Negentropic Value Currency: A Research Program for Re-Anchoring Monetary Value in Net Order Creation

1. Introduction

Where does monetary value come from? Metallism appeals to the scarcity of precious metals; credit theory to sovereign guarantee; the quantity theory to the medium-of-exchange function; cryptocurrencies to algorithmic scarcity. Beneath these different answers lies a common premise: monetary value must rest on some form of scarcity. Scarcity, however, is a negative concept. It creates no order; it merely restricts possession. This paper pursues one question: can the monetary value anchor migrate from scarcity to net order creation? Put differently, can monetary value be grounded in the accounting of objective order increments, and does such a migration survive scrutiny at the three levels of theoretical coherence, microeconomic incentives, and empirical evidence?

Two developments give the question its urgency. The first is cosmological. The second law of thermodynamics implies that civilization, a dissipative structure, must continuously draw low-entropy resources from its environment in order to persist [33,35]; value creation is, at bottom, an ordering process that counteracts entropy. A terminological clarification is in order. Strictly speaking, statistical physics knows no entity called “negentropy,” only entropy-reducing processes and net entropy flows. Throughout this paper the term is used as an abbreviation, after Brillouin [6], whose precise meaning is “the net increment of systemic order.” The second development is economic. The technology cluster of strong artificial intelligence, controlled nuclear fusion, and atomic manufacturing is pushing the marginal cost of basic material provision toward zero, while value creation migrates into non-material domains: knowledge, ecology, trust, meaning. As the civilizational task shifts from allocating scarcity to creating order, the mismatch of scarcity-anchored money becomes harder to ignore. Global financial assets exceed 400% of GDP, of which less than 10% directly serve the real economy [5], and the Bitcoin network consumes more electricity each year than Argentina while producing no order [7].

Our starting point is the value equation W = √(E×T) proposed by Cheng and Cheng, and the framework builds on their Value Consensus Currency construction (Cheng [9]), whose currency unit the present framework renames the negentropic value currency (NVC; Definition 2). The paper is written in a conditional register from the outset. We distinguish model posits from physical facts, and we go one step further by distinguishing physical quantities from physically inspired accounting conventions. The Eτ and Tτ factors are operationalized through engineering and psychophysical constructs (exergy, ecological entropy debt, flow-state composites) that any party can recompute from public conventions, though the conventions themselves are not belief-free. The Iτ factor depends on a reference distribution that embodies scholarly consensus. The proposed anchor is therefore a two-layer structure in which the lower layer is not free of convention but free of unilateral discretion: independently recomputable under stated conventions, with the residual convention-dependence democratized and audited at the calibration layer (the α-coefficients). Section 15.1 collects the limitations that follow from this design choice.

The contributions, in order of weight. Because the paper is formally dense, we state at the outset where we take the originality to lie, so that the formal apparatus is read in its intended supporting role. The first and primary contribution is conceptual: the move from “entropy as cost” to “negentropy as value” — treating the net increment of systemic order, rather than any form of scarcity, as the ontology of monetary value (Section 3). This is a normative and conceptual step, defended philosophically in Section 3.4 and Section 3.5, not a theorem and not a consequence of physics. The second contribution is architectural: the demonstration that such an ontology can be made operational as a two-layer anchor, in which a measurement layer of recomputable order increments is separated from a calibration layer of democratized conventions, so that what the architecture eliminates is unilateral discretion rather than convention as such (Section 4, Section 10, Section 11). The third contribution is the formal apparatus that disciplines the architecture: screening criteria that select the functional form of the value equation within the twice-differentiable class (Lemma 1 — an application of standard functional-equation classification, not new mathematics), a robustness bound on ordinal rankings (Proposition 4), an error budget that converts the measurement objection into numerical tolerances (Proposition 5), and the incentive theorems of Section 6, Section 7 and Section 8, whose decentralization logic is standard and whose novel element is the measurability program that supplies observable shadow prices. The fourth contribution is evidential and deliberately modest: component-level consistency checks on the information dimension (Section 12), accompanied by a pre-registered extension protocol rather than by claims of identification. The fifth contribution is a disciplined statement of boundaries: what the theory claims, what it does not, and the premise layer on which each result rests (Table 5).

