Viridian Key insights summary
Each week, Viridian publishes insights and analysis on completed capital raise transactions in the prior week, focusing on all equity and debt deals. Our analysis includes:
- Summary
- Outlook
- Best & Worst Perfromers
Quick Links
Summary
- YTD capital raises totaled $1.71B, down 10.3% from last year’s $1.91B. From an LTM view, capital raises totaled $2.12B, down 8.5% from the same period in 2024. Debt as a percentage of capital raised worldwide is 90.8%, compared to 56.6% in the previous year. U.S. raises YTD accounted for 91.1% of total funds, up from 68.6% at the same point in 2024. Raising from outside Canada and the U.S. accounted for 4.4% of total funds raised, slightly below the average of 5.3% over the six previous years.
- Public company raises accounted for 92.2% of total raises in the LTM period, the highest in at least the last 7 years.
![]()
- YTD capital raises for the cultivation and retail sector total $1.28B, up 43.3% from last year’s $890.5M. For the LTM period, the capital raised in the cultivation and retail sector was $1.54B, 33.3% higher than in 2024, which in turn was 160% higher than in 2023.
- Debt accounts for 94.5% of the funds raised over the last 12 months (LTM). Large debt issues (over $100M) accounted for 56.0% of capital raised, up from 0% in 2023.
![]()
- Cannabis equity prices (as measured by the MSOS ETF) fell by 4.42% for the week.
![]()
Market Commentary and Outlook
VIRIDIAN INSIGHTS
-
- THE ADVENT OF REVOLVING CREDIT AGREEMENTS IS A MAJOR ADVANCE IN CANNABIS CREDIT
- Over the last two weeks, Verano and Curaleaf have each announced new revolving credit agreements —the first of their kind in cannabis — and a sign of the maturation of the cannabis debt capital market.
- They are not notable for their size, and it’s tempting to dismiss them because they do not seem to make much of a dent in either company’s 2026 refinancing needs.
- But that is not their intent, and misses the real importance of these deals. The real key is that they offer an off-balance-sheet source of liquidity that will allow these companies to avoid holding quite as large cash balances and achieve better asset efficiency.
- Most major US companies maintain undrawn revolving credit lines to augment liquidity for unforeseen cash requirements. Still, until now, cannabis companies have been unable to attract this type of unfunded commitment.
- The revolvers for Verano and Curaleaf also signal investor confidence that they will be able to refinance their 2026 maturities. This is particularly true for Verano. The Chicago Atlantic agented revolver matures in 2028, after the required refinancing of the Chicago Atlantic agented term loan. Would CA offer this revolver if there was any real doubt that Verano would be able to roll its 2026 maturities? Conversely, the Needham bank revolver for Curaleaf matures before the company’s 2026 maturities, providing weaker vote of confidence.
- TARGET BEGINS SELLING THC BEVERAGES IN TEN TRIAL STORES IN MINNESOTA.
- The state has approved all 72 Target stores in Minnesota to sell hemp-derived products. The initiation is important because it further legitimizes a product category that the public clearly wants. If Target feels safe in carrying these products, why not grocery stores or convenience stores?
- It is unclear, however, how this will play against the growing tide of state legislation that threatens to eliminate intoxicating hemp products. Never mind what might end up in the new Farm Bill.
- We find it quite interesting that despite the FDA sending a warning letter to a hemp derived THC gummy manufacture as recently as July 2025, citing its position that both d8 and d9 were not allowed as food additives, no such warning letters have been sent to any of the growing number of hemp based THC beverage manufactures, despite the fact that these drinks are frequently sold via mail order.
- Why the difference? And for that matter, why has the FDA not proceeded against companies like Wanna and Wyld that sell d9 gummies through the dispensary channel?
- VIREO AND CHICAGO ATLANTIC USHER IN A NEW AGE OF CANNABIS RESTRUCTURING WITH THEIR SCHWAZZE TRANSACTIONS
- On October 2, 2025, Vireo entered into a definitive agreement to acquire approximately 86% of Schwazze’s outstanding senior secured convertible notes for approximately $62M payable in Vireo’s subordinated voting shares, valued at $0.54 per share. We estimate that the notes have PIK’d up to about $110M from the $103.1M figure in Schwazze’s 12/31/23 10K
- The Vireo transaction should be viewed in the context of a larger picture: the overall restructuring of Schwazze using credit bids via Article 9 of the UCC, a technique also being utilized in AYR, which we believe will be the future path of major cannabis restructurings. We refer readers to two other documents/ news stories for deep background; see the footnotes at the end.
