Whether you like or loathe vanilla, you can’t deny that it’s everywhere and in everything. You might think that it would be great for cannabis to be “the new vanilla.”
It wouldn’t. It would be terrible for North America’s cannabis industry, and perhaps even worse for cannabis farmers in developing countries.
I will explain this by analogy between cannabis and the “vanilla flavoring” industry. Most otherwise-unsourced claims about the vanilla flavoring industry were borrowed from Mark Schatzker’s 2015 book The Dorito Effect (pages 41-50) with other online sources as linked. All factual errors are my own.
Originally, all vanilla flavoring came from the bean-pods of the Mexican tropical orchid Vanilla planifolia, which has since been transplanted around the world. Today, Madagascar — a large island off the south-east coast of Africa — produces the world’s best vanilla bean-pods and flavorings (mostly extracts). 80-85% of all farmed vanilla flavoring comes from Madagascar, where approximately 80,000 smallholder vanilla farmers grow the vanilla plant. Farming, harvesting, and processing these bean-pods into high-quality vanilla flavoring is an extremely complex and labor-intensive craft (video). Production of 1 kg of vanillin requires approximately 500 kg of vanilla pods, corresponding to the [hand-]pollination of approximately 40,000 flowers. Josephine Lochhead, President of the Cook Flavoring Company, stated that Madagascar’s farmers “grow and cure these singular pods with such painstaking hand labor and care that we in the West can scarcely fathom it.”
Globally, there are few “boutique” vanilla farms like this one, and some tourism-oriented vanilla farms like this one, but their production is a drop in the ocean.
Lessons for cannabis:
Vanilla farming requires as much deep knowledge, broad experience, and hard work as cannabis farming does, yet vanilla farmers are extremely poor. When cannabis prices collapse globally, cannabis farming will shift away from the developed world to the developing world, and even there, become as unprofitable as vanilla farming. North American cannabis farmers will not be able to compete.
Yes, Big Cannabis could prohibit the importation of poor foreign cannabis farmers’ products into North America, but such prohibition won’t help North American cannabis farmers, for the reasons discussed below.
In the 1800’s, chemists discovered that vanilla’s aroma and flavor came primarily from a single chemical compound, which they named “vanillin.” (The chemical structure of vanillin is shown at the top of this article.) Starting in 1874, other chemists discovered ways to make artificial vanilla from inputs that were much cheaper than vanilla bean-pods. The cheapest process — synthesizing vanillin from petroleum — was commercialized 100 years later, in the 1970’s.
Vanillin (4-hydroxy-3-methoxybenzaldehyde) is exactly the same chemical compound whether it is made by the chemical reactions occurring inside a vanilla plant or made by the chemical reactions inside a laboratory. What makes some vanillin “natural” and some “artificial” is not what it is, but how it is made. We’ll come back to this point shortly.
Artificial vanillin supplies more than 99% of the global demand for vanilla flavoring (20,000 metric tons), with less than 1% being supplied by extracts from farmed vanilla (40-50 metric tons). This is not surprising, given the huge price difference: in 2010, farmed vanilla extract cost $1,200/kg vs. $10/kg for vanillin(120x). After Madagascar was recently hit by cyclones and drought, the cost of farmed vanilla extract shot up to US$11,000/kg (1,100x)…. yet Madagascar’s farmers remain poor.
Lessons for cannabis:
Artificial cannabinoids are being manufactured by the metric ton, legally, in North America today. Production is high-grade and small-scale (merely “metric tons,” not container-loads), so their prices are higher than they will be, later.
As these factories scale up and expand from pharmaceutical-grade to food-grade, North America could start producing artificial THC, CBD, and other valuable cannabinoids at “vanillin-like” costs: $10/kg (vs. $3,000/kg today for CBD isolated from farmed cannabis). These cost reductions won’t happen overnight, but over perhaps a decade. Competition is likely to be fierce.
Just as with vanillin, artificial THC, CBD, CBG, etc. are likely to be “good enough” to satisfy 99% of global demand. All of the world’s cannabis farmers will be fighting over the remaining 1% of global demand, and only the world’s best cannabis farmers, in its poorest developing nations, will survive…if you call extreme poverty “surviving.”
Right about now, you may be thinking, “Single cannabinoids are much less effective than whole plant extracts. A single artificial cannabinoid can’t out-compete a whole-plant extract. North America’s cannabis farmers will be saved by the Entourage Effect. I don’t need to worry after all! Yea!”
Not so fast.