What this paper does not do. The program is a trilogy, and this paper is deliberately incomplete in two directions. First, it does not establish that a currency built on the negentropic anchor is macrostable. A self-consistent measuring rod is not yet a stable currency: whether an issuance rule proportional to audited order increments can deliver a stable unit of account is the central open problem of the program, and it is treated in the companion paper on macrostability [54], which constructs the full monetary life cycle — issuance, circulation, extinction — and rediagnoses inflation as the product of four identifiable integrity failures. Strictly speaking, no claim of the present paper survives the failure of that one. Second, this paper does not specify who governs the calibration layer. The α-coefficients carry value judgments, and the question of who sets them, under what procedures and constraints, is a constitutional question treated in the companion governance paper [61], which formalizes the consensus-generation problem and designs a bicameral algorithmic democracy with constitutional firewalls. Wherever the present paper invokes results of the companions, the invocation is flagged in the text, and every load-bearing element used here is restated in place.

2. Literature Positioning

2.1. From the Economics of Entropy: Cost Theory and Value Theory

The closest precursor of the negentropic theory of value is the bioeconomics of Georgescu-Roegen [19], which introduced the second law of thermodynamics into economic analysis and portrayed the economic process as the irreversible dissipation of low-entropy resources. Daly’s [14] steady-state economy and the exergy accounting of Ayres and Nair [2] pushed resource measurement toward physical rigor. This paper differs from that tradition in one specific respect. Where the economics of entropy treats entropy production as a cost constraint, a physical limit on growth, we elevate the net increment of order to the ontology of value. The move from “entropy as cost” to “negentropy as value” is the paper’s core conceptual contribution, and it should be read as a conceptual and normative step rather than a consequence of thermodynamics. Thermodynamics constrains which accounting schemes are physically meaningful; it does not select which of them defines value. The valuation strand of ecological economics supplies the closest empirical cousins of the accounting proposed here: the contingent and participatory valuation of ecosystem services (Costanza et al. [12]) already monetizes order stocks that scarcity-based accounts treat as zero, and the present program can be read as an attempt to generalize that precedent from ecosystems to the whole sphere of value creation.

2.2. The Incommensurability Challenge: Engaging the Critics, Not Only the Allies

The dialogue with ecological economics cannot rest on its friendly wing alone. The sharpest internal critique of the valuation tradition comes from within ecological economics itself, and it aims precisely at the move this paper generalizes. Martínez-Alier, Munda and O’Neill [57] argue that ecological economics rests on weak comparability of values: the dimensions of social value are irreducibly plural, no common numéraire exists that does not do violence to at least one of them, and the monetization of ecosystem services is therefore a category error dressed as measurement. Munda’s [58] social multi-criteria evaluation operationalizes the critique: where value dimensions conflict, the honest method is a multi-criteria matrix opened to deliberation, not a scalar index. Spash [59] radicalizes the objection into a methodological charge: cost–benefit monetization imports the utilitarian ontology of neoclassical economics into a field founded to escape it, and the pretense of a single metric suppresses the distributional and ethical content of environmental choice. O’Neill [60] supplies the philosophical root, distinguishing constitutive value — what a thing is for a form of life — from instrumental value, and arguing that market commensuration systematically confuses the two. If this critique is correct, then a monetary anchor built on a scalar measure of order increment does not merely risk measurement error; it repeats, at the level of the currency itself, the original sin of pricing the priceless.

The objection is the strongest one the present framework faces from its natural readership, and the answer must be structural rather than rhetorical. Four replies delimit what the framework actually commensurates.

First, the target of commensuration is the unit of account, not the spectrum of values. Money is a scalar instrument by construction: any monetary system, including one anchored in gold, sovereign credit, or a multi-criteria basket, must ultimately express obligations in a single dimension. The incommensurability critique correctly establishes that social value is plural; it does not establish that no scalar accounting unit is legitimate, because it proves too much — on the strong reading, not only NVC but every monetary economy, including the steady-state economy the critics themselves defend, is committed to the same sin whenever it issues a price. The question is therefore not whether a scalar exists but what the scalar is anchored to and who controls its drift. Scarcity-anchored scalars answer: to the market’s existing power structure. The proposal here answers: to publicly recomputable order increments, with the residual conventions democratized.

