Descheduling vs. Rescheduling: Where Cannabis Law Lands by 2035

August 25, 2026

Photo by Syed F Hashemi via Unsplash.

Pull up the Federal Register entry from April 28, 2026, and you'll find something the Controlled Substances Act was never designed to produce: cannabis with two different federal identities at the same time. Buy a Schedule III product through an FDA-approved pathway or a state-licensed medical dispensary, and you're operating in a different legal universe than the guy buying the same plant, grown the same way, from a recreational storefront three blocks over. That recreational purchase is still, as of this writing, a Schedule I felony under federal law. This is not a metaphor for regulatory confusion. It is the literal, current, operative status of cannabis in the United States.

Two tracks are running in parallel right now, and they don't end in the same place. One is administrative: a DEA rescheduling process, triggered by Executive Order 14370 and pushed through an expedited evidentiary hearing in the summer of 2026, that can move cannabis within the CSA's existing tiers but cannot remove it from federal control. The other is legislative: a cluster of competing bills in Congress -- the Cannabis Administration and Opportunity Act, the MORE Act, STATES 2.0, and a forthcoming House GOP proposal -- aimed at varying degrees of full descheduling. These aren't two paths to the same destination. Rescheduling and descheduling are structurally different outcomes, and confusing them is the single most common mistake in how this moment gets covered.

This piece isn't going to predict a date when either track finishes. Nobody credible can, and anyone who tells you cannabis will be legal by a specific year in Congress is selling something. What we can do is model where each track plausibly lands seven to fifteen years out, using the closest historical analogues we have: how alcohol actually got re-legalized after Prohibition, how hemp's narrow 2018 carve-out from the CSA created gray zones that are still being litigated, and how state medical marijuana programs spent over a decade as the only legal lane before recreational markets followed. History doesn't hand us a date. It hands us a shape -- and the shape is worth taking seriously.

The Split Screen: What Actually Happened in 2026

Photo by David Trinks via Unsplash.

Start with the order that actually changed something. Executive Order 14370, signed December 18, 2025, directed the Department of Justice and DEA to expedite their review of cannabis's Schedule I status -- a review that had technically been pending, in various forms, since a 2016 petition and a subsequent HHS recommendation years later. Expediting a review isn't the same as completing one, but it forced the agencies to move on a timeline they'd been avoiding.

What came out the other end, on April 23, 2026, was narrower than a lot of people expected. Acting Attorney General Todd Blanche's order, effective April 28, moved exactly two categories of cannabis to Schedule III: FDA-approved drug products and state-licensed medical marijuana. That's it. Everything outside those two lanes -- which, by sales volume, means most of the country's actual cannabis market, since recreational sales dwarf medical sales in nearly every mature state system -- stayed Schedule I. That's not a typo or an oversight; it's the deliberate scope of the order.

This creates a bifurcation the CSA has genuinely never produced before. A substance normally gets one schedule. Cannabis, as of April 28, 2026, has one schedule for medical and pharmaceutical uses and a different one for recreational use of the identical plant. Layer on top of that two more lanes the order left completely untouched: synthetic THC analogues, which have their own tangled scheduling history, and hemp-derived cannabinoids, which occupy the separate legal space carved out by the 2018 Farm Bill. Four lanes, one plant, one statute.

The part worth sitting with is that this whole structure is administrative, not legislative. Blanche's order is a DOJ/DEA action under existing statutory authority, the same kind of authority that produced EO 14370 itself. A future attorney general, under a future administration with different priorities, could narrow it, freeze it, or attempt to reverse it through the same rulemaking channel -- something considerably harder to do to an act of Congress. Anyone treating April 28 as a permanent settlement is reading more durability into it than the mechanism actually provides.

The DEA Hearing: Who Showed Up and What It Signals

The DEA didn't just issue a rule and walk away -- it went through a full evidentiary hearing, the kind of proceeding federal agencies use when a rule change is contested enough to need a formal record. That hearing ran from June 29 to July 15, 2026, in front of Chief Administrative Law Judge Derek C. Julius, and the witness list tells you almost everything you need to know about where organized opposition to full Schedule III treatment is coming from.