In the late 1970’s, the McCormick spice company sought to develop an inexpensive vanilla flavoring that more accurately matched that of farmed vanilla. First, McCormick identified the thirty-odd chemical compounds — in addition to vanillin — that most-strongly contributed to the flavor and aroma of pure extracts of vanilla bean-pods. Next, McCormick mixed cheap versions of these chemical compounds (most of them artificial) together in exactly the same proportions as they occurred in the world’s best farmed vanilla. McCormick’s resulting “Imitation Vanilla” sells at a wholesale price that is perhaps 5 times more ($50/kg) than vanillin alone ($10/kg), but that’s still 20 to 200 times less expensive than farmed vanilla extract. The flavor and aroma of Imitation Vanilla is is a “perfect match” (that is, subjectively indistinguishable) from the pure extract of Madagascar’s best farmed vanilla (according to The Dorito Effect, page 49).
Lesson for cannabis: Chemists are very likely to be able to blend artificial cannabinoids, terpenes, and flavonoids into an imitation extract that’s subjectively indistinguishable from, and as effective as, the extracts from the very best farmed Gorilla Glue, Jack Herer, Charlotte’s Web, or any other cannabis strain. Furthermore, such an “Imitation Cannabis Extract” may eventually cost an “Imitation Vanilla-like” wholesale price of perhaps $50/kg (5 cents/gram) — much less that the cost of the equivalent extract from farmed cannabis.
Sounds incredible, right? As incredible as “potato chips that taste like lasagne, or like lobster hot plate”… but those exist right now, and they are the result of precisely the same combination of chemical analysis and artificial synthesis.
Maybe now you’re thinking, “But Imitation Cannabis Extract would be artificial, and no one likes anything artificial nowadays. The big brands are going ‘all-natural.’ North America’s cannabis farmers will be saved by the natural foods movement! I don’t need to worry after all! Yea!”
Why bother with biosynthesis, when biosynthetic vanillin costs nearly 50 times more than vanillin produced using the proven, industrialized processes of artificial chemistry? And when civil society organizations have opposed biosynthetic vanillin?
The main attraction of biosynthesis is that the US Food and Drug Administration (FDA) currently defines of the word “natural” as follows (emphasis added):
The term natural flavor or natural flavoringmeans the essential oil, oleoresin, essence or extractive, protein hydrolysate, distillate, or any product of roasting, heating or enzymolysis, which contains the flavoring constituents derived from a spice, fruit or fruit juice, vegetable or vegetable juice, edible yeast, herb, bark, bud, root, leaf or similar plant material, meat, seafood, poultry, eggs, dairy products, or fermentation products thereof, whose significant function in food is flavoring rather than nutritional.
‘Natural flavouring substance’ shall mean a flavouring substance obtained by appropriate physical, enzymatic or microbiological processesfrom material of vegetable, animal or microbiological origin either in the raw state or after processing for human consumption by one or more of the traditional food preparation processes listed in Annex II [which includes fermentation].
By these definitions, vanilla flavoring (whether just vanillin or a more complex mix of chemical compounds) may legally be described as “natural” if its flavor chemicals were extracted from:
a vanilla plant, or
yeast fermentation products, including the fermentation products of genetically-modified yeast.
You read that right: Vanilla flavoring whose ingredients are extracted from the fermentation products of genetically-modified yeast strains can, legally, be labelled as “natural vanilla flavoring.”
[Solvay’s biosynthetic] Rhovanil® US NAT is a highly pure, naturally-sourced ingredient that is easy-to-use and serves as a one-for-one drop-in replacement for synthetic vanillin. This means that US food and beverage manufacturers are better positioned to deliver on-trend ‘natural’ and ‘clean label’ products without compromising on consumer appeal.
But surely, this could never happen to cannabis, right?
A few months ago, Canada-based cannabis giant The Cronos Group invested US$122 million in US-based Ginkgo Bioworks. Under the terms of the deal, Ginkgo will genetically engineer yeast to produce, through fermentation, the eight cannabinoids THC(A), CBD(A), CBC(A), CBG(A), THCV(A), CBGV(A), CBDV(A), and CBCV(A). The target price is US$1,000/kg — 1/3rd of the price of CBD extracted from farmed cannabis today — but that’s just to start. Economies of scale could push the price much lower.
Jason Kelly, CEO of Ginkgo, stated that brewing cannabinoids from yeast is “cheaper, it’s not subject to weather conditions, the price isn’t all over the place, it’s not different if you grow it in Morocco or somewhere else, it’s just a much better product. The reality is that brewery economics is going to wipe the floor with farming economics.”