Second, the two-layer architecture relocates plurality rather than suppressing it. The measurement layer produces a scalar Wτ; the calibration layer — the α-coefficients generated through the democratic procedure of Section 8 — is precisely where the plural, conflicting, incommensurable social values enter and remain visible as objects of public contestation. In Munda’s [58] terms, the framework does not replace the multi-criteria matrix with an index; it assigns the matrix and the index to different institutional layers, with the matrix governing the index’s weights. What is commensurated is the anchor; what is deliberated is the direction. The companion governance paper [61] carries the full institutional argument.

Third, the constitutive/instrumental distinction of O’Neill [60] is preserved by the stratified ontology of Section 3.2. The maintenance layer of negentropic value — care, stewardship, the holding of systems in a living state — is accounted not because care is instrumentally productive but because the definition of value itself counts the preservation of order as value. Constitutive worth is thereby brought inside the accounting boundary rather than priced at the margin from outside it. Whether the operationalizations of Section 4.2 do justice to this intention is an empirical question about the measurement layer, and we flag it as such; but the framework is not open to the charge that it recognizes only instrumental value.

Fourth, the residual agreement deserves equal emphasis. The critics are right that every scalar embeds a convention, and this paper does not claim otherwise: the reference distribution of Iτ, the ecological entropy debt, and the flow-state composite are all conventions, publicly auditable but conventions nonetheless (Section 3.3). Where we part from the incommensurability school is not over the existence of convention but over the remedy. Their remedy — multi-criteria deliberation without a scalar — abandons the unit of account to the default scalar, which is the existing scarcity money whose biases motivated the critique. Our remedy is to make the scalar’s conventions explicit, recomputable, and democratically revisable. Between an implicit scalar captured by incumbent power and an explicit scalar under audit, the second is the weaker evil and the stronger instrument.

2.3. Promissory Anchors and Their Alternatives

Hayek [22] and the free-banking school [36] anchor value in issuer reputation and redemption promises; Modern Money Theory [26,43] identifies the tax-driven mechanics of sovereign money; central bank digital currency research [4] retains the sovereign credit anchor. These are promissory anchors: value depends on the stability of subjects’ beliefs, and beliefs can form and dissolve. The collapse of the gold standard, sovereign defaults, and episodes of preference drift all testify to this. The two-layer anchor proposed here differs in structure rather than in kind. Its measurement layer, recomputable order increments under public conventions, depends on no promise; its calibration layer, the α-coefficients, is set through democratic procedure. The stability requirement on belief is thereby reduced from the whole anchor to the calibration layer of the anchor. This is an improvement in robustness, not the disappearance of convention, and we state the point precisely so that the claim cannot be over-read.

One further tradition deserves explicit engagement, because it shares the diagnosis while rejecting the remedy. Post-Keynesian endogenous-money theory [56] dissolves the exogenous-money myth and locates money creation in credit demand; on its account, the quantity of money was never the locus of value to begin with. The present program agrees that money is created rather than dug up, but draws the opposite conclusion about anchoring: endogenous credit creation answers how money enters, while leaving open what one unit of it is worth. The negentropic anchor can be read as supplying exactly the missing object — a discretion-free unit of account compatible with endogenous creation — and the companion paper [54] shows where the two traditions must then part: the credit-multiplier channel has to be structurally closed, not merely regulated.

2.4. Physical-Quantity Anchors: Energy Standards and Demurrage

A further lineage must be confronted directly, because it is the closest structural relative of the present proposal. Soddy [49] argued that monetary wealth should be anchored to the physical flow of available energy; the technocracy movement of the 1930s proposed replacing price money with energy certificates denominated in units of work [52]; Kümmel [50] demonstrated that exergy input carries genuine explanatory power in production functions; and the thermoeconomics tradition has long priced goods by their embodied exergy. These proposals share the conviction that monetary value should answer to physics. They also share a defect that the present framework is designed to avoid: energy throughput as such rewards dissipation. A currency issued against joules consumed would subsidize waste heat as readily as order creation. The anchor proposed here is not energy input but the net order increment that energy input helps to produce, with the ecological entropy debt of the source deducted at the point of accounting (Section 4.2). Exergy enters our equation as one factor among three, not as the value substance itself. The same discipline applies reflexively: the ecological entropy-debt term of Section 4.2 is convention-dependent in much the way we criticize in the energy-standard tradition. The difference is locational, not categorical — our conventions sit in a published, audited, democratically revisable accounting layer rather than in the denomination itself — and we do not claim more.