- We view the set of transactions through three lenses:
- The acquisition of Schwazze debt is a huge win for Vireo. Controlling 86% of the fulcrum security places Vireo in control of the credit bid for the majority of the assets of Schwazze. We estimate that the principal value of the senior secured converts has PIK’d up to approximately $110M from the original $95M and the $103.1M shown in the 2023 10K. The purchase of 86% of the outstanding debt for $62M implies a price of 65.5% of par. More importantly, we believe the ongoing EBITDA of the Schwazze assets subject to the credit bid should be approximately $35 million. Grossing the $62M to account for 100% of the issue, we get $72M for the senior secured debt, which would give us only 2.1x EBITDA on the purchase price. We think, however, that one also needs to add the $45M senior lender refinancing of seller notes to get the total enterprise value of $ 72M + $45M = $117M. It is also likely that the senior lenders will end up offering the Schwazze equity holders a carrot to help this transaction cross the finish line. Still, Vireo appears to be able to acquire the core Schwazze assets for only about 3.3 times EBITDA without using any cash. The transaction is tremendously accretive.
- Chicago Atlantic’s strategic position as both a lender to Schwazze and a significant shareholder in Vireo positions it as a lender/operator hybrid with an ability to more aggressively assert creditors’ rights than typical cannabis lenders, who typically avoid direct ownership due to federal illegality, regulatory hurdles, and operational complexities. This unwillingness to take control of operators is a key reason why so much “Amend, Extend & Pretend” has occurred in cannabis. Chicago Atlantic, through its close relationship with Vireo, is different. It can play hardball when necessary, knowing it has a place to park and rehabilitate problem assets. CA doesn’t control Vireo; we estimate that CA will only own around 12% of Vireo post-closing of the transaction. However, the fact that the CEO of Vireo was a senior partner at CA tightens the relationship beyond what the raw numbers suggest. Chicago Atlantic and Vireo are uniquely positioned to roll up distressed U.S. cannabis operators.
- The transactions envisioned by Vireo and Chicago Atlantic usher in a new era of cannabis restructuring:
- Article 9 credit bids instead of receiverships
- Asset partitioning: selecting only the desirable assets and discarding the undesirable ones.
- Maintaining ongoing core operations rather than piecemeal liquidations
- An increasing use of “Liability Management Exercises”, also known as Creditor on Creditor Violence. Sharp-elbow techniques, common in the high-yield world, where senior secured lenders maximize their returns at the expense of other creditors, are making their way into cannabis restructurings. Subordinated debt is already scarce in cannabis, but it’s about to become even more so.
- Footnotes
- Altmore, another creditor at Schwazze, filed a complaint on May 3, 2025, in the District Court of Denver County Colrado, alleging that during discussions following the Company’s default on Altmore’s $15M senior secured loan, it was notified by Schwazze’s bankers that the Company and its secured lenders intended to pursue a restructuring transaction called a strict foreclosure under which the company’s assts would be divided into “good collateral” and “bad collateral” and a “NewCO” would acquire the “good collateral” while the “bad collateral” would be left behind. The banker informed Altmore that the bondholders and the company viewed Altmore’s collateral as “bad collateral,” but that it could possibly be included in the “good collateral” pool if Altmore agreed to reduce its principal balance by more than 70%. Altmore also alleged that the company had moved some of its collateral, including a preroll machine, out of the subsidiaries that Altmore is secured by. Altmore requests that a receiver be appointed to protect its creditors’ rights.
- A Bloomberg article on August 29, 2025, outlined a restructuring plan under which senior secured creditors would provide a $65 million cash infusion, allocating $45 million to refinance seller notes and the remaining $20 million to working capital. The senior lenders will then use a credit bid under Article 9 to acquire performing properties, while the others will be left behind to be wound down.