This isn’t “blue sky” research. Ginkgo has already genetically-engineered yeast strains to biosynthesize other chemical compounds. Furthermore, Ginkgo is not alone. Many other companies are researching the same thing. Among them, the technological problems will eventually be solved. This is going to work…and soon.
The eight cannabinoids listed in the Cronos-Ginkgo deal are only the start. Eventually, yeast biosynthesis will almost certainly be able to produce all of the cannabinoids, terpenes, and flavonoids needed to produce a vape oil that is subjectively indistinguishable from the extract from your favorite farmed cannabis strain — and the biosynthetic vape oil can legally be marketed as “pure and natural.”
You might imagine that North America’s leading cannabis farmers could organize to exclude extracts from vat-brewed genetically-modified organisms from the US FDA’s definition of “natural.” This will be difficult, because those same “cannabis industry leaders” are funding the development of biosynthesized cannabinoids! Cronos invested in Ginkgo Bioworks; Organigram in Hyacinth Biologicals; Aurora Cannabis bought Anandia Laboratories; etc.
Some Non-Governmental Organizations (NGOs) are pushing back against the legality of using “natural” to describe the products of biosynthesis (see here and here). The stated goal of one of them, the ETC Group, is to “address the socioeconomic and ecological issues surrounding new technologies that could have an impact on the world’s poorest and most vulnerable people… especially in Africa, Asia and Latin America.” It is not clear how much these NGOs might care about any impacts on North American cannabis farmers.
Weaknesses of the analogy
No analogy is perfect. Vanilla is not cannabis.
There are fewer strains of vanilla than of cannabis (half a dozen vs. an uncountable plethora).
The vanilla plant does not produce industrial fiber, nutritious seeds, or other such potentially-valuable outputs, as cannabis does.
Most farmed vanilla is consumed in as extracts, whereas most cannabis is consumed by smoking, vaping, or eating cured flower material (although that is changing — see item #3 below).
Strengths of the analogy
The vanilla-cannabis analogy has many strengths:
Like vanilla, cannabis is consumed recreationally.
Like vanilla, cannabis is also consumed medically.
The preference of cannabis consumers is shifting rapidly away from flowers to extracts in non-smoked products (oil drops, vape oils, dabs, edibles, etc.). Once the flower is replaced with extracts, the extracts are easily replaced by non-farmed cannabinoids.
The willingness of consumers to smoke non-cannabis plant matter imbued with synthetic cannabinoids (K2, Spice, etc.) demonstrates that demand exists that can be met profitably (if, at present, illegally) by non-farmed cannabinoids.
The economics are the same for cannabinoids as for vanilla: It’s cheaper to make them through industrial chemistry than through biosynthesis; it’s cheaper to biosynthesize them than to import them; and it’s cheaper to import them than to farm them domestically.
Under current rules, consumers’ demand for “natural” products can be met by biosynthesized vanillin and cannabinoids/terpenes/flavonoids.
The cannabis industry’s over-focus on just two cannabinoids — THC and CBD — strengthens the analogy.
If market demand for oils, extracts, and isolates is met primarily by non-farmed cannabis, then the only market demand remaining for farmed cannabis will be (a) connoisseurs smoking the traditional trimmed cannabis flower, and (b) manufacturers whose branding requires them to use “all natural” ingredients but not biosynthetic cannabis. This demand is much smaller than the multi-billion-dollar projections being made by the “cannabis industry boosters.”
Farming cannabis flowers, like farming vanilla bean-pods, is very labor-intensive, so it is likely to shift to nations with very low labor costs, which would then export cannabis to North America (and elsewhere).
When artificial vanilla was first synthesized in 1874, it was only slightly cheaper than farmed vanilla. In the 144 years since then, the relevant science and engineering have improved so much that the wholesale price of vanillin is 1/1000th the price of farmed vanilla. This price decline took a more than century.
As another data point, in the late 1970’s, it took McCormick four years to research and develop the formula for Imitation Vanilla.
Today, however, the pace of technological change is much faster (see The Law of Accelerating Returns). All else being equal, the price of artificial and/or synthetic cannabinoids could reach 1/1000th the price of farmed cannabis within twenty years, and maybe ten — much less than “a century.” Likewise, modern technology could accelerate the development of “Imitation Cannabis Extract.”
Consider this interview (video, starting at 6:27) of Cronos’ CEO regarding Cronos’ investment in Ginkgo’s biosynthetic cannabinoids. In it, Cronos’ CEO describes biosynthetic cannabinoids as costing 1/10th the cost of farmed cannabinoids to start. As his interviewer then observed (same video, 7:20), “That implies that this massive build-out of cannabis [farming] capacity is already overdone.” (Emphasis added.)