A second member of this lineage concerns the velocity of money rather than its backing. Gesell’s [46] stamped scrip imposed a periodic holding cost on idle balances, an idea Keynes treated with qualified respect in the General Theory [47], and the Wörgl experiment of 1932–33 supplied a brief practical demonstration before its suppression by the central bank [51]. Our intrinsic metabolism rate δ is demurrage redesigned rather than demurrage renamed: it is constitutionally bounded, publicly calibrated, and, unlike Gesell’s stamp, fully avoidable without penalty through deposit in the Civilization Time Bank, so that saving is redirected into patient capital rather than punished (the constitutional design is treated in the companion governance paper [61]). We flag this ancestry here because the intellectual honesty of the program requires it: what is new is not the holding tax but its integration into a complete issuance-and-extinction life cycle anchored to measured order creation.

Finally, within the cryptocurrency tradition, proof-of-useful-work and regenerative-finance proposals gesture at issuance tied to verified external benefit. They remain, however, anchored in algorithmic scarcity at the level of supply caps, and none specifies how heterogeneous useful outputs are to be made commensurable. The measurement architecture of Section 11 is our answer to that commensurability problem — and Section 2.1a records the price that answer pays to the incommensurability critique.

2.5. The Scarcity Paradigm and Its Blind Spots

Metallism, the labor theory of value, marginal utility theory, and MMT differ in form yet share one structural feature: the value anchor depends on the stability of collective belief or on discretionary issuance. The macroeconomic costs of scarcity-based money are extensively documented, from credit idling and asset bubbles to ecological overdraft (global ecological footprint at 1.75 times carrying capacity [44]). Order increments differ in one specific respect. Their measurement layer is publicly recomputable under stated conventions and does not fluctuate with the mood of belief. Recomputability, rather than any alleged freedom from convention, is the source of the robustness claimed here.

3. The Negentropic Theory of Value

3.1. Concept

Definition 1 (net order increment). Value is the net increment of systemic order achieved by an open system, through the exchange of energy, matter, and information, within a specific time window: the integral measure of local entropy reduction, net of the entropy cost exported to the environment. The physical picture requires care, because the second law forbids the shortcut an earlier formulation of this program once took. For any value-creating process, the entropy of the system itself may fall, dSsystem < 0, while the entropy production of the system together with its environment remains non-negative, σ = dSsystem + dSenvironment ≥ 0, with equality only in the reversible limit. Ordering is always purchased at an environmental entropy price; nothing in the theory requires or permits σ < 0. Schrödinger’s [35] phrase “feeding on negative entropy” denotes precisely this lawful export of entropy, not a negative joint production. The economic content of Definition 1 is therefore double-sided: the credit side is the local entropy reduction embodied in the ordered structure, and the debit side is the environmental entropy cost of producing it. Both sides reappear in the operationalized energy factor of Section 4.2, where the exergy efficiency measures the credit and the ecological entropy debt T0ΔSeco charges the debit; the definition and the accounting formula are two statements of one structure.

Definition 2 (negentropic value currency, NVC). A currency whose value ontology is the net increment of systemic order, whose unit of value is calibrated by the negentropic action, and whose issuance is anchored to audited negentropic-creation events.

Terminological discipline: “negentropy” means net order increment; “negentropic creation” means human activity producing net entropy reduction at the system level under the second-law constraint just stated; “negentropic value” means the monetized measure of that increment.

3.2. Stratified Connotation

The connotation of negentropic value has three layers. The first is physical maintenance: life care, ecological stewardship, infrastructure operation, the work of holding a system in a living state far from equilibrium and preventing its slide toward thermodynamic equilibrium. The value of care and watchkeeping labor (Folbre [18]; UN Women [41]) belongs to this baseline layer. The second is productive generation: manufacturing, construction, agriculture, the transformation of disorderly raw materials into functional ordered structures, the value layer of traditional industry. The third is cognitive ascent: scientific discoveries, technological paradigms, knowledge systems. Cognitive creation does not merely produce ordered structures; it produces templates for creating order. Knowledge is the only form of negentropy that can reproduce itself, accumulate across generations, and amplify nonlinearly, and once a theory is established its value lies in the permanent compression of humanity’s cognitive uncertainty about the universe rather than in the exergy it consumed. This third layer, the pivot of civilization utility theory (Section 5), guides the productive layer, which in turn supports the maintenance layer; the three are cumulative rather than substitutive.