- DOES THE TRUMP VIDEO POST ACTUALLY MEAN SOMETHING? IN THE ABSENCE OF ANY REAL SIGNS OF PROGRESS, TWEETS, OFF-HAND COMMENTS, INFLUENCERS, AND VIDEO POSTS DRIVE THE MARKET – IS IT DIFFERENT THIS TIME?
- The chart below shows aggregate enterprise value to next twelve months consensus EBITDA projections for the six of the top MSOs, including Curaleaf, GTI, Trulieve, Verano, Cresco, and TerrAscend from 8/25/23 (the Friday before the HHS announcement) through 9/29/25 (The Trump Video Post)
- THE ADVENT OF REVOLVING CREDIT AGREEMENTS IS A MAJOR ADVANCE IN CANNABIS CREDIT
![]()
- Note that immediately after the last “real news” in October 2024, when the DEA ostensibly followed the HHS recommendation (we should know better by now), the market entered a funk and slid into a nearly linear downward glide path for almost 11 months. The slide was punctuated on the downside by the election, which dealt two surprises: Trump and the Florida debacle. We view the latter as the more serious and damaging of the two, primarily because companies had put a lot of eggs in that basket, but also because the market had truly come to believe it was a foregone conclusion.
- Valuation multiples plummeted to levels worse than before S3 was even a glimmer in HHS’s eyes. Then, in July 2025, tweets by influencers supposedly close to the White House began to appear, claiming that real progress was being made behind closed doors. Multiples soared by about 79% from the end of June through August 11, when we received Trump’s announcement that a rescheduling decision would be made in “several weeks.”
- The announcement predictably gave the market still more push, albeit without any clear direction. Over the next three weeks of silence, the market drifted lower, and multiples declined from 7.0x to 5.8x, only to be cranked up again by the video.
- So does the video really signal that S3 is a done deal and about to be announced? We have always thought that S3 was a political plaything for Trump, to be trotted out when it best served him, but with no driving motivation in either direction. The video doesn’t change our minds about that. What sort of serious political messaging gets done by releasing somebody else’s video on social media? If we just wanted good political theater, we would bring the comedy team of Schumer & Booker back to the stage.
- No one seems to be addressing the fact that this video has virtually nothing to do with cannabis per se, directed as it is toward Farm Bill-compliant CBD. There is a passing reference to medical cannabis, but all of the big stuff is about government reimbursement for elderly CBD. Undoubtedly, that is like the first spring flowers after a multi-year winter for the CBD industry, but what about cannabis? And in particular, what about the biggest part of the cannabis business, the adult rec market? Crickets.
- We still think Trump is likely to follow through on his campaign statements in support of S3, but this video doesn’t give us quite the warm-and-fuzzy feeling we crave. Obviously, the market vigorously disagreed for at least a day.
- On the positive side, even after the pump from the “several weeks” quote and the video, the market is still trading at multiples that are lower than for most of cannabis history, except for late 2024 and the first half of 2025. That gives us some comfort that significant upside still exists if something more real than a tweet were to occur, while also limiting the downside from more time drift and investor skepticism.
- What would it take for us to become enthusiastic? How about a less-than-hostile mention of adult-use cannabis, which makes up around 70% of retail sales? Sharp legal minds like Dentons have assured us (via their recent write-up) that S3 would by necessity apply to all cannabis, as it is the substance that is classified, not the application. But we can’t help but think Trump would leave adult rec behind if he could.
- For a longer-term reference on cannabis multiples and their reactions to various legislative/regulatory announcements, please see the graph below.
![]()
- WE UPDATED OUR NEW YORK ANALYSIS IN THE JOINT VIRIDIAN/HOODIE WEBINAR ON SEPTEMBER 30. REACH OUT TO US IF YOU MISSED IT. WE HAVE BECOME MORE CONFIDENT ABOUT THE NY GROWTH PROSPECTS.
- The courts, for the time being, have set aside the specter of proximity-gate, and with strong results now in for August, we are feeling pretty good about our $1.6 billion forecast for 2025 retail sales. In fact, if anything, we might be a little too conservative. However, one of the conclusions of our report —that substantial excess cultivation capacity exists —is coming home to roost. Price compression is clearly happening in NY, and we expect it to continue.