“To be clear, the industry may end up being overbuilt substantially if biosynthesis emerges as a scalable technology.”
That is: Time is not on the side of cannabis farmers. Unless your farm producers boutique quality at developing-world costs, you’re already Blockbuster Video in a Netflix world.
[Update (24 July 2019): The Cronos Group recently announced its acquisition of a “fermentation and manufacturing facility” to “produce cultured cannabinoids at commercial scale with high-quality and high-purity.” This development suggests that yeast-synthesized cannabinoids will hit the North American market, in commercial quantities, in 2020 — at least a year earlier than I had expected. The profit-maximizing strategy for Cronos is to sell these cannabinoids to the pharmaceutical industry at prices that are only slightly lower than farmed isolates, until Cronos’ capacity ramps up, at which point Cronos can flood the market at prices low enough to drive their existing competition (including cannabinoid farmers in the developed world) out of business and dry up the supply of capital that would otherwise fund new competitors. This whole sequence could play out in five years, or even faster.]
The human body’s endocannabinoid system makes it own cannabinoids to balance the body’s internal functions. The human body has no “endovanillin system.” Cannabis is simply more complex than vanilla.
One measure of “plant complexity” is the number of chemical compounds it contains. Farmed cannabis contains 483 chemical compounds — nearly three times as many as farmed vanilla (170), and about the same as chocolate (400), but only half those in coffee (1000).
Scientific studies support the “complexity” counter-argument, by validating the Entourage Effect, indicating that cannabis’ chemical compounds interact with each other and with the human body in complex ways.
[Update (24 July 2019): I just learned about a “100% natural” synthetic whiskeynamed Glyph that won a gold medal and three silver medals at recent blind tastings. That’s clearly a “successful” synthesis. This, despite the flavor of whisky being affected by at least 4,271 unique chemical compounds, which is nearly ten times as many as farmed cannabis. The successful synthesis of whiskey, despite its being an order of magnitude more complex than cannabis, is strong evidence against the “cannabis is too complex to synthesize” argument.]
Will cannabis’ complexity save North America’s cannabis farmers? No, for three reasons.
Look at the ingredients label of your favorite soda or energy drink. You are likely to find “caffeine,” but you are unlikely to find “coffee.” Similarly, Coca-Cola may someday add CBD to its sodas, but is much less likely to add “full-spectrum cannabis extracts.” Most of the caffeine in caffeinated beverages is synthetic. The same argument applies to THC-infused beer. [Bio]synthetic single cannabinoids are likely to be “good enough” to be ingredients in other products… especially if the price is right.
Despite the complexity of coffee, you can buy artificial coffee flavoring by the barrel. Same with chocolate. Odds are good that you have already consumed a lot of these, in one product or another…and that they were good enough that you didn’t notice, or cheap enough that you didn’t care.
If Coca-Cola did choose to add farmed cannabis extract, it would almost certainly buy those extracts from cannabis farmers in the developing world — again, to get a cheaper price — as described here.
It has also been suggested that cannabis consumers will avoid the artificial and/or biosynthetic cannabinoids, avoid the inexpensive imports, and instead pay a substantial premium for “the good stuff.”
Will a willingness to pay extra for premium-quality craft cannabis save North America’s cannabis industry?
No, for two reasons.
First, “price” is the top-listed buying criterion of cannabis consumers, outscoring “familiar brand” by 5:1. As in every other market, at any given level of quality, price is the primary determinant of sales. There will, of course, be some “cannabis connoisseurs” who will be willing to pay a premium for “the good stuff.” Which bring us to the next point.
Second, the price premium commanded by “the good stuff” isn’t big enough to overcome North America’s higher costs of production. As described here, the price premium for:
Craft beer is 4:1
Specialty chocolate is 17:1
Fine wine is 21:1
The global wholesale commodity price of cured psychoactive cannabis flower is, as discussed here, likely to fall to roughly 1¢/g (that is, $4.50/pound). Let’s imagine that the best North American boutique cured cannabis flower attracted a price premium equivalent to fine wine’s 21:1 over the global commodity price. That implies a wholesale price of 21¢/g for “boutique” cannabis.
Unfortunately, as described here, the cost of growing cannabis in North America’s lowest-cost growing region is 70¢/g — more than three times cannabis’ future “boutique price.”
As the old Danish proverb says: “It is difficult to make predictions, especially about the future.” Nevertheless, despite this warning, I shall persist in making predictions based on the vanilla analogy to cannabis, with the above caveats, as follows.