3.3. What the Theory Claims, and What It Does Not

To keep the propositions defensible, the boundaries must be drawn first. The theory claims, first, that value creation has observable physical correlates: any act of value creation consumes usable energy, occupies time, and produces information structure. Second, the monetary value anchor can be built on the accounting of net order increments under public, recomputable conventions, and thereby escape dependence on unilateral discretion and open-ended promises. Third, beyond serving as a medium of exchange, a monetary system can perform a civilizational navigation function.

The theory does not claim that value is a purely physical quantity. The value posited here is an ontological assumption, not a physical theorem, and the commensuration of energy, time, and information derives from the logical constraints of axioms and screening criteria, not from conservation laws. Nor does it claim that the measurement layer is convention-free. Exergy efficiency, ecological entropy debt, the flow-state composite, and the reference distribution of Iτ all embed finite conventions; what the architecture eliminates is unilateral discretion over issuance, not convention as such. The two-layer structure is therefore one of democratized convention rather than physics without interpretation — and it is precisely this feature that answers the incommensurability critique of Section 2.1a: the framework does not pretend to have dissolved the plurality of values into physics; it assigns the plurality to the calibration layer and the scalar to the measurement layer. Finally, the theory does not deny the importance of subjective preference. Preferences are fully respected at the market-exchange layer (Section 10.6); the negentropic theory of value constrains the navigation layer of monetary incentives, not the expression layer of preferences.

3.4. Normative Foundations

The sharpest challenge arrives at once: on what grounds should civilization’s meta-objective be order-maximization? Are freedom, happiness, and equality not equally legitimate goals? Our answer rests on a lexicographic hierarchy of goals.

Proposition 1 (priority of survival). Freedom, happiness, and equality are questions of value distribution within a system, and their logical prerequisite is the existence of the system. A civilization that ceases to maintain its own order disintegrates thermodynamically, and all second-order goals lose their bearer at the same moment. Writing P(N) for the survival probability of civilization as a function of its order stock N, survival requires the net negentropy flow to remain above a critical value; below it, P(N) → 0. The expected realization of any second-order goal equals P(N) multiplied by its level, so any rational civilizational planning must treat N ≥ Ncritical as a hard constraint. This ordering does not belittle second-order goals. It installs them at the correct logical location: they operate at the policy layer (α-weights, UBI design, tax design), while the order objective operates at the constitutional layer. The two stand in the relation of constraint and optimization, not of competition.

Proposition 2 (self-support). The nihilist challenge (“why survive at all?”) dissolves in reflective equilibrium: any value judgment presupposes the existence of a judging subject, and judging subjects presuppose ordered structures. “Maintain and expand order” is thus the minimal postulate that does not beg the question, since rejecting it dissolves the very possibility of asking. Its epistemic status should be stated with care. This is a transcendental-pragmatic consistency argument, not an empirical hypothesis, and we make no Popperian falsifiability claim for it. The associated historical conjecture, that civilizations adopting net-negentropy-depleting paths tend to perish, is a probabilistic claim to be weighed against history, not a testable prediction in the strict sense.

Proposition 3 (incompleteness of alternatives). Competing meta-objectives are incomplete when pursued on their own. Freedom-maximization without the survival constraint dissolves the material basis of social cooperation; happiness-maximization without order-accumulation degenerates into fleeting consumption; equality-maximization without a growth engine reduces to the equalization of poverty. Each must be embedded within the hard constraint of survival-order in order to be achievable. Order-maximization is not one good among many but the physical precondition of all other goods, dominant at the meta-level rather than competitive at the same level.

