Version 1.0 of this paper argued that the carbon price exists for one reason and dies for one reason. Version 2.0 corrects that thesis, because a program whose product is verified measurement must apply verification to its own arguments first. Carbon pricing serves several purposes at once: it internalizes climate damages, caps and ratchets emissions, shapes the operation of existing assets, reaches sectors with no clean substitute yet, raises public revenue, and anchors broader climate policy. As of 2026 it covers just over 29 percent of global greenhouse gas emissions and raised more than 107 billion dollars for public budgets in 2025 [5b]. It is expanding, not retiring.
What this paper defends is a narrower and still enormous claim: where the CTMP platform delivers fully substitutable, firm, low-carbon electricity and products below the full cost of fossil alternatives, without depending on carbon revenue, the carbon price ceases to be the marginal economic driver of substitution in those sectors. In those sectors, and only those, the instrument is succeeded rather than defeated. The paper then specifies the successor: CTMP Verified Metered Avoidance, an attestation record built from metered delivery, externally published grid data, transparent and separately reported consequential-impact calculations, corresponding-adjustment discipline under Article 6.2 of the Paris Agreement, and independent verification designed to make misconduct detectable, reviewable, and consequential.
The carbon ledger of a single 300 GW module is rebuilt in this edition on a mutually exclusive energy allocation of 2,496.6 TWh of modeled net annual generation, with displacement reported through an illustrative, unbounded scenario set rather than a single number: total avoidance of approximately 0.46 to 1.34 GtCO2e per year in the low, central, and fossil-heavy cases, falling toward roughly 0.02 GtCO2e in a zero-displacement floor case where only the firm industrial basket survives and the upstream deduction is still charged. Those totals already include the firm industrial substitution basket of roughly 60 MtCO2e, computed net of the platform's own modeled lifecycle intensities; conditional synthetic fuels and concrete carbon storage are reported separately and are never added to avoidance. The grid-export allocation behind every displacement figure is a residual placeholder pending a completed integrated-module energy balance. Engineered removal is treated as an energy-availability case, not a schedule: the platform's compounding surplus could power removal whose energy component alone falls to one to two dollars per tonne at owner cost, against an opportunity cost of 30 to 65 dollars per tonne of foregone electricity sales at the target tariff, while capture plant, sorbents, transport, storage, and verification remain real, unbuilt, and unclaimed. Every reported result in this paper is classified as measured, externally sourced, calculated, modeled, or targeted, and the calculation trail is published so each claim can be reproduced, challenged, and revised.
Every financial instrument is a response to a condition. Version 1.0 of this paper claimed the carbon price responds to exactly one: clean energy has historically cost more than dirty energy. That was part of the story presented as the whole of it, and the correction belongs at the front of this edition, not in a footnote.
Carbon taxes and emissions trading systems are designed to internalize climate damages that markets otherwise externalize, to place a declining cap on total emissions, to shape consumption and dispatch decisions on assets that already exist, to reach emissions for which no clean substitute is yet available, to raise public revenue, and to anchor wider climate policy [5, 5b]. The World Bank reports that direct carbon pricing now covers just over 29 percent of global greenhouse gas emissions and mobilized more than 107 billion dollars for public budgets in 2025 External. An institution that is expanding is not empirically approaching retirement, and this paper does not pretend otherwise.
What changes when firm, zero-carbon supply arrives below fossil cost is one specific function of the price: its role as the marginal economic driver of substitution. The CTMP platform is designed to produce firm baseload electricity at a modeled owner cost of $0.0008 per kilowatt-hour Modeled [1] and to sell it at a target posted tariff of $0.025 per kilowatt-hour Target [2]. Published new-build benchmarks place onshore wind near $0.034 per kilowatt-hour and unabated fossil generation between roughly $0.072 and $0.134 [3, 4] External. The target tariff sits below every published new-build fossil benchmark. It does not automatically undercut the operating cost of existing, depreciated, or subsidized assets, and new-build LCOE does not capture full system value; those comparisons are site-specific and are treated as such throughout.
Where the substitution is full, firm-for-firm, delivered, and cheaper, the tilt that the carbon price provides is no longer what moves the decision. In those sectors the instrument is not defeated. It is succeeded, and the succession should be designed in public before it arrives. That design is the remainder of this paper.
