A Federal Cannabis Tax Could Redraw the State Price Map
Photo by TechPhotoGal via Pixabay.
As of October 2026, there is still no federal excise tax on cannabis. There's also no signed comprehensive federal cannabis law, despite a claim circulating from a source called CannIntel that one was enacted on July 2, 2026 — a claim both the Marijuana Policy Project and Marijuana Herald have specifically disputed in their reporting. What does exist is a traffic jam: at least 22 cannabis-related bills currently sitting in Congress, several of which borrow their tax architecture almost line-for-line from the federal tobacco excise tax.
That borrowing is the actual story here, more than the question of whether any of these bills becomes law. Everyone arguing about cannabis legalization tends to treat federal legality as the finish line. It isn't. The design choice buried inside the tax section — a flat percentage of the sale price, a fixed amount per milligram of THC, or no federal tax at all — will do more to reshuffle which states can sell cannabis cheaply than whether the plant is rescheduled or descheduled. States already run wildly different effective tax burdens: Michigan just added a 24% wholesale layer on top of its existing retail excise, while Oklahoma's medical market sits around 11.5% combined. A federal tax doesn't land evenly on that patchwork. It stacks unevenly on top of it, and depending on which bill's structure wins out, it could either flatten state-by-state competitiveness or make the gaps far more dramatic. This piece works through the actual bill text — the Cannabis Administration and Opportunity Act (CAOA), the MORE Act, the STATES Act, and STATES 2.0 — plus the Yale Budget Lab's potency-tax modeling, to reason out who ends up winning and losing shelf-price competitiveness over the next decade.
The Tobacco Tax Template Congress Keeps Returning To
Federal tobacco taxation runs on a hybrid model that most cannabis-tax drafters are now consciously imitating: a fixed per-unit federal excise layered on top of state excise taxes that range from under $0.20 a pack in low-tax states to more than $4.50 a pack in the highest-tax states. The federal number is flat and specific. The state number is all over the map. That combination is exactly what CAOA, reintroduced July 16, 2026 by Senators Cory Booker, Chuck Schumer, and Ron Wyden, is built to replicate for cannabis.
CAOA's tax structure starts at 10% of the removal sales price in year one, stepping up five percentage points annually until it hits 25% in year four. At that point, rather than continuing as a percentage, the tax converts to a hybrid system: a per-ounce rate for flower and a per-milligram-THC rate for extracts and concentrates. That pivot from a percentage-of-price (ad valorem) tax to a fixed, potency-based tax isn't an arbitrary design flourish. It mirrors what Congress has done in past federal excise overhauls of tobacco taxation, when lawmakers moved toward fixed per-unit rates specifically because ad valorem taxes erode as prices fall.
That erosion problem is already cannabis's problem, not a hypothetical one. Legal cannabis prices have been falling for roughly a decade across nearly every mature state market, driven by cultivation efficiency gains, oversupply in markets like Oregon and Michigan, and basic competitive pressure as more licensed operators enter. A 10% tax on a falling price base collects less every year in real terms, which is exactly the trap federal tobacco tax writers were trying to avoid as cigarette prices stagnated relative to inflation in earlier decades. CAOA's year-four pivot to a per-ounce and per-milligram structure is best read as a hedge against that same price-collapse dynamic — a bet that taxing a stable physical unit (an ounce of flower, a milligram of THC) protects revenue in a way that taxing a shrinking dollar figure cannot. Whether that hedge actually works depends on how Congress indexes those per-unit rates going forward, since an un-indexed fixed tax has its own long-run erosion problem — inflation, not price competition, eats it instead.
Three Bills, Three Very Different Tax Futures
Photo by nmnm via Pixabay.
Lump these bills together as generic legalization and you miss that they encode three genuinely different fiscal futures. CAOA is the most tobacco-like and the most aggressive long-run revenue design of the bunch: 10% rising to 25% over four years, then converting to that hybrid per-ounce/per-milligram-THC structure described above. It's built by senators who are thinking in decades, not sessions.
The MORE Act, carried by Rep. Jerry Nadler, takes a visibly different approach. It starts at just 5% — a fraction of CAOA's eventual 25% — and stays closer to a flat, modest rate rather than ramping aggressively. That's not an oversight; it reflects a priority ordering that puts equity provisions and expungement infrastructure ahead of revenue maximization. A lower tax rate means lower retail prices, which matters directly to the legal market's ability to compete with illicit sellers, a point we'll come back to.
The STATES Act goes further in the other direction: no federal excise tax at all. It essentially codifies the current state-by-state patchwork as a permanent feature rather than a transitional one, leaving price competitiveness exactly where it sits today — determined entirely by each state's own excise structure, with no federal layer to either compress or widen the gaps between them.
STATES 2.0 (H.R. 2934) pairs a federal excise tax with public health and safety standards, which suggests a regulatory-first model: tax follows compliance infrastructure rather than preceding it. That sequencing matters for timeline forecasting — a bill that conditions tax collection on built-out safety and testing regimes will likely move slower in practice than one that just flips a tax rate on at signing.
None of these four bills has passed either chamber of Congress as of October 2026. That's worth stating plainly given how much noise is circulating. The widely shared claim that a comprehensive federal cannabis law was signed on July 2, 2026 is contradicted by reporting from both the Marijuana Policy Project and Marijuana Herald, and should be treated as unverified until a primary legislative record confirms it. Nothing in this piece assumes that claim is true.
What the Yale Budget Lab Numbers Actually Say
Under the Yale Budget Lab model, federal cannabis tax revenue over 10 years nearly doubles—from $58 billion at current state legalization levels to $111.3 billion if all remaining states legalize.
The Yale Budget Lab has run one of the more detailed public models of what a potency-based federal cannabis tax would actually generate, and the numbers are specific enough to be useful even though they're explicitly a projection, not a locked-in outcome. Their model applies a tax of $0.00625 per milligram of THC. Translated into retail terms, that works out to roughly $1.31 in tax on a gram of marijuana priced at the current average of $8.59 per gram — about a 15% increase in the tax-inclusive price a consumer actually pays at the counter.
Scaled to federal revenue, Yale projects nearly $58 billion collected over ten years under current state legalization patterns — meaning the states that have already legalized adult-use or medical cannabis, as they exist today. If every remaining state legalized during that window, the ten-year figure rises to $111.3 billion. That gap is the single most important number in the model: roughly half of the larger projected revenue figure depends entirely on future state-level legalization that hasn't happened yet and isn't guaranteed to happen on any particular timeline. Treat the $111.3 billion figure as conditional, not as a baseline expectation.
It's worth being explicit about what this model is and isn't. It's a projection built on stated assumptions — a specific potency-tax rate, a specific baseline of which states are counted as legalized, and an assumed consumption volume that doesn't account for behavioral shifts the tax itself might cause. Actual federal revenue, if any version of this tax becomes law, will move with whatever final bill text, effective date, and phase-in schedule Congress actually settles on. CAOA's four-year ramp, for instance, would generate meaningfully less revenue in its early years than a tax that hit 25% immediately.
One structural consequence is worth flagging because it's a genuinely testable prediction rather than vague speculation: a potency-based tax mathematically hits high-THC concentrates and vapes harder in dollar terms than it hits flower, since those products pack more milligrams of THC into a smaller unit. If retailers pass that cost difference through to shelf prices — and tobacco and alcohol pricing history suggests they generally do — the reasoned expectation is a consumer shift toward flower and lower-potency products, since the tax burden per unit of enjoyment becomes comparatively cheaper there. Current research doesn't yet confirm this shift has happened anywhere, because no jurisdiction has run a potency tax at this scale long enough to measure it. It's a mechanism-based forecast, not an observed trend.
The Patchwork a Federal Tax Would Land On
Whatever federal structure eventually passes, it lands on a state patchwork that's already extremely uneven, and getting more uneven by the month. California's cannabis excise tax actually hit 19% in mid-2025 before Governor Newsom signed legislation on September 22, 2025 resetting it back down to 15% through July 2028 — a direct response to industry complaints that the higher rate was pushing consumers toward unlicensed sellers. Stack state excise on top of local cannabis taxes, which vary city by city, and effective combined rates in California run anywhere from 27% to 38% depending on jurisdiction.
Colorado runs a different kind of double layer: a 15% retail excise tax paid at the point of sale, plus a separate 15% wholesale excise tax levied earlier in the chain, at the first transfer from cultivator to processor or retailer. That's effectively two 15% taxes hitting the same product at two different supply chain points, which complicates any simple price comparison with states that only tax retail.
Michigan is the most dramatic recent mover. Starting January 1, 2026, the state added a 24% wholesale tax on producer-to-dispensary sales, stacked on top of its existing 10% retail excise. This is a serious reversal for a market that had built its national reputation on some of the lowest ounce prices anywhere in legal cannabis — Michigan growers flooded the market for years, and consumers benefited from the resulting price collapse. The new wholesale tax is already drawing industry litigation, with operators arguing the levy was imposed too abruptly for a market built around thin margins.
Oklahoma sits at the opposite end entirely. Its medical-only market carries a 7% excise tax plus the state's 4.5% general sales tax, landing around 11.5% combined — light by comparison to nearly every adult-use state on this list.
The important structural point: a flat or percentage-based federal tax doesn't flatten this patchwork, it compounds it. The same federal rate represents a much bigger relative price jump in a low-tax state like Oklahoma, where it's being added to a small existing base, than in an already high-tax market like California, where consumers and retailers alike are numb to effective rates near a third of the shelf price.
Who Gets Squeezed, Who Gets Room to Compete
States that have marketed themselves on low effective tax burdens — Michigan on historically rock-bottom prices, Oklahoma on its lighter 11.5% combined rate — stand to lose that competitive edge fastest once a federal layer gets added, for a simple arithmetic reason: the percentage gap between them and high-tax states like California narrows as a flat federal rate gets added to both sides. If Oklahoma's 11.5% and California's higher combined rate both get an additional federal rate stacked on, Oklahoma's relative price advantage shrinks proportionally more than California's, even though the dollar amount added is identical in percentage terms.
California, somewhat counterintuitively, may have less to lose in relative terms. Consumers there are already acclimated to effective combined rates in the high-20s to high-30s percent range, so an added federal layer changes the competitive math less dramatically — though it does risk pushing California toward having the highest absolute prices in the country, which carries its own demand risk independent of tax share.
Tobacco's own multi-decade history offers a useful, if imperfect, precedent here. States with the lowest cigarette excise taxes — concentrated in parts of the South, averaging under $0.60 per pack — still maintain meaningful, durable price advantages over states with the highest state cigarette taxes, where the state tax alone can exceed $4.35 per pack, even after decades of a stable federal per-pack tax sitting on top of both. That persistence suggests state-level rate differences matter more than the federal layer over the long run, since the federal tax is the same everywhere and doesn't erase the state-level gap — it just adds a constant to both sides of an unequal equation.
The meaningful counter-case is CAOA's own structure. If its year-four hybrid per-ounce/per-milligram-THC model becomes law, it could compress some of these gaps, because a fixed per-unit tax hits a cheap ounce of Michigan flower and an expensive ounce of California flower in nearly identical dollar terms — which means it burdens the low-price market proportionally harder. That's a genuinely different outcome than a percentage tax produces, and it's the clearest reason cultivators and investors should be tracking CAOA's specific unit-tax design rather than treating all four bills as interchangeable.
A further-out scenario worth flagging as speculative: if interstate cannabis commerce eventually becomes legal, plausible under a STATES 2.0-style framework, cultivation could migrate toward whichever states combine the lowest total tax burden with the lowest production costs — echoing how tobacco growing concentrated historically in a handful of states with favorable growing conditions and cost structures. This pattern is worth watching, but interstate cannabis commerce isn't legal today, so it remains a forward-looking hypothesis, not a trend already underway.
The Illicit Market Risk Nobody's Pricing In
Photo by MART PRODUCTION via Pexels.
Tobacco's tax history comes with a documented side effect that cannabis tax drafters don't seem to be pricing in carefully enough: every major federal or state tobacco tax increase has produced measurable cross-border and illicit trade effects. States that carry the nation's highest cigarette taxes consistently show up in multiple Tax Foundation studies as having some of the highest estimated rates of untaxed cigarette consumption in the country — smuggled, cross-border, or otherwise off the books.
Cannabis enters this dynamic with a disadvantage tobacco never had during its decades of tax escalation: a large, already-entrenched illicit market operating at scale well before legalization even began. California in particular has a longstanding unlicensed cannabis market that predates legal sales by decades and currently undercuts licensed retail prices by avoiding all taxes — not just a federal one, but the state and local layers too.
Stack a new federal excise tax on top of California's existing 27-38% effective combined state-and-local rate, and the risk is that total tax burden crosses a threshold wide enough to meaningfully widen the price gap between legal and illicit sellers. California's own cannabis industry has already cited exactly this dynamic — tax burden pushing consumers toward unlicensed sellers — as a driver of the state's unexpectedly slow legal-market growth, which is part of why Newsom's 2025 rollback to 15% happened in the first place.
There's a genuinely open question here that current data doesn't settle, and it's worth being honest that this is reasoned speculation rather than established fact: does a potency-based tax, which hits high-THC concentrates and vapes harder, nudge consumers toward lower-potency legal flower as described earlier — or does it instead push them toward untaxed illicit concentrate, where no tax of any kind applies regardless of potency? Both outcomes are mechanically plausible, and no jurisdiction has run this experiment at sufficient scale and duration to show which force wins.
This uncertainty argues for watching the final bill's effective date and phase-in schedule as closely as its headline rate. CAOA's four-year ramp to 25% gives the legal market time to adjust pricing, build margin cushion, and compete, which is a meaningfully different risk profile than a single abrupt rate change landing all at once — the kind of shock that historically tends to produce the sharpest illicit-market responses in tobacco's own tax record.
The headline question everyone asks — will Congress legalize cannabis federally — is less consequential for state competitiveness than a question almost nobody outside tax policy circles is asking: which of these four tax architectures ends up in the bill that actually passes. An ad valorem percentage tax, a potency-based per-milligram tax, and no federal tax at all produce three different maps of which states get cheaper and which get more expensive, independent of whether legalization itself happens. Cultivators, investors, and state regulators would do better tracking CAOA's year-four hybrid per-ounce/per-milligram-THC pivot specifically than waiting for a generic legalize-or-not headline.
Michigan's 2026 wholesale tax hike is worth treating as a preview, even though it's a state-level change rather than a federal one. It shows, in miniature and in real time, how quickly a price advantage can erode once a new tax layer stacks onto an existing one — and the litigation already underway there is a useful early signal of the kind of industry pushback pattern that will almost certainly repeat at the federal level if CAOA or something like it ever reaches a floor vote.
None of this is settled, and it shouldn't be treated as if it were. Every revenue figure and price projection in this piece — Yale's $58 billion and $111.3 billion ten-year estimates included — is a model built on stated assumptions about tax rates, legalization timelines, and consumption patterns that haven't happened yet. The honest position is to watch which of these four bills, if any, gets a floor vote, and to read the fine print of its tax section before drawing conclusions about who wins.