3.5. Replies to Five Counter-Positions

Reply 1 (subjective value theory). Subjective utility explains exchange ratios within a given institutional framework, but it struggles to anchor the unit of account itself: in the post-marginal-revolution framework utility is ordinal, and an ordinal system provides no natural cardinal accounting anchor. Mises’ regression theorem supplies an auxiliary historical-evolutionary perspective here; we cite it as supporting argument, not as decisive proof, since modern institutions can of course create accounting units by design. Our claim is comparative, namely that a recomputable order anchor is more robust to discretionary drift than the available alternatives. Subjective preference and negentropic value operate at different levels: the former answers what people want, the latter what deserves to be rewarded by the monetary incentive layer if civilization is to persist.

Reply 2 (labor theory of value). The core insight of the labor theory, that value derives from the process of creation rather than from things, is retained. Its fatal defect, the incommensurability of heterogeneous labor, is addressed by the net order scale, on which physical, cognitive, emotional, and care labor become comparable and aggregable, with the comparability resting on public measurement conventions rather than on a physical theorem.

Reply 3 (state credit theory). Sovereign credit is endogenous to the production and energy base; a sovereign whose fiscal capacity is exhausted, as the history of hyperinflations shows, is an anchor that fails. The order anchor grounds monetary value in publicly recomputable physical accounting rather than in political promises.

Reply 4 (algorithmic scarcity). Algorithmic scarcity is manufactured scarcity. Proof-of-work consumes enormous energy while producing no order (Iτ ≈ 0, χ ≈ 0, hence Wτ ≈ 0), so crypto-mining automatically loses monetary-issuance qualification under the proposed system.

Reply 5 (nihilism). See Proposition 2. Taken together, the five replies show that the negentropic theory of value excludes no other value dimension; it supplies the precondition of survival and a common measuring scale. Its posited status is declared openly, and its empirical consequences are open to testing. This is the maximum rigor available to it.

4. Derivation of the Negentropic Value Equation

4.1. Axioms and Screening Criteria

The Cheng equation W = √(E×T), developed from the connotative-value formulation of Cheng and Cheng [8], rests on two axioms. Axiom 1 (cost necessity): any value creation consumes energy and occupies time. Axiom 2 (orderliness): the core attribute of value is the net elevation of systemic order. Axioms alone do not determine a functional form, and four screening criteria are added: the necessity chain (value approaches zero as any factor approaches zero); dimensional compatibility (which excludes arithmetic and harmonic means); scale-invariance, f(λE, λT, I) = λf(E, T, I); and diminishing marginal returns. The most general family satisfying all four criteria is the power-function form

Wτ = k·Eτ^a·Tτ^1−a·Iτ^c, 0 < a, c < 1

(1)

The symmetric benchmark a = c = 1/2 (hence b = 1−a = 1/2), with k = 1, gives Wτ = √(Eτ·Tτ·Iτ). The benchmark gives symmetric exponents to the two dimensional core factors, comparable elasticity contributions at typical scales of the normalized Iτ ∈ [0,1], and a square-root law that preserves recognition of stepwise increases in Iτ such as paradigm transitions. The exponent restriction b = 1−a is not an arbitrary normalization: it is forced by the scale-invariance criterion C2 of Lemma 1, which requires Eτ and Tτ to enter jointly homogeneously of degree one. Within the twice-differentiable class the functional form is unique, and only the exponents a and c remain calibration freedom to be estimated from CRN historical data. We note for the record that the passage from the screening criteria to the power family is an application of the standard functional-equation argument behind Cobb-Douglas-type forms; the derivation organizes the choice but imports no new mathematics.

The constant k carries the dimensions required by the chosen exponents. If Eτ and Tτ are measured in J and s, Iτ is normalized dimensionless, and Wτ is expressed in √(J·s), then [k] = J^1/2−a·s^a−1/2. In the symmetric benchmark a = 1/2 the constant is dimensionless and set to unity by convention. Equivalently, one may normalize Eτ = E0·Ẽτ and Tτ = T0·T̃τ and write Wτ = W0·Ẽτ^a·T̃τ^1−a·Iτ^c with W0 ≡ k·E0^a·T0^1−a, which absorbs k into the reference scales; W0 reduces to √(E0·T0) exactly at the benchmark.