Carbon pricing is not one market but three instrument families sharing a vocabulary, and they must be diagnosed separately. Carbon taxes set a fixed price per tonne and create no tradable instrument at all; they are fiscal policy, and a substantial share of the 107 billion dollars mobilized in 2025 flows through them [5b] External. Allowance markets, emissions trading systems led by the EU ETS, traded on the order of 850 to 950 billion dollars of allowances in recent years [5]; they are closed cap-and-trade systems in which governments mint the currency and covered entities must surrender it, so no external party can flood a market whose currency it cannot mint. Crediting mechanisms are the third family: baseline-and-credit systems that issue offsets against modeled counterfactuals, spanning compliance-eligible credits and the voluntary carbon market, which transacted roughly 2 billion dollars at its 2021 peak and roughly 723 million dollars in 2023, with annual retirements on the order of 100 to 160 million tonnes [6] External. This paper's structural critique applies to the third family. The first two are policy instruments whose futures are set by legislatures and regulators, not by anyone's cost curve, and this paper claims nothing about their retirement.
Version 1.0 called that contraction "epistemic, not cyclical." That was a causal conclusion stated as fact. The defensible statement is this: the contraction was driven substantially by an integrity and confidence crisis, alongside macroeconomic conditions, uncertainty over corporate claims, regulatory uncertainty, and inventory overhang. Ecosystem Marketplace's own 2025 reporting describes a market in transition, with 2024 retirements relatively steady even as transaction volumes fell [6b] External.
The integrity evidence remains serious and must be stated fairly. A prominent 2023 investigation alleged that more than 90 percent of the examined rainforest credits from the market's leading registry lacked the claimed climate value [8]. Peer-reviewed analysis of 26 REDD+ project sites found that most did not significantly reduce deforestation and that reductions, where observed, were generally below credited levels [9]. The Berkeley Carbon Trading Project reached related conclusions across project categories [10]. Verra disputed the analyses and their methodology, and researchers published rebuttals in turn [8, 9] External. A rigorous paper reports the dispute, not just the headline.
The structural point survives the fairness edit. Every offset is a claim about a world that did not happen, and counterfactuals cannot be metered, only modeled. And the market's gatekeeping concept, additionality, carries a self-limiting logic. Additionality can be tested several ways: investment analysis, barriers, regulatory surplus, common practice, performance benchmarks, and positive lists [11]. But across all of them, as a technology becomes universally profitable, common, and legally required, its eligibility for additional crediting generally shrinks. The major registries restricted most grid-connected renewables in 2019 on exactly these grounds [12], though eligibility rules have continued to evolve since. The narrower formulation is the durable one: the better clean energy's economics become in a sector, the less of that sector the offset market is permitted to touch.
This section exists so that it may be quoted, permanently, against any future misrepresentation.
Three reasons, each architectural. First, additionality: a platform designed to earn its keep at a $0.025 tariff with no carbon revenue is, if it works as modeled, among the least additional energy projects ever conceived; under the market's own doctrine its avoidance could not generate legitimate credits, and selling them would reproduce the exact failure that broke the market's credibility in 2023. Second, the platform's published carbon accounting has declared its claim type from the start: avoided emissions, not carbon credits, no double counting [14]. Third, the Stewardship Charter's invariants forbid an entity from producing, pricing, selling, and verifying its own carbon instrument [15]. What CTMP produces is measurement plus transparent impact calculation. What independent parties produce is verification. What the world does with verified measurement is Section 5.
Consider the decision the carbon price influences on the margin: a buyer choosing between fossil and clean supply. Under the old condition, clean cost more, and an EU allowance at 60 to 90 euros per tonne adds roughly 2.4 to 3.6 cents per kilowatt-hour to unabated gas [5], narrowing the gap from the dirty side. Under the CTMP condition, if the platform performs as modeled, firm clean power at 2.5 cents undercuts new-build gas and coal before any carbon cost exists. In that transaction, the tilt is no longer what decides.
