany has about three

On 17 October 2018, Canada legalised cannabis for recreational purposes. The medical cannabis system still exists. But, a much larger recreational cannabis market now operates alongside it, selling products via licensed stores and websites.

My research

I’m a university professor who studies Canada’s cannabis legalisation. So far I’ve examined the recreational market’s initial shortages of dried cannabis, the importance of having enough licensed stores, and the continued sales growth despite pandemic disruptions.

In my newest study, I examine the relationship between medical cannabis and alcohol. The study is based on data from April 2015 until September 2018, the period from when medical cannabis sales started rising, until just before recreational cannabis sales began. I compared sales of legal medical cannabis to those of beer, wine, and liquor stores. My calculations account for alcohol’s seasonal sales patterns and price fluctuations, and adjust for changes in economic activity, unemployment rates, and impaired driving penalties.

So, did access to medical cannabis decrease alcohol sales? The analysis found a small negative correlation between medical cannabis sales and alcoholic beverage sales. For each dollar of legal cannabis sold, alcohol sales were roughly $0.74CAD lower than expected. Consequently, Canada’s 2017-2018 alcohol sales were about 1.8% lower than they otherwise would have been.

I would note that the study showed that cannabis sales increases were correlated with alcohol sales decreases. While it did not prove they caused the decreases, the results do suggest that medical cannabis use replaced some drinking. 

Broader implications?

These findings in turn imply more generally that reduced alcohol use could partly offset cannabis legalisation’s harms and benefits. For example, increased cannabis use after legalisation presumably causes more cannabis-related health problems. But, it might also mean fewer alcohol-related ones. Similarly, while governments are considering the new tax revenues they could collect from legal cannabis sales, they might simultaneously be collecting less money from alcohol taxes.

Of course, it’s unknown whether my Canadian results will apply to cannabis legalisation in other countries, like the UK, Germany, or Luxembourg. Government regulations and consumer preferences differ from place to place. So, while my study seems to be the first comparison between fully legal cannabis sales and alcohol, it’s unlikely to be the last.

Michael J. Armstrong is an associate professor who studies the economic aspects of cannabis legalisation at Brock University in Canada.