- Of course, the perennial challenges to the social equity framework continue. An August 12th ruling from the Second Circuit in the case of Variscite NY Four LLC v New York State Cannabis Control Board, which held that under the Dormant Commerce Clause, New York’s licensing preferences tied to New York-specific marijuana convictions are discriminatory and cannot stand. So now license applications that were given preference on those grounds may come under challenge.
- Why is this critical? Our retail sales analysis rested heavily on the continued rapid pace of dispensary rollouts. We still feel comfortable with our estimates of $1.6 billion in total retail sales for 2025 and $2.5 billion for 2026, which should place NY in the top five. But to reach the $4B we have projected for 2030, we need a dispensary count of over 1,000, implying a CAGR of over 10%. Many things could still go wrong, warranting close monitoring of the data.
- To be cynical, no one has ever gone wrong by underestimating the speed of the NY rollout; however, we are confident that the state will crack the top five in 2026.
![]()
-
- STATES CONTINUE TO CRACK DOWN ON HEMP IN ADVANCE OF THE FARM BILL, WITH LITTLE REGARD TO THE LAW OF UNINTENDED CONSEQUENCES
- Last week, we presented a table with the latest attempts by six different states to rein in intoxicating hemp.
- However, states must be cautious not to push consumption back into the illicit market.
- The most recent move in California is the passage of AB 8, moving intoxicating hemp products into California’s licensing and regulatory system (because that is working so well)
- Did anyone consider why consumers were buying these products in the first place, given that they had ample opportunity to purchase products from dispensaries?
- As we see it, there are two main drivers:
1) convenience. Pre-ban estimates suggest that as many as 10,000 convenience stores and smoke shops sold intoxicating hemp products, or nearly 3x as many as licensed dispensaries. Many intoxicating hemp products were also available by mail order. Many Californians live in cannabis deserts due to the municipal retail opt-outs. For many, the convenience of hemp was the driving factor.
2) Price. Similar to the illicit market, hemp products are close substitutes that are available at substantial discounts to dispensary products. The illicit market shows that many consumers do not consider the perceived greater safety of testing and regulatory oversight to be worth the cost. They prefer cheaper
- Put these two factors together, and you will get a picture of what is likely to happen: more illicit market sales. A significant part of hemp sales will not be transported to the dispensaries but will go back into the illicit channel.
- Regulators often fail to anticipate the unintended consequences of their actions.
- STATES CONTINUE TO CRACK DOWN ON HEMP IN ADVANCE OF THE FARM BILL, WITH LITTLE REGARD TO THE LAW OF UNINTENDED CONSEQUENCES
- RESCHEDULING, DESCHEDULING, LEGALIZATION…
- The potential impacts on the industry seem difficult to wrap your head around, so we wanted to go back to the fundamentals of the market structure of cannabis using Michael Porter’s Five Forces Framework, which decomposes the forces operating on an industry into the threat of substitutes, the threat of new entrants, the power of suppliers, and the power of buyers, which together influence the internal rivalry and degree of price competition in the industry.
- . At the bottom of each force discussion, we comment on how legalization will impact the force and whether that impact will foster consolidation.
- We believe eventual legalization will spur tremendous consolidation, which in turn will benefit the industry in several ways: greater price stability through an oligopolistic structure, centralization of production and distribution, both of which improve margins and capital efficiency, thereby increasing ROIC.
- The news isn’t all good, though. If descheduling were to occur without a phase-in period and without attempting to buffer the impact on limited-license state programs, it would be highly disruptive, likely causing a wave of facility shutdowns, write-offs, and layoffs. The retail side of the business would fare better, at least initially, as we would expect DTC sales to take some time to develop. This is why we believe the most important thing we’ve learned in our years in cannabis is “be careful what you wish for.”
![]()
-
Threat of Substitutes – Very High
- Illicit Market: The largest and most immediate substitute. Illicit operators face no compliance costs or taxes, and can undercut legal players by 30–50%. Consumer willingness to buy illicit products remains high due to habit, price sensitivity, and, in some cases, quality parity.
- Hemp-Derived Intoxicants: Delta-8, delta-10, and THCA products (often sold in gas stations and online) are federally legal under the 2018 Farm Bill loophole and are virtually perfect substitutes for cannabis flower and vapes. They carry lower production costs and fewer regulatory burdens.