The historical analogy remains instructive if kept in its lane. Mechanical refrigeration did not crash the natural ice trade; it made the condition the ice market served cease to exist, and the market followed its condition into memory [16]. The analogy describes sectors CTMP fully displaces. It does not describe aviation, agriculture, land use, legacy assets mid-life, or jurisdictions the platform has not reached, where carbon pricing retains every one of the purposes listed in Section 1.
CTMP's strategic posture is therefore non-entry at speed: no sales into the voluntary market, no attempt to influence compliance prices, no celebration of the incumbent's decline. Deliver energy, molecules, and materials below fossil parity where the platform operates, meter every delivery, publish the calculations, and let each sector set the prosthetic down when its limb bears weight.
Version 1.0 said "what replaces a market in counterfactuals is a registry of facts." That sentence was wrong, and the correction is the intellectual center of this edition:
The meter establishes what CTMP generated, when, where it was delivered, what product was produced, and who received it. The meter cannot observe what would have happened without CTMP. Avoided emissions are, and remain, a consequential estimate: which generator would otherwise have run, whether new fossil capacity would have been built, whether renewables would have been curtailed, how dispatch and long-run investment respond. The Attestation therefore has two strictly separated layers.
Layer one, the metered record Measured (once built): a time-stamped, tamper-evident record of delivered energy or product, by hour, by delivery point, by counterparty, with the platform's own upstream and lifecycle emissions published against it.
Layer two, the consequential estimate Calculated: net displacement computed against externally published marginal emissions factors, using operating-margin methods for short-run effects and build-margin methods where the change is structural, consistent with the GHG Protocol's consequential-accounting work now in revision, with final standards expected in 2027 [17, 18]. A 300 GW module is plainly structural, and the paper says so rather than hiding behind an operating margin.
Corporate use follows the same separation. Buyers of CTMP power receive hourly delivery records: high-quality evidence for market-based Scope 2 inventory accounting of their purchased energy. Avoided-emissions figures are consequential impacts, reported separately, and are not deducted from any buyer's inventory [17]. Scope 3 value for CTMP products requires product carbon footprints, lifecycle boundaries, chain-of-custody, and assurance, all of which the vertical masterfiles must supply; hourly displacement data alone do not create Scope 3 evidence, and this paper claims none.
Sovereign use requires equal precision. Host states control authorization of internationally transferred mitigation outcomes under Article 6.2, and corresponding adjustments address double counting between participating national accounts [19]. The Paris Agreement does not by itself settle private-law ownership of mitigation outcomes; ownership and benefit allocation depend on domestic law, contracts, and the cooperative approach. And corresponding adjustments alone do not eliminate double issuance, double registration, double use, double claiming, or overlapping boundaries; separate registry, authorization, unique-identifier, and claims controls are required, and the Attestation architecture includes them. CTMP's standing posture is that sovereign value flows to hosts, implemented through those instruments rather than asserted as automatic ownership.
This section replaces its v1.0 predecessor entirely. Two errors are corrected in public. First, v1.0 computed Year One avoidance on 2,628 TWh (a 100 percent capacity factor) while computing compounded growth on 2,496.6 TWh (the 95 percent basis). One generation basis now governs everything: 2,496.6 TWh of modeled net annual generation (300 GW nameplate, 8,760 hours, 95 percent capacity factor) Modeled. Second, v1.0 counted the full generation as grid displacement while separately counting industrial products made from that same electricity. A megawatt-hour spent making hydrogen cannot also be a megawatt-hour displacing grid fossil generation. Every megawatt-hour is now assigned to exactly one destination.
| Destination | Basis | TWh / yr | Class |
|---|---|---|---|
| Metered grid export | Residual after internal allocations | 2,200.0 | Modeled |
| Green hydrogen, merchant | 2.0 Mt at ~52 MWh/t incl. compression | 104.0 | Modeled |
| Ammonia, methanol, SNG | Embedded hydrogen, synthesis, carbon feed | 90.0 | Modeled |
| Green steel | 15 Mt at ~3.6 MWh/t | 54.0 | Modeled |
| Desalination | 2.0 billion m³ at ~3.5 kWh/m³ | 7.0 | Modeled |
| Concrete, transport, auxiliaries, other verticals | Indicative pool pending masterfiles | 41.6 | Modeled |
| Net generation, Year One | 300 GW × 8,760 h × 95% | 2,496.6 | Calc |
These allocations are indicative and will be superseded line by line as vertical masterfiles publish audited energy intensities. The 2,200 TWh export figure in particular is a residual placeholder, not the output of a completed integrated-module energy balance; that balance, covering auxiliary loads, transmission losses, curtailment, and inter-vertical flows, is a required deliverable of the platform masterfile, and every displacement figure in Sections 6.2 and 6.4 inherits its uncertainty in full. The rule the table obeys is permanent: allocations are mutually exclusive and must sum to net generation, with transmission losses, curtailment, and removal energy added as explicit lines when metered.