Lemma 1 (uniqueness of the functional form). Let f: ℝ³₊ → ℝ₊ be twice continuously differentiable and satisfy: (C1) the necessity chain, f(E, T, I) → 0 whenever any factor tends to zero; (C2) dimensional homogeneity, f(λE, λT, I) = λf(E, T, I) for all λ > 0; (C3) factor separability, the logarithmic cross-partials ∂²ln f/∂xi∂xj vanish for distinct factors; (C4) diminishing marginal returns, f strictly increasing and strictly concave in each factor. Then f(E, T, I) = k·Ea·T^1−a·Ic with k > 0, a ∈ (0,1), c ∈ (0,1). Proof. (C3) reduces f to a product of univariate functions, f = φ(E)·ψ(T)·χ(I); (C2) forces φ(E)ψ(T) to be homogeneous of degree one in (E, T), and by the Aczél classification of homogeneous twice-differentiable functions [55], φ(E)ψ(T) = k·Ea·T^1−a; (C1) excludes additive intercepts, and (C4) confines the exponents to the open unit interval. □ The screening criteria therefore do not merely suggest the power family: within the twice-differentiable class they select it, and only the symmetric benchmark a = c = 1/2 (hence b = 1−a = 1/2) remains a convention rather than a theorem.

On the status of Lemma 1. Because the lemma carries the word “uniqueness,” its role in the argument should be stated with exactness. The result is a classification theorem in the sense of Aczél [55]: it shows that four desiderata, once adopted, leave no freedom in the functional family — it does not show that the desiderata themselves are forced by nature. The originality of this paper does not reside in this mathematics, which is standard, but upstream of it: in the identification of net order increment as the object the equation should measure, in the choice of the three factors, and in the two-layer architecture that makes the resulting accounting auditable. Lemma 1’s contribution to that program is defensive rather than foundational: it guarantees that within the chosen class there is exactly one form to calibrate, so that all subsequent empirical and institutional argument can focus on the exponents and the measurement layer rather than on an open-ended search over functional forms. Readers who regard the screening criteria as contestable should read the lemma as conditional on them — which is how we read it ourselves, as the caveat on C3 below makes explicit.

One caveat on C3 is due, because it is the strongest of the four criteria. Vanishing logarithmic cross-partials rule out interaction terms among the factors; if evidence were to favor interactions — information gain raising the marginal product of energy, say — the uniqueness conclusion would fail and the admissible family would widen beyond the power form. Uniqueness is therefore claimed conditional on C3, not derived from more primitive axioms, and the criterion is empirically examinable: cross-partials of ln Wτ estimated on CRN data that differ significantly from zero would falsify separability and force a richer functional class. Proposition 4 bounds the ranking damage from whatever misspecification a false separability assumption might hide, which is why the benchmark can be retained honestly while C3 awaits adjudication.

Proposition 4 (robustness of ordinal rankings to exponent calibration). Let x = (E, T, I) and x′ = (E′, T′, I′) be two activity profiles whose benchmark values satisfy W(x) > W(x′), and write ρ ≡ ln(W(x)/W(x′)) > 0 and L ≡ max{|ln(E/E′)|, |ln(T/T′)|, |ln(I/I′)|}. Then the ranking is preserved under every exponent perturbation with 2|Δa| + |Δc| < ρ/L (admissible perturbations respect the constraint b = 1−a, so that Δb = −Δa). Proof. The log-ratio of the two values is a·ln(E/E′) + b·ln(T/T′) + c·ln(I/I′); a perturbation shifts it by Δa·ln(E/E′) + Δb·ln(T/T′) + Δc·ln(I/I′) = Δa·[ln(E/E′) − ln(T/T′)] + Δc·ln(I/I′), whose absolute value is at most (2|Δa| + |Δc|)·L < ρ, so the log-ratio stays positive. □ The proposition turns the conventionality of the benchmark into a quantitative statement. For profiles differing by a factor of two in value (ρ = ln 2) and by at most a factor of e in each factor (L = 1), any perturbation with 2|Δa| + |Δc| below 0.69 leaves the ranking intact (for example, |Δa| < 0.35 when Δc = 0). Rankings between near-tied activities are correspondingly fragile — which is exactly why fine discrimination is assigned to the calibration layer (α) rather than to the equation.

Two closing remarks delimit what this section establishes. First, the coherence claimed for the value equation is layer-relative: the definitions, the functional form, the robustness and error propositions, and the dimensional accounting cohere within the premise layers L0–L4 mapped in Table 5. This is the consistency of an auditable accounting convention (Section 3.3), not the unconditional necessity of a physical law, and the difference is a matter of design rather than an admission of defect. Second, coherence of the accounting layer is not coherence of the monetary anchor: whether an issuance rule built on Wτ stabilizes a price level is the macrostability problem of Section 13, which remains open.