- Other Recreational Substitutes: Alcohol, tobacco, and in some cases, prescription or over-the-counter pharmaceuticals compete for the same consumer leisure or therapeutic dollar.
- Overall: This force is very strong — substitutes put a permanent cap on pricing power and force legal operators into constant margin pressure.
- Consolidation Drivers/Impact: Larger, more efficient operators are better able to withstand ongoing margin compression. Descheduling/ legalization will accelerate the drive for roll-ups and M&A to create centralized production, distribution, and marketing. Consolidation will eventually foster a more oligopolistic price structure.
Threat of New Entrants – Moderate to High (State Dependent)
- Licensing Barriers: In theory, cannabis markets are shielded by limited-license regimes. In practice, states often over-issue cultivation or retail licenses (e.g., Michigan, Oklahoma), creating oversupply.
- Capital Requirements: Building an indoor facility costs ~$400 per sq. ft., with millions in capital expenditures required for cultivation
- Regulatory Hurdles: State-by-state rules slow new entry, but hemp-derived entrants largely circumvent these barriers.
- Overall, entry into licensed markets is difficult, but trivial in the case of hemp-derived substitutes or illicit channels — making this force stronger than it appears on paper.
- Consolidation Drivers/Impact: After descheduling/legalization barriers fall and tobacco, alcohol, CPG, and Pharma will enter, shifting power to larger players with capital and nationwide distribution networks. Existing MSOs merge to bulk up before these new players enter the market.
Bargaining Power of Suppliers – Moderate
- Labor: Fragmented, largely non-unionized, and specialized but replaceable. Worker turnover is high, limiting labor’s bargaining power.
- Nutrients/Fertilizer/Equipment: These suppliers are dependent on cannabis demand and lack leverage to raise prices materially.
- Energy: The most powerful supplier. Indoor grows are energy-intensive, and cannabis operators are price takers on local utility rates. Energy cost spikes directly erode margins.
- Real Estate/Financing: With federal illegality, cannabis companies often rely on expensive sale-leaseback arrangements (e.g., IIPR) and high-cost debt, which gives landlords and lenders unusual power relative to other industries.
- Overall: Supplier power is moderate — weak in raw materials, strong in energy and capital access.
- Consolidation Drivers/Impact: Descheduling/legalization reduces the cost of capital through uplistings and improved access to banking & securities. Larger firms tend to benefit differentially from these moves.
Bargaining Power of Buyers – Increasing
- Retail Consumers: Currently, brand loyalty is weak. Consumers shop on price and THC percentage, and can easily shift to cheaper or illicit alternatives.
- Retail Dispensaries: In vertically integrated markets, MSOs control both cultivation and retail operations, thereby reducing the leverage of buyers. But in wholesale-heavy states, retailers wield power by choosing which brands to stock.
- Large Buyers: Emerging multi-state wholesale distributors (and in hemp-derived products, convenience store chains) exert significant bargaining power by controlling shelf space.
- Overall, buyer power is rising, particularly as oversupply provides consumers and retailers with numerous
- Consolidation Drivers/Impact: Descheduling/ legalization will eventually allow national retailers to sell cannabis products. Only firms with nationwide logistics will effectively supply those buyers, reinforcing consolidation pressure.
Industry Rivalry – Very High
- Commodity Dynamics: Cannabis is a perishable commodity, as unsold flower loses potency, aroma, and value. Like airline seats, the incentive is always to sell inventory even if it requires price discounting.
- Fixed/Semi-Fixed Costs: Cultivation has high fixed and semi-fixed costs (especially energy for lights and HVAC). Once a grow is built, operators tend to run it at full capacity, covering variable expenses and making a contribution towards fixed costs. This tends to intensify oversupply.
- Exit Barriers: Specialized grow facilities and equipment have little alternative use value, locking weak competitors into the market and prolonging price wars and supply gluts.
- Growth and Profitability: Industry growth is slowing. 2025 is expected to show declining revenues among top MSOs, with only ~7% rebound projected in 2026 as Pennsylvania, Virginia, or Florida flip to adult use. Long-term growth is CPG-like (population + inflation) as the illicit market is converted to a legal one.
- Overall, the rivalry is cutthroat due to oversupply, price compression, and trapped capacity.