Version 1.0 applied a single marginal emissions factor of 0.60 tCO2e per MWh to all generation. For a project of this scale that is a scenario input, not an outcome: a 300 GW module is a structural change to any grid it touches, build-margin effects are material, and the platform's own success drives the factor down. Displacement is therefore reported per scenario, computed on the 2,200 TWh export line with the platform's published upstream deduction of 0.02 tCO2e per MWh subtracted first [14].
| Illustrative case | Consequential factor, tCO2e/MWh | Net grid displacement, Gt/yr |
|---|---|---|
| No demonstrable displacement | 0.00 | (0.044): upstream deduction still charged |
| Low displacement | 0.20 | 0.396 |
| Central | 0.40 | 0.836 |
| Fossil-heavy grid | 0.60 | 1.276 |
| Site-specific verified | Hourly consequential model | Published post-PoC |
All figures are Calculated on Modeled inputs. None is "the module's carbon output." The set is illustrative, not exhaustive, and asserts no floor above zero: where displacement cannot be demonstrated, the ledger prints zero and still charges the upstream deduction against itself. The fossil-heavy figure is honest for early modules on fossil-dominant grids and dishonest as a fleet constant, for the reasons Section 9 makes explicit.
Version 1.0 multiplied product output by the incumbent's intensity, which silently assumed CTMP products have zero lifecycle emissions. They will not. Every vertical now follows one formula: avoided emissions equal output times (incumbent lifecycle intensity minus CTMP lifecycle intensity), with the comparator matched to the specific route displaced. CTMP lifecycle intensities below are modeled placeholders pending each vertical's audited masterfile.
| Vertical | Output | Comparator route [21, 22, 25] | CTMP intensity, modeled | Net avoided, Mt/yr |
|---|---|---|---|---|
| Green steel | 15 Mt | BF-BOF, ~2.33 t/t (global avg ~1.92; scrap-EAF ~0.69) | ~0.25 t/t | ~31.2 |
| Green hydrogen | 2.0 Mt | Unabated SMR, ~9 to 10 t/t | ~0.5 t/t | ~17 to 19 |
| Green ammonia | 2.267 Mt | Conventional, ~2.4 t/t | ~0.35 t/t; N2O and leakage tracked by end use | ~4.6 |
| Green methanol | 2.133 Mt | Fossil methanol | Conditional: atmospheric or biogenic carbon feed required; combustion returns the carbon | up to ~3.5, conditional |
| Synthetic natural gas | 0.8 Mt | Fossil gas incl. combustion | Conditional: same carbon-source rule; methane leakage deducted | up to ~1.9, conditional |
| Green concrete | 45M m³ | Conventional, ~250 kg/m³ embodied | ~120 kg/m³ modeled | ~5.9 |
| Concrete mineralization | 45M m³ | Storage, not substitution [26] | 10 to 25 kg CO2/m³ injected | 0.45 to 1.1 stored |
| Firm substitution basket (steel, hydrogen, ammonia, concrete substitution) | ~59 to 61 | |||
| Conditional synthetic fuels (methanol, SNG): zero until carbon-source conditions are met | 0, up to ~5.4 | |||
| Concrete mineralization: carbon storage, reported separately, never added to avoidance | 0.45 to 1.1 stored | |||
Synthetic fuels get one paragraph of unmissable honesty. SNG and methanol release their carbon when burned. They enter this ledger as low-carbon only where the carbon feedstock is atmospheric or biogenic, production energy is platform-metered, methane leakage is measured and deducted, and the carbon-cycle boundary is published. Absent those conditions their line is zero, and the ledger will print zero.