Seven outside parties were granted designated-participant status: the National District Attorneys and Sheriffs Association, the Tennessee Bureau of Investigation, Smart Approaches to Marijuana, and the states of Nebraska, Idaho, Indiana, and Louisiana, plus DUID Victim Voices and physicians Kenneth Finn and Phillip Drum. Every single one of them opposed or expressed serious doubt about full Schedule III rescheduling. The DEA itself, notably, argued in favor of it -- meaning the agency conducting the hearing was on the opposite side of the table from most of the people it invited to testify. That's an unusual posture for a rulemaking body, and it signals genuine institutional disagreement rather than a rubber-stamp process.

Closing briefs were due August 17, 2026. From there, Julius issues a recommendation to the DEA Administrator, who holds final decision-making authority and isn't bound by the recommendation. There's no statutory clock on any of this. Observers have penciled in late 2026 or early 2027 for a final decision, but that's a projection built on how long comparable proceedings have taken, not a deadline written into law.

History argues for skepticism about that timeline. The original 2016 rescheduling petition process dragged on for years and ultimately ended in denial -- the agency simply declined to act on the timeline outside observers expected, and then declined to act at all. Agency proceedings involving controlled substances have a documented pattern of running long past initial estimates.

And even a final Administrator decision won't end the fight. Three petitions -- one from SAM, one jointly from Nebraska, Indiana, and Louisiana -- are already challenging the legality of the April order itself, separate from the hearing on the merits. That litigation proceeds regardless of what Julius recommends or what the Administrator ultimately decides, which means courts, not just the DEA, will have a say in when or whether this settles.

Path One: Schedule III as the Long-Term Status Quo

Photo by Johannes Schröter via Pexels.

Assume, for a moment, that the Schedule III order survives litigation, the hearing process resolves in DEA's favor, and nothing bigger happens in Congress for the next several years. What does that world actually look like a decade out? Cannabis remains a federally controlled substance, full stop. Schedule III doesn't touch the CSA's prohibition on interstate commerce in controlled substances, so a California cultivator still can't legally ship product to a Nevada dispensary under this framework, no matter how similar the two states' regulatory regimes look. And crucially, the recreational majority of the market -- the part generating most of the actual tax revenue and employment in states like Colorado, Michigan, and Illinois -- stays Schedule I indefinitely under the April order's current scope.

The tax picture splits along the same lines. Section 280E of the tax code, which bars businesses trafficking in Schedule I or II substances from deducting ordinary business expenses, would plausibly no longer apply to state-licensed medical operators once Schedule III sticks. That's real relief for a meaningful slice of the industry. But the recreational-only operators -- again, the bulk of the market by revenue in most legal states -- get nothing from this unless states find a way to restructure their programs as medical-adjacent, which is a heavier lift than it sounds given how differently medical and adult-use licensing regimes are built in most states.

The closest precedent for what a narrow, technical schedule change actually produces is hemp's 2018 Farm Bill removal from the CSA. That was a clean statutory carve-out, arguably more decisive than an agency reschedule -- and it still left a gray zone around intoxicating hemp-derived cannabinoids that state legislatures and courts are still fighting over years later. A partial fix, even a well-intentioned one, tends to generate its own downstream litigation rather than closing the book.

There's a real case for this arrangement lasting a long time. Agency rules move faster than statutes and don't require sixty Senate votes or a House majority willing to bring a bill to the floor -- if Congress stays gridlocked on cannabis, which it has been for most of the last decade, a Schedule III-only world could simply persist by default. The counter-case is just as real: three legal challenges are already filed, a future DOJ could reverse the order the same administrative way it was issued, and state markets that have built billion-dollar recreational sectors aren't going to stop lobbying Congress just because medical operators got relief. Watch which companies benefit fastest -- likely those pursuing FDA drug approval pathways in the tradition of Marinol and Epidiolex, since that lane has a clearer regulatory road than waiting on state-license products to catch up.

