With only twelve states having legalized marijuana for recreational use, it’s hardly surprising that an estimated 84% of 2018 cannabis sales in the U.S. were illegal. What might be more surprising is that at least 7% of the money spent on illicit purchases nationally took place in one of those twelve states where weed is legal.
Much has been written about how ending the prohibition on recreational use of cannabis in many states has resulted in a thriving parallel illegal market, leading many politicians as well as police departments to decry the advent of legalization. The blame has often been directed to the lack of supply relative to the boom in demand. This is true, to a point, but as any Economics 101 student will tell you supply and demand will reach equilibrium over time if allowed to do so. Right now there’s a significant barrier to achieving that equilibrium, and it’s far from being out of the control of those same politicians: a shortage of licensed retail outlets.
Of course, we recognize that there are reasons to not to toss out unlimited numbers of licenses indiscriminately. But let’s take a look at what happens when the number of licensed outlets grows large enough to supply the demands of the marketplace.
We have adequate data to run these analyses, drawn together from multiple sources, for five states that have legalized cannabis for recreational use: California, Oregon, Washington, Nevada, and Colorado.[i] Admittedly this number is too small to be a projectable sample, but the data provides a more than solid demonstration that the predictions of economics are in fact playing out in the marketplace. In these five states, the percent of cannabis sold illegally ranges from 14% in Colorado to 57% in Nevada.
Figure 1 graphs per capita sales of legal weed versus the price of an ounce for the five states. (This is the price per ounce for high-quality cannabis, which is the only price information available for all states; however, it’s fair to assume that higher prices for good weed are indicative of higher prices overall.)
Figure 1 demonstrates the most basic premise of supply and demand: as the price of a good increases, sales tend to decline. However, price alone doesn’t explain everything that’s going on here. Colorado is a big outlier, with the highest per capita sales despite a mid-level price, With Colorado, the trend line provides a fairly good fit to the data; without Colorado the fit is excellent.
If there are good substitutes for a product, higher prices can also make those substitutes more appealing – and illegal weed is in many ways a good substitute for the legal kind. It’s no great surprise, then, that where the price of cannabis is higher a greater percentage of marijuana is sold illegally. Figure 2 demonstrates this, with Colorado a bit of an outlier once again. (It’s entirely possible that Coloradans are just more likely to be law-abiding citizens.) But Figure 2 certainly supports the point that high prices can push more people into the black market. (Admittedly, some of them never left.) But why are prices so much higher in some states than others?
A lot of factors influence the price of a good, but the ease of availability is a significant one, and Figure 3 shows just how important this is in the cannabis market. The price of legal weed is strongly correlated with the number of stores licensed for recreational sales per 100,000 people.
The R2 statistic, which statisticians use to measure the strength of the correlation, is 82%. (An R2 of 100% represents perfect correlation.) In other words, 82% of the variation in the price by state can be mathematically explained by the variation in licenses per capita by state. That’s evidence
of a pretty strong, consistent relationship – there aren’t any real outliers here.
The first three graphs, and our main point, are essentially summarized in Figure 4A, which relates the percentage of weed sold illegally to the number of licenses. The R2 is a solid 57%.
But the truth is that we’re not limited to only five data points for this analysis, because we know the values of these two variables for virtually all states. In the large majority the number of licenses is zero and the percentage of weed sold illegally is 100%. And so we added this data point – really 42 data points sitting one on top of the other – to create Figure 4B. This raises the R2 for the relationship between these two variables to an even more compelling 84%.
And so if a state is serious about reducing the size of the illicit market for recreational cannabis, the first step on the most effective path to a solution probably isn’t blaming legalization. Better results would be obtained by expediting the process of granting more licenses to responsible would-be sellers, bringing down the price and making the black market relatively less appealing.