Totals below include the firm industrial basket; conditional fuels and concrete storage are excluded and reported separately above. Wherever a single range is quoted from this paper, it is these totals, and nothing is added to them.
| Illustrative case | Grid, Gt | Industrial firm, Gt | Total, Gt/yr |
|---|---|---|---|
| Floor: no demonstrable displacement | (0.044) | ~0.06 | ~0.02 |
| Low | 0.396 | ~0.06 | ~0.46 |
| Central | 0.836 | ~0.06 | ~0.90 |
| Fossil-heavy | 1.276 | ~0.06 | ~1.34 |
An illustrative range of roughly 0.02 to 1.34 gigatonnes per module-year, with no asserted floor and every dependency named, is smaller than v1.0's single 1.60 and worth more, because every line of it survives review. For calibration against the flow it addresses: global energy and industry emissions run at approximately 37 to 38 Gt per year [29].
Engineered carbon removal has remained marginal partly because of its energy bill. Direct air capture requires on the order of 1,200 to 2,600 kilowatt-hours of combined thermal and electric energy per tonne of CO2 [27], and current all-in costs run roughly 600 to 1,000 dollars per tonne [28] External. At the CTMP modeled internal transfer price of $0.0008 per kilowatt-hour, the energy component alone falls to between $0.96 and $2.08 per tonne Calculated on Modeled inputs. That is the input cost, not the economic cost: the same 1,200 to 2,600 kilowatt-hours could instead be sold at the target tariff of $0.025 per kilowatt-hour, so each removed tonne carries an opportunity cost of $30 to $65 in foregone electricity revenue, and any removal program must justify itself against that figure, not against the owner cost alone.
Version 1.0 let that arithmetic carry more than it can hold, and this edition pulls it back to what is defensible. Energy is a dominant and highly reducible cost component of direct air capture, but it is not the whole cost: capital, sorbent manufacture and replacement, contactor area, water and heat systems, compression, transport, characterized storage formations, injection capacity, monitoring, liability, and permitting all remain, and published system studies show that even large energy-cost improvements move total costs by fractions, not orders of magnitude, until the rest of the system scales [27, 28, 28b]. The correct claim is narrower: the platform is designed to remove energy price and energy firmness from the list of binding constraints at its sites. The remaining constraints are engineering and governance programs in their own right.
The energy-availability arithmetic, for the record: the compounding mandate grows capacity 10 percent per year [2] Target. With Year One as the baseline, surplus generation reaches approximately 524 TWh by Year Three, approximately 826 TWh by Year Four, and approximately 1,524 TWh by Year Six Calculated; v2.0 misassigned these figures to Years Three and Five, an off-by-one now corrected. Removal at 0.4 Gt per year would demand roughly 600 to 800 TWh at 1,500 to 2,000 kWh per tonne. The energy covers the full requirement by Year Four if the platform performs as modeled; whether a 400 Mt per year capture and storage industry can physically be constructed on any comparable timeline is a separate question this paper does not answer and therefore does not claim. The v1.0 commitment of 0.4 Gt removed by Year Three to Four is withdrawn.
Storage pathways remain as described in the literature: geological storage in saline formations and mineralizing basalts with durability measured in millennia [27]; mineralization into platform concrete at 10 to 25 kg per cubic meter, permanent as calcium carbonate within the material [25, 26]; and brine-derived alkalinity streams from the desalination refinery as mineralization feedstock [7]. No removal tonnage is claimed for any of them here. And the standing rule extends unchanged: CTMP will not sell removal credits. Removal tonnes, when they exist, are metered at injection or mineralization, attested identically to avoidance, and channeled through Section 5.
The platform's master plan includes district-scale electrified transit within the platform footprint. This paper assigns it a tonnage of zero and a measurement rule instead: transport displacement enters the ledger only when metered, computed per pod-kilometer against the published intensity of the mode verifiably replaced, using national transport inventory factors [20]. A ledger that refuses to estimate is a ledger that cannot be accused.