- Consolidation Drivers/Impact: Rivalry becomes national, not state-by-state. The weakest competitors exit, mid-tier firms merge for survival, and scale advantages (lower cost per pound, better logistics, stronger brands) drive consolidation
Summary: What the Five Forces Tell Us About Consolidation
- Before legalization: Fragmented, state-by-state markets with regulatory moats allow lots of small to mid-tier operators
- After descheduling: Forces flip toward economies of scale, lower capital costs, and national distribution requirements
- Result: The Five Forces indicate heavy consolidation, with parallels in the beer, tobacco, and pharmaceutical industries.
- Thousands of small local operators with a handful of dominant national and international firms. Survivors include some of today’s largest MSOs with strategic entrants from alcohol, CPG, and tobacco.
- THE 2026 DEBT MATURITY WAVE IS LOOKING LESS THREATENING: THE COMPANIES WITH THE RISKIEST REFINANCINGS HAVE ALREADY GONE INTO RECEIVERSHIP, NEGOTIATED AN EXTENSION, OR ARE PREPARING AN ARTICLE 9 SALE OF ASSETS TO CREDITORS
- Much has been made of the upcoming wave of cannabis debt maturities in 2026. The sheer size is undoubtedly intimidating. The companies pictured on the graph below collectively have approximately $1.5 billion of debt maturing in 2026. (IAnthus maturities are actually in 6/27, but close enough!). This figure used to be over $2.3 billion, before Gold Flora, 4Front, and AYR flamed out, and we recently added Shwazze to the list. Cresco agreed to refinance its secured term loan. Putting that figure into perspective, $1.53B is still more than total cultivation & retail sector capital raises for any year since 2019, except for 2021 ($4,8B) and 2022 ($1.7B)
- Viridian is more constructive about the issue than most other industry observers. We observe that in the high-yield bond market, it is virtually never the case that debt is paid off in cash. It is generally refinanced, OR the company is forced to restructure. Obviously, given the lack of prepackaged bankruptcy (or any bankruptcy, for that matter), restructuring is rightfully a prospect to be feared in the Cannabis Industry.
- So, how do we gauge the risk of something going wrong in 2026? Refinancing risk is a peculiar mixture of market psychology and financial realities.
- The green bars show the 10/3/25 market-implied asset coverage of total liabilities. We arrive at this by viewing the equity as a call option on the asset value of the firm, with a strike price equal to its liabilities, and assuming maturities of 2026, as well as a 30% volatility and a risk-free rate of 4.25%. This provides us with all the elements of the Black-Scholes option pricing formula except for the current asset value. By iterating on the solution of the BS model, we can find the market’s assumption for asset value. The importance of this data point should be obvious. For companies with less than 1x asset coverage of liabilities, debt providers are effectively making an equity bet. They do not have adequate asset value coverage to fall back upon.
- The red line represents the Viridian Capital credit ranking, which considers four key credit factors: Liquidity, Leverage, Profitability, and Size. Refinancing will be more difficult for weaker credits (higher numbers). Companies with ranks of under 12 are in the top half of the Viridian-ranked universe of credits.
- The black dots represent the multiples of market cap that the 2026 debt maturities represent. Clearly, the larger the debt maturities relative to the market cap, the more difficult we would expect refinancing to be.
- The companies to the right of Jushi (JUSHF: OTCQX) on this graph represent higher risk. They have less than 1x asset value coverage, generally poor Viridian Credit Ranks, and several, such as IAnthus and Body & Mind, have maturing debt that is a multiple of their market capitalization. Companies in this position represent only about $170M of the maturing debt.
- Conversely, the companies on the left-hand side of the graph represent lower refinancing risk. They have solid asset coverage, strong Viridian Credit Ranks, and maturing debt that is less than 1.25x times their market capitalization. These companies represent $1.36 billion of the $1.53 billion total (70%), and we believe they should all be able to refinance their maturities without undue hardship. The offered side quotes on Curaleaf, Verano, and Trulieve, the solid credit MSOS with the most remaining 2026 maturities, all tightened significantly since the Cresco deal. Investor psychology appears to have shifted, with a growing belief that these names will be able to refinance their maturities without undue hardship.
- But market psychology can be fragile. There is always the possibility of an “accident” where negative psychology meets illiquid funding markets, and refinancings that appear favorable on paper fail to materialize. An adverse ruling on rescheduling could raise the threat level back to previous peaks.