It is likely, within the next seven years, give or take, that:
Artificial cannabinoids will be offered to the market as both (a) single-cannabinoid isolates and as (b) blends of cannabinoids, terpenes, and flavonoids that most consumers will consider to be “good enough” for most applications. These will cost approximately 1/20th (or less) of the price of equivalent farmed-cannabis derivatives. They will be labeled as “artificialingredients.”
Biosynthetic cannabinoids will be offered likewise, but at price that is in between the lower-priced artificial cannabinoids and the higher-priced farmed-cannabis derivatives. The biosynthetics will be labeled as “naturalingredients.”
The above-described “non-farmed cannabis products” will, between them, capture most of the non-flower market for cannabinoids and cannabis extracts. (Not 99%, as with vanilla, but most of it.)
Foreign-farmed cannabis imports will (for the reasons discussed here) capture more than 90% of the North American market for cannabis flower, at wholesale prices that are considerably lower than the North American cannabis farmer’s cost of production.
Implications for the North American cannabis industry
As a consequence of the above, it is extremely likely that, in North America (and the OECD nations more generally), within the next seven years, give or take:
Prices: The price of non-farmed single cannabinoids and of blends of cannabinoids, terpenes, and flavonoids will collapse by 80-90%. The price of farmed cannabis extracts will fall by almost as much.
Non-farmed: Nearly all market demand for cannabinoids in North America will be met by a combination of (a) non-farmed single cannabinoids and blends of non-farmed cannabinoids, terpenes, and flavonoids, and (b) imported farmed cannabis and/or extracts thereof.
Medical: Nearly all market demand for medical cannabis in North America will be met by non-farmed single cannabinoids and blends of non-farmed cannabinoids, terpenes, and flavonoids.
Recreational: Nearly all of the farmed cannabis (and farmed cannabis extracts) imported to North America will be for the adult-use (recreational) market.
Stock prices: The market value of most North American cannabis cultivators will collapse by 95-99% from their current levels.
Farming: The commercial farming of cannabis in North America will collapse, leaving only “lifestyle” businesses operating tourism-oriented boutique cannabis farming and perhaps hemp farming for seed & fiber (with low-quality flower as a by-product).
Implications for the global cannabis industry
Within the next 7 years, give or take, cannabis will be commercially farmed only in the few developing nations that can produce and export top-quality, internationally-certified cannabis flowers and their extracts at the world’s lowest prices.
Suggestions for investors in the global cannabis industry
A. Ask cannabis industry CEOs:
[Cultivator] Should imports of foreign farmed cannabis and extracts into your firm’s home country be prohibited? If prohibiting imports also prohibits exports, do you still favor prohibition?
[Cultivator] According to your firm’s strategic planning, what percentage of global demand for cannabinoids will be met by non-farmed cannabinoids three years from now? Five? Seven? What is your firm’s strategic plan for addressing the shrinking market share of farmed cannabis?
[Cultivator] According to your firm’s strategic planning, in what year will the price of non-farmed THC and/or CBD fall below the cost at which your firm can produce it from domestically-farmed cannabis?
[Retail, Distribution, Delivery] After a bump right after legalization, average wholesale cannabis prices in your jurisdiction have been falling. If prices continue to fall, all else being equal, at what price do you start losing money? What is your firm’s plan to address this issue?
[Ancillary] What is your firm’s strategic plan for surviving, and profiting from, the relocation of many links in the cannabis industry’s global supply chain to low-wage, climate-appropriate foreign jurisdictions?
[Any] Should labeling as “natural” any cannabinoids biosynthesized by genetically-engineered yeast be prohibited?
[Any] According to your firm’s strategic planning, which links in your global supply chain are you shifting overseas first, when, where, and why there?
B. Consider selling shares in:
Any cannabis company whose CEO cannot give a compelling answer to any of the above questions.
Any company whose margins fall when the price of cannabis falls (e.g., retailers, wholesalers, distributors, delivery services, etc.).
Any company whose primary business is selling products or services to the above company categories operating in OECD nations (e.g., “ancillary” products and services).
Any cannabis company that is more focused on branding than on reducing costs.
C. Consider buying shares in:
Developers of genetically-modified organisms for, and/or manufacturers of, biosynthetic cannabinoids (especially if regulators and/or judges decide that the word “natural” MAY be used to describe the outputs of genetically-engineered organisms).
Manufacturers of artificial cannabinoids (especially if regulators and/or judges decide that the word “natural” may NOT be used to describe the outputs of genetically-engineered organisms).
Any link in the global cannabis supply chain that is located in one of the few developing nations that can produce and export top-quality, internationally-certified cannabis flowers, their extracts, and the products made from them, at globally competitive prices.