Earlier platform materials extrapolated per-module avoidance linearly to an 80-module program, yielding approximately 122 gigatonnes avoided per year. That figure was arithmetically faithful to its assumptions and physically impossible: global energy and industry emissions total approximately 37 to 38 Gt per year [29], displacement saturates at the size of the thing displaced, and the saturation is endogenous, since every module lowers the marginal factor of the grids around it. The brine refinery's masterfile already caps mineral monetization at market-share corridors for the identical reason [7]; this section applies that discipline to carbon, permanently.
The corrected fleet-level statement: platform-wide avoidance claims are capped by externally published, annually updated consequential factors and by the remaining stock of displaceable fossil generation and industry. At full deployment, aggregate avoidance asymptotically approaches, and can never exceed, the residual fossil flow, declining toward zero as the transition completes. A metric engineered to abolish itself is the only honest metric of transition.
Removal is different, and stating why requires more care than v1.0 gave it. Version 1.0 wrote that removal is "bounded only by the atmospheric stock," citing a 950 Gt figure that conflated distinct quantities. The quantities, separated: cumulative anthropogenic CO2 emissions since 1850 are on the order of 2,400 Gt and rising [30] External. Roughly 44 percent of emitted CO2 has remained airborne; atmospheric concentration stood near 428.6 ppm in April 2026 against a conventional pre-industrial reference of about 280 ppm [29b] External, an atmospheric excess of very roughly 1,100 to 1,200 Gt of CO2 mass Calculated. The removable legacy is smaller than the excess implies, because ocean and land carbon partially re-equilibrate against removal. And removal is constrained by far more than stock: energy, capital, storage, materials, land, water, ecological effects, carbon-cycle feedback, social acceptance, monitoring, and governance.
Within that corrected frame, the strategic point stands. The IPCC's assessed pathways require cumulative carbon dioxide removal in the hundreds of gigatonnes this century [30], and identify cost and energy among the reasons deployment lags requirement [27, 28, 30]. As avoidance saturates, the compounding surplus has a destination: it transfers, module by module, from displacement to drawdown, if and as the removal industry of Section 7 is actually built and verified. Avoidance is the errand. Removal, subject to every constraint named above, is the career.
Version 1.0 claimed a structure "incapable of self-dealing." No governance architecture can honestly claim that misconduct is impossible, and this edition retires the vocabulary of impossibility for the vocabulary of consequence:
The mechanisms, extended beyond v1.0: CTMP operates meters and publishes raw telemetry; consequential factors are sourced exclusively from independent, externally published datasets, never computed by CTMP [17]; attestors serve under staggered, publicly disclosed appointments with mandatory rotation, compensation fixed in advance, and a prohibition on contingent fees; audit assignments are randomized or multi-party; attestors carry defined liability; a whistleblower process and an appeal mechanism exist with published outcomes; audits are replicable from raw-data access; the methodology committee is independent of operations under the Charter's separation-of-duties rules with multi-party change control and tamper-evident logging [15]; conflict disclosures are public; and uncertainty and dissent are published alongside findings, not beneath them. Alignment with the ICVCM Core Carbon Principles on transparency, quantification, and no double counting is maintained not because the Attestation is a credit, but because those principles codify honest measurement [11]. Any tonne whose telemetry, attestation, or adjustment status is not current is reported as Not CTMP Verified, automatically [15].
One: "Your avoidance isn't additional." Correct, and disqualifying only for a product we refuse to sell. Section 3 forecloses offset sales permanently; the Attestation claims metered delivery plus a separately reported consequential estimate against external factors. One cannot fail an admissions exam for a school one is not applying to.
Two: "This is greenwashing with better typography." Greenwashing substitutes narrative for measurement. The Attestation publishes meters, external factors, deducted lifecycle emissions, scenario bounds, independent attestation, and an automatic Not Verified default. Name the specific meter, factor, deduction, or scenario you dispute; the architecture makes criticism become arithmetic to proceed.
Three: "You will double count." Corresponding adjustments address double counting between participating national accounts [19]; registry, authorization, unique-identifier, and claims controls address double issuance, registration, use, and claiming; corporate delivery records attach to inventory accounting while impact figures are reported separately and never netted [17]. Five failure modes, five named controls, all public.