![]()
-
- HOW MUCH OF THOSE 280E SAVINGS WILL FIRMS ACTUALLY BE ABLE TO KEEP?
- Every analysis we have seen regarding the impact on credit capacity, valuation, or growth from the elimination of 280E begins with a simple yet probably incorrect assumption: that the tax savings from eliminating 280E will flow directly to free cash flow.
- This is clearly not the case, however. Some part of the savings, and possibly the majority of it, is likely to be passed through to consumers via lower prices.
- One historical example that demonstrates this point is the 2022 California elimination of the cultivation tax. The $161 per pound tax equated to roughly 10-15% of the value when pounds were clearing at $1,000-$1,500 per pound. However, according to figures from MJBIZ, California production revenue dropped nearly 27% in 2022, more than offsetting the benefit from the excise tax. The hoped-for shot in the arm for California cultivators never happened.
- The ability to retain tax savings varies significantly across markets. Operators in markets with many competitors and a vibrant illicit market are likely to keep very little of the savings. On the other hand, lower prices may allow a recapture of sales from the illegal market. The net impact can be approximated based on two variables: the number of competitors in the market and the price elasticity of demand.
- The Cournot model is a simple economic model that describes competition when firms sell a relatively undifferentiated product. Each firm chooses the quantity it will produce, assuming rivals’ outputs are fixed, knowing that the market price will be whatever clears the market at the total amount produced. The strategic “best responses” to one another’s quantities define a Nash equilibrium: with few firms, each has market power (the price stays above the cost), and as the number of firms N increases, the outcome drifts toward perfect competition. In the symmetric, linear-demand case, it gives clean rules of thumb: firms will retain only 1/(N+1) of the tax savings, where N is the number of competitors in the market.
- There is another side to the story, though. The other variable in our analysis, the price elasticity, determines the degree to which firms will recapture sales from the illicit market. If the elasticity is below 1, sales react weakly to a change in price, whereas if e > 1, sales react strongly to a price change. We have modeled e at 0.6, 0.9, and 1.2 to correspond with our assumptions of N = 5, 10, and 50. Each set of beliefs about N and e defines a type of market.
- The chart below shows that in oligopolistic markets with low numbers of competitors and low price elasticity, the net impact will be that firms will retain around 22% of the tax savings. Around 17% of that comes from tax savings not passed through to consumers, while the other 5% comes from the extra FCF from incremental sales from the illegal market. The situation is reversed in markets with many competitors and high elasticity: only 2% of tax savings is retained, but incremental FCF equivalent to 12% of the tax savings comes from recaptured illegal sales.
- The bottom line is that modeling based on reasonable assumptions suggests that, while S3 remains quite positive for the industry, the impacts are likely overstated.
- HOW MUCH OF THOSE 280E SAVINGS WILL FIRMS ACTUALLY BE ABLE TO KEEP?
![]()
-
- MEASURES OF MSO VALUATION, LEVERAGE, AND LIQUIDITY
- The first two graphs present different versions of EV/EBITDA on the vertical axis and Debt/EBITDA on the horizontal axis.
- The first graph presents our latest view of the most appropriate valuation and financial statement-based leverage metrics: Adjusted Enterprise Value (EV) / 2026 EBITDAR and Adjusted Net Debt / 2026 EBITDAR. In calculating Adjusted Net Debt, we make several key assumptions: 1) Leases that are included on the balance sheet are considered debt. We view most leases in the cannabis space as equivalents to equipment loans or mortgage loans. While it is true that a lease default does not necessarily trigger a cascade of events leading to bankruptcy, the distinction is often meaningless in the Cannabis Industry due to the mission-critical nature of many long-term leases and the absence of bankruptcy protection in This Sector. 2) We consider any accrued taxes (including uncertain tax liability accounts listed as long-term liabilities) in excess of the most recent quarterly tax expense to be debt. Our calculation of enterprise value is now market cap plus debt plus leases plus tax debt minus cash. We now use EBITDAR rather than EBITDA, as lease expense is deducted prior to calculating EBITDA.
- Our adoption of new metrics tends to make the companies look less cheap and more leveraged.
- The second graph utilizes EBITDA and employs traditional calculations for both debt and enterprise values, excluding leases and taxes.