4.2. Operationalization of the Three Factors

Eτ (net negentropy flow input):

Eτ = max{0, ηex·Ephysical − T0·ΔSeco}, 0 < ηex ≤ 1

(2)

Exergy is the carrier of ordering potential, and a positive net Eτ is the condition of accounting entry. Equation (2) now formalizes this entry condition as a truncation at zero: an activity whose ecological entropy debt exhausts its exergy input records Eτ = 0, hence Wτ = 0 by the necessity chain, with the negative balance posted to the entropy-debt ledger introduced at eq (4). The truncation also keeps Wτ real-valued under non-integer exponents, since the power form is evaluated only on the non-negative orthant. Balances recorded in the entropy-debt ledger never enter Wτ; they serve audit, future offset against subsequent negentropic creation, and the negative-value record, so that the “net” in net order increment is accounted on both sides — the credit side in Wτ, the debit side in the ledger. Two clarifications matter for physical consistency. First, ηex is a second-law (exergy) efficiency, bounded above by unity in all cases without exception. The advantage of renewable sources enters not through an efficiency above one, which would violate the second law and undercut the very physical foundation the paper claims, but through a near-zero ecological entropy debt T0·ΔSeco, the cumulative climate and ecosystem damage of the source expressed in entropy units. Second, the debt term itself must be estimated under public conventions. It is convention-dependent, publicly recomputable, and auditable, and we classify it as such rather than as a raw physical quantity. Clean and polluting sources are thereby priced differently at the source of value: by physical law for the efficiency component, by audited convention for the debt component. The subtraction of the debt term is also where Definition 1 becomes operational: the double-sided structure of entropy export stated there is charged here, in the one place where the equation touches the environment.

Tτ (non-equilibrium maintenance time):

Tτ = ∫ (||∇S||/||∇S||max)·χ·dt

(3)

Here χ is a flow-state composite index built from EEG α/β/γ power ratios, heart-rate variability, eye-tracking concentration, and behavioral logs [13,31]. The maintenance-order contribution of care labor becomes measurable in principle on this construction. High-frequency trading has χ ≈ 0 and ||∇S|| ≈ 0, hence Tτ ≈ 0. This is an operationalized empirical proposition resting on psychophysical measurement conventions, and we present it as such rather than as a purely physical measurement. The normalization ||∇S||max is taken over a declared CRN reference frame of comparable activities, not within each activity: an activity-internal maximum would make every activity maximal in its own units and destroy comparability, whereas a frame-level maximum fixes a shared scale. Tτ is therefore an intra-frame measure, and cross-frame comparisons require renormalization against a common reference maximum — a convention of the same status, and the same auditability, as the reference distribution of Iτ in eq (4).

Iτ (structural information gain):

Iτ = max{0, DKL(pinitial||pref) − DKL(pfinal||pref)} / Href

(4)

Iτ measures, in the Shannon [37] sense of information content, the reduction in KL divergence (Kullback and Leibler [29]) relative to the empirical distribution of the field’s accepted knowledge base, the negative of Bayesian surprise [24]. Residual dependence on the reference distribution is examined through cross-corpus sensitivity analysis, and cross-civilizational application requires reference-distribution calibration. The implication for the anchor claim should be faced squarely: Iτ is convention-dependent through pref, so the physical layer of the two-layer anchor is genuinely physical only for the Eτ component and partially for Tτ. The information component is an intersubjectively stabilized convention. The robustness claim of the architecture is accordingly the modest one stated in Section 2.2: elimination of unilateral discretion, with democratized convention, not elimination of convention itself. Negative values of the bracketed difference represent information loss and do not enter negentropic value accounting; they are recorded instead in an entropy-debt (negative-value) ledger. The clipping at zero is a deliberate accounting choice: it renders information-destroying activities valueless alike (Wτ = 0 by the necessity chain), and the price of this simplicity is that degrees of information loss are distinguished only in the ledger, not in Wτ. The normalization constant Href is set as the supremum of the attainable KL reduction within