Four: "Your marginal emissions factor is self-serving." No single factor is claimed and no floor is asserted. Displacement is published as an illustrative scenario set that includes a zero-displacement case, pending site-specific hourly consequential modeling, and every factor is externally sourced. A skeptic disputing the set is invited to publish a better model; the ledger will adopt it.
Five: "You destroyed the voluntary carbon market." We never entered it. If it contracts further, the causes will include its own documented integrity crisis [8, 9, 10] and the arrival of clean supply cheaper than fossil in specific sectors, the outcome its own additionality doctrine always defined as the finish line. Obsolescence by success is not sabotage. It is graduation, and it is sector by sector, not universal.
Six: "The 122-gigatonne claim proves you exaggerate." It proved our earlier extrapolation needed the saturation correction our mineral economics already carried, and this paper issued it publicly, unprompted, together with a corrected capacity-factor basis, a mutually exclusive energy allocation, and a withdrawn removal schedule. A program that audits its own headline numbers before critics do is exhibiting the property the Attestation exists to certify.
Seven: "Removal at one to two dollars per tonne is fantasy." The figure is the energy component only, labeled as such in the same sentence it appears, computed as 1,200 to 2,600 kWh per tonne [27] times a modeled $0.0008 per kWh, and the same energy carries a disclosed opportunity cost of $30 to $65 per tonne at the target tariff. Total removal cost includes capital, sorbents, transport, storage, and verification, remains far higher today [28], and will be published in the sequestration masterfile before any removal tonne is claimed. The claim is not that removal becomes cheap. The claim is that its energy line becomes small at platform sites, and that claim is arithmetic on stated inputs.
Eight: "The verifiers will be captured." Possibly; no structure makes capture impossible, and we no longer pretend otherwise. The structure fixes fees in advance, prohibits contingency, randomizes assignment, rotates appointments, imposes liability, opens raw data to replication, publishes conflicts and dissent, and defaults unattested tonnes to Not Verified. We do not ask the world to believe our referees are honest. We ask it to inspect a structure in which dishonesty is detectable, reviewable, and consequential.
Nine: "Your foundational numbers are unproven." Correct, and labeled. The owner cost, the capacity, the tariff, and every downstream figure are classified Modeled or Target until the proof-of-concept converts them, which is the entire purpose of the modular pathway: 50 MW proof-of-concept, then 250 MW, then the full core. This paper's job is to make the claims falsifiable in advance, so that when meters exist, they confirm or destroy something specific.
The people who built the carbon markets were engineers of the possible, working inside a condition where clean energy cost more and someone had to pay the difference. Registries, verifiers, and project developers accumulated exactly the skills the successor era needs most: measurement, attestation, standard-setting, and the institutional memory of every way a climate claim can fail. The Attestation economy will need thousands of such professionals as independent attestors, consequential modelers, and Article 6 practitioners, and the platform's standing invitation is that the incumbent market's honest infrastructure convert, not collapse. Carbon pricing itself, meanwhile, retains real work in every sector and jurisdiction the platform does not reach, and nature-based restoration retains a dignity no meter diminishes: ecosystems are not carbon instruments, and restoring them will remain worth doing when the last offset has expired, for reasons older than accounting.
For forty years, one of carbon pricing's jobs has been to make pollution expensive where clean energy could not yet be made cheap. Where the CTMP platform performs as designed, that specific job ends, sector by sector, delivery point by delivery point, and the honest successor is not a new tradable claim but a public record: metered delivery, published lifecycle emissions, transparent consequential estimates reported separately from anyone's inventory, verified by parties structurally arranged so that misconduct is detectable and costly, disciplined by illustrative scenario sets that include a zero-displacement case and assert no floor, and pointed, past the saturation of avoidance, toward the accumulated legacy overhead, on terms this paper refuses to promise before the machines exist.
The real contribution of this document is not the announcement that carbon markets are obsolete. They are not; they are expanding, and they will keep real jurisdiction for decades. The contribution is a demonstration of where the center of climate accounting moves when physical substitution becomes cheaper than financial permission: from purchasing claims to proving delivery and impact.
Per the change control provisions, the published edition of this document will carry retrievable records for every reference: URL or DOI, page or section numbers, access dates, and archived copies. The list below is the citation skeleton pending that conversion.