- Surprisingly, seven of the companies on the enhanced metric chart are still above 3x leverage, which we have identified as the boundary of sustainability in a 280e environment. Five companies now exceed 4x leverage, which we believe will be close to the maximum sustainable post-280E.
- Glass House is a valuation outlier. We have been positive on Glass House for quite a while, but the multiple spread to the nearest competitor is straining our resolve. We note GLASF’s $25M at the market equity issuance facility as another factor likely to restrain price appreciation. Finally, the sharp downward revisions to revenues and EBITDA for the second half of 2025 are concerning, as they certainly do not align with the picture the company originally painted.
- MEASURES OF MSO VALUATION, LEVERAGE, AND LIQUIDITY
![]()
![]()
- The third graph examines leverage through the lens of total liabilities to market capitalization. We believe this is the single best measure of leverage because it is a direct reflection of the market’s assessment of a company’s assets in excess of its liabilities, and it is sensitive to changes in the market’s perception of a company’s future prospects.
- On the bottom left are companies with an Adj. EV/2025 EBITDAR ratio of under 7x and total liabilities to market cap of under 2x. The group includes GTI, Trulieve, Verano, Cresco, and Vext. Companies in this quadrant are right to consider stock repurchases or using cash in acquisitions. They can afford some additional debt and can take advantage of the ongoing dislocation in equity prices.
- Between 2x and 5x total liabilities to market cap, we find MariMed and Jushi. Jushi is interesting because its liabilities include nearly equal amounts of 2026 maturities of debt and uncertain tax liabilities.
- On the right lies Cannabist and Ascend, both between 6x and 10x, a range that signals stress if not distress. Ascend appears to be looking much better than it did three weeks ago, thanks to the rally in its stock.
- AYR, 4Front, and Schwazze are now off the chart to the right, befitting their distressed status. Our recent work, which utilized option modeling of equity prices, showed that the market believes each of these companies has significantly less asset value than its liabilities.
![]()
- The fourth graph introduces the free cash flow adjusted current ratio liquidity measure into the mix. Note that we have recently modified our treatment of this ratio by removing uncertain tax liabilities from current liabilities, where they were previously placed. The result is that only Jushi and Cannabist are significantly below 1x free cash flow adjusted current ratio.
- On the top left, we find companies with adequate liquidity and low market leverage, including GTI, Trulieve, and Glass House. Companies in the lower right generally have constrained liquidity and high leverage, a potentially dangerous combination in a capital-constrained environment. The only company presently on the chart with these symptoms is Cannabist.
![]()
- CANNABIS STOCK VOLUME DRIFTS LOWER AS S3 WAIT DRAGS ON
- The average daily dollar volume of $23 million for the week ending October 17th is down from the 52-week peak of $68 million and from last week’s $ 24 million. The current Days to Trade the Market Cap (DTTMC) of 627 remains significantly better than the 52-week average of 1201. A DTTMC of 627 means that an investor who acquired a 5% position in the stock, assuming they wanted to be less than 25% of the average daily dollar volume, would require 125 days to trade out of their position.
![]()
- GIVING CREDIT WHERE CREDIT IS DUE
- The chart below displays our updated credit rankings for 25 U.S. cannabis companies as of October 17, 2025. We have reduced our ranking set from 30 to 25 companies by eliminating AYR, CLS, FFNT, GRAM, and SHWZ. Each of these companies is either in receivership (FFNT, GRAM), in restructuring discussions (AYR & SHWZ), or has out-of-date financials (AYR, FFNT, GRAM, SHWZ). Additionally, each trade is significantly out of the money, with stock price movements driven primarily by volatility rather than valuation.
- The blue squares show the offered-side trading yields for each Company.
- We find it curious that Cresco, TerrAscend, and Ascend are all trading within 25bps of each other. All three have refinanced their 2026 maturities, but we rank Cresco four notches better than Ascend and five notches better than TerrAscend.
![]()
Best and Worst Stock Performers
- TRADING RETURNS FOR PUBLIC COMPANIES BY CATEGORY
-
- Plant-touching categories are still negative for the LTM period, but have clearly reduced their losses YTD.
![]()
Best and Worst Performers for the week:
![]()
![]()