Marijuana and Tax: It’s Complicated

The other day I was walking my dog through a park and I saw two children gleefully playing on a seesaw. As I returned home and got back to work the image of the seesaw stuck with me. The seesaw is a great metaphor for the legalized marijuana tax conundrum because it demonstrates the challenge of balancing interests. If we imagine California’s marijuana tax scheme as the seesaw, we can see the Golden State’s two primary objectives sitting at each end: generating state revenue on one end and reducing externalities (social harms attributed to marijuana use) on the other. Although both objectives can play nice, if the interests are not balanced and cognizant of the other, one end of the seesaw will crash into the dirt while the other soars sky high. And the dirt in this case isn’t the playground kind that can be dusted off; it is muddied public policy that would incentivize the black market.

The good news is that balancing these objectives is attainable and reasonable. First, generating state revenue is attainable once a system to tax marijuana is in place. Colorado is a great example of this: it gained a new source of revenue by implementing a multi-faceted tax approach that generated $52,570,081 in 2014. The recreational marijuana market generated $40,968,203 alone. Second, tax can serve as a mechanism to offset some of the most salient social harms attributed to marijuana use such as youth access and use, drug crime, and the health risks of ingesting unregulated products.

The difficulty of deciding how to tax marijuana is not only in trying to balance interests. The idiosyncrasies of the marijuana market add to the difficulty of creating a balanced tax structure. We have very limited data on the marijuana market due to its status as an illegal drug. As such, it is difficult to gather reliable information about marijuana users and how they respond to market changes. This creates a lot of guesswork and “best estimations” which may be wrong and could create problems once in play within a burgeoning market. In addition, the marijuana plant itself is unlike anything else we have taxed before. Alcohol and tobacco are the nearest in kind to marijuana but both substances are easier to tax by nature. We are able to tax distilled spirits by alcohol content because alcohol is a homogeneous material that lends itself easily to testing. Marijuana, on the other hand, is not homogenous and technology still needs to be developed for reliable potency testing processes. The heterogeneity of marijuana plants also distinguishes it from tobacco because taxing marijuana by unit or weight is not as effective as it is to tax cigarettes (details of this are discussed below).

In order to meet our objectives we need to establish the type of tax, basis for the tax, the rate, the collection point, and the market structure that will best achieve our objective(s). None of these variables is mutually exclusive and there are various ways to stack them. For example, marijuana can be taxed by weight, potency, price, or even by product type (e.g., raw material, edibles, concentrates). The tax can be imposed on growers, processors, retailers, consumers, or some combination of these. Each of these combined, however, has different degrees of efficacy. If we really want to meet both objectives (revenue and reducing negative externalities), we’ll need to consider the different benefits and drawbacks to each approach. Each approach will require different tradeoffs and will have real-life effects on consumer usage, safety, and even business sustainability. The discussion on how to tax marijuana is lengthy and complicated. Today, the goal is to start the conversation by looking at the various tax bases and how each can help California meet its tax objectives.

Let’s move now to discussing how the various tax bases fare on the seesaw of generating revenue and offsetting externalities. First, a tax based on price is easy to administer but does not necessarily help control prices. This will not bring in much revenue if post-legalization prices plummet as predicted by many in the industry. Revenue aside, there are numerous other reasons price control is desirable. The most salient is thwarting youth access. It is well established that youth substance use rates are particularly price sensitive. Youth access and use decrease as prices increase. One important caveat to a price-based tax is that it only remains easily administrable if it functions in a horizontally integrated system. Rather than have “jacks of all trades,” horizontally integrated systems feature distinct entities with limited roles within the market (e.g., growers, processors, and retailers). The RAND study “Considering Marijuana Legalization” explains that a horizontally integrated system is preferable when imposing price-based taxes because they depend on the existence of a real price—normally, on an arm’s-length sale between unrelated parties. On the other hand, if prices do start and remain high, a price-based tax can be too high and push consumers into illegal black markets. Going back to the seesaw image, it is apparent that a price-based tax alone cannot achieve both objectives of legalizing marijuana. The seesaw of marijuana tax objectives can only come close to being balanced if we compromise the amount of revenue generated for more positive impacts on social harms. Further, the outcome of a balanced seesaw depends on multiple variables aside from price alone.

Marijuana can also be taxed by weight. Similar to a price-based tax, a weight-based tax is also easy to administer. It also has the distinct feature of bringing in a steady amount of revenue since it isn’t determined by market price. The difficulty of a weight-based tax is in the details. For instance, the weight of each marijuana plant changes over time due to drying and trimming and weight changes aren’t exact among each plant. We have to establish timelines for when marijuana plants should be weighed and our timelines may not coincide with the initial objectives of taxing the product. The main concern, however, is the risk that it will incentivize production of higher potency products. This is a disadvantage for three reasons. One, consumers may unwittingly ingest more potent products, which may lead to health and safety issues. Two, the tax base wouldn’t encourage consumers to purchase less potent products. Third, potential revenue for higher potency products would never be captured within the market.

Another option is a tax based on potency or product type which has the benefit of addressing more externalities but would be difficult to administer given our current lack of knowledge regarding marijuana compounds. It’s important to clarify that for purposes of brevity in this post, I am conflating potency and product type (some edibles might be less potent, others might be more). A tax based on potency and/or product type can be used to shape usage patterns and implicitly impose limits or restrictions on how marijuana can be consumed. Therefore, it is advantageous in the sense that it addresses more social harms related to marijuana use. For example, California could impose a higher tax on products that typically yield higher potency and are more difficult to dose appropriately (e.g., edibles and concentrates). This could encourage users to purchase products, such as raw marijuana, which is typically smoked. This could be a desirable outcome because our familiarity with this form of marijuana enables us to better predict behavioral outcomes and leads to shorter durations of intoxication. In contrast, California could promote lung health and discourage smoking marijuana by taxing the raw material higher than other ingestible forms. Potency could be measured by the amount of THC, CBD, or THC:CBD ratio. However, the difficulties and inaccuracies in current technology for testing potency pose an immediate obstacle to the feasibility of such a tax. Plus, even a single marijuana plant can have inconsistent potency throughout. Finally, our experiences with the tobacco and alcohol industries have taught us that higher taxes do not always have positive impacts on public health.

At its inception, any potency-based tax should have at least three features. First, a potency-based tax would have to be set in ranges. These potency ranges would need to be set high enough to discourage divergence to black markets. Second, any potency tax would need to include a provision for revision based on timely reviews of any developments in new information regarding potency ranges and/or limits. Third, the previous feature necessarily requires at least some of the revenue be dedicated to research that can advance the system. Research could explore various topics such as testing modules, scientific improvements in measuring potency, and studies on the effects different products and potency had on users’ intoxication levels.

This discussion of the various ways to tax is only the tip of the iceberg and will be covered in more detail in future posts. Of particular interest is the history of alcohol and tobacco taxes. Both can serve as a great resource for what works and what doesn’t in terms of taxing to affect use and offset social harms. After we cover the lessons learned from above, you can join me in the related discussion of which market structure (vertical or horizontal integration) better facilitates particular tax methods. Join me next time for a discussion on sin taxes and the impact they can have on the burgeoning marijuana market.

Alexa Quinn for Drug Law and Policy

Follow Drug Law and Policy on Twitter @DrugLawPolicy or Alexa @aquinn_dlp. You can also contact Alexa Quinn by email:

Comments are closed.