Path Two: Full Descheduling and the Bills Racing Toward It

The legislative track looks nothing like a single coordinated push -- it looks like several bills competing for the same lane, which is itself informative. The Cannabis Administration and Opportunity Act, filed July 16, 2026 by Senators Booker, Schumer, and Wyden along with fourteen other Democrats, would remove cannabis from the CSA entirely while explicitly preserving states' authority to set their own rules, and it pairs that removal with expungement provisions and social-equity and taxation frameworks. Thirteen days later, Senators Bennet and Hickenlooper reintroduced their own version -- a sign that even within the Democratic caucus, there's competition over whose descheduling bill becomes the vehicle, not just disagreement with Republicans over whether descheduling should happen at all. The MORE Act tracks closely with CAOA on the core descheduling mechanism and equity provisions, giving Democrats effectively two overlapping full-descheduling options in play at once.

STATES 2.0 is a genuinely different animal. Backed by eight bipartisan House sponsors, it doesn't touch the CSA's scheduling of cannabis at all -- it instead shields state- and tribal-legal cannabis activity from federal enforcement. That's closer to a truce than a resolution: it doesn't answer what cannabis is, federally, it just tells federal agencies to leave state-compliant operators alone. It's a meaningfully lower lift, politically, than descheduling, which is exactly why it's the one that's had the most staying power across multiple congressional sessions.

The most interesting development might be the House GOP nationwide descheduling bill expected before the midterms. Reporting suggests it's modeled on the MORE Act's descheduling core but stripped of the social-equity and expungement provisions that Democrats consider essential. That split matters: it shows descheduling itself now has support crossing party lines, even though the two sides want it for different reasons and are unwilling to pass the exact same bill to get there. Nearly 100 members of Congress currently back some version of a descheduling proposal -- a real, notable number, though still well short of what either chamber needs to actually pass something.

The historical analogy worth holding onto here is alcohol. Prohibition ended through the 21st Amendment in 1933 -- a full constitutional reversal, not an agency reclassifying liquor within some intermediate tier. Descheduling is the closer cousin to that kind of clean break: it ends federal control outright. Rescheduling, by contrast, resembles how other controlled substances have been shuffled between CSA tiers over the decades without ever exiting federal control -- a real change, but a change of degree rather than kind.

Modeling the Decade: Three Plausible Sequences

Illustrative projections suggest cannabis policy reform will unfold gradually, from a possible DEA Schedule III decision as early as 2027 to a full descheduling vote not likely before 2034—highlighting how each legal and legislative milestone builds on the last over roughly a decade.

Line up what we know, and three sequences seem plausible over the next decade -- none of them a certainty, and none of them tied to a hard date.

Scenario A, and the one the current trajectory most supports: the DEA finalizes Schedule III for the medical/FDA lane sometime around 2027, give or take the delays agency proceedings routinely produce. Recreational cannabis stays Schedule I for years afterward. Congress, rather than jumping straight to full descheduling, passes something closer to STATES 2.0 first -- an enforcement truce that doesn't resolve federal status but stops the pretense of enforcement against state-compliant operators. This mirrors almost exactly how medical marijuana rolled out state by state for well over a decade before recreational markets followed in most of those same states -- incremental legitimacy first, full resolution much later, if ever.

Scenario B: the litigation already filed by SAM and the Nebraska/Indiana/Louisiana coalition succeeds in stalling or vacating the April order on procedural grounds, freezing the current bifurcated status quo and pushing the entire fight back to Congress. This has real precedent -- courts have repeatedly delayed federal drug-scheduling actions on procedural challenges before, and an agency record built under an expedited hearing process is exactly the kind of thing a reviewing court might find vulnerable.

Scenario C, flagged explicitly as the most speculative of the three: a unified bipartisan bill emerges from the GOP and Democratic descheduling proposals converging on the core CSA removal while splitting off the equity and expungement provisions into a separate fight. This is the SAFE Banking pattern -- bipartisan language that got introduced, attached to must-pass legislation, and stalled repeatedly across multiple sessions before ever getting a real vote. For Scenario C to happen at all, you'd likely need a specific political trigger, like a shift in Senate composition that changes who controls the floor calendar. Nothing currently on the table guarantees that trigger arrives within this window.

None of these scenarios comes with a hard deadline attached, and any year mentioned above is a reasoned projection based on comparable historical timelines, not a forecast anyone should build a business plan around. What's worth noting is how differently each path resolves the things businesses actually care about: interstate commerce, banking access, and 280E relief land in very different places depending on which fork wins, and depending on whether it wins cleanly or through the kind of partial, litigated settlement that seems more likely based on the precedents above.

What Businesses and States Should Actually Plan For

Photo by Michael Hystead via Unsplash.

The most useful planning assumption right now is that fragmentation isn't a temporary condition to wait out -- it's the operating environment for the foreseeable future, regardless of which track wins. Multi-state operators should plan around continued state-by-state variation for years to come, because neither the current Schedule III order nor most of the pending bills, STATES 2.0 included, actually forces interstate commerce open. Even in an optimistic scenario, cross-border shipping between legal states likely requires a separate, later legislative or administrative step beyond whatever resolves scheduling itself.

On taxes, the caution is specific: 280E relief is not guaranteed for recreational operators under a Schedule III-only world, since the April 2026 order's protections currently run to FDA-approved products and state-licensed medical marijuana, not adult-use sales. Business models that have been penciling in automatic 280E relief on the assumption that rescheduling solves it for everyone need a contingency plan built around the slower, less certain full-descheduling timeline instead -- and need to be honest internally about which of those two outcomes their financial projections actually depend on.

Banking access deserves its own line item, separate from scheduling entirely. SAFE Banking-style relief has been introduced and stalled across multiple congressional sessions now, and that pattern tracks much closer to the STATES 2.0 truce model than to either full rescheduling or full descheduling. That repeated stall-out is itself a signal worth monitoring on its own terms, independent of whatever happens with the DEA hearing or the CAOA-style bills -- banking relief has its own political logic and its own coalition, and it may move (or stay stuck) on a completely different timeline than the scheduling fight.

Companies with international ambitions have an additional variable to track: how the EU and Canada treat a US market operating under Schedule III rather than full descheduling. Export and import rules for controlled substances generally require compatible scheduling status between trading partners, so a partial US fix could still leave cross-border commercial pathways constrained even if domestic operators see relief.

In the near term, the safest capital allocation bet is toward states with mature medical programs, since that's the lane the April 2026 order has already confirmed will move to Schedule III. It's the one part of this whole picture that isn't hypothetical.

It's tempting to frame this as chaos versus order -- a messy present giving way to a clean resolution once the DEA rules or Congress finally acts. That framing doesn't hold up. Neither live track actually produces a complete fix. Schedule III leaves interstate commerce closed and recreational operators outside 280E relief; the descheduling bills on the table, even in their most ambitious form, still leave banking and cross-border commerce as separate fights to be won later. The real choice isn't between chaos and order. It's between two different flavors of incomplete federal policy, each of which solves some problems while leaving others exactly where they started.

History's pattern here is pretty consistent, and it doesn't point toward a single clean legislative moment. Alcohol's path ran through over a decade of state-level nullification and enforcement erosion before the 21st Amendment ever passed. Hemp's 2018 exit from the CSA was a real fix that still left cannabinoid gray zones courts are litigating years later. State medical marijuana programs ran for more than a decade as the only legal lane before recreational markets followed, and even now the two coexist under different rules in the same states. The through-line is overlapping half-measures, stacked on top of each other over years, rather than one bill or one ruling that settles everything at once. There's no strong reason to expect cannabis policy to break that pattern in the next decade.

Which means the businesses and states that come out ahead won't be the ones that guessed the right date. They'll be the ones that treated the current fragmentation as the actual terrain -- building compliance structures, tax strategy, and market entry plans around the fork as it exists today, with contingencies for both branches, rather than parking capital and planning on the assumption that a single clean federal fix is coming on anyone's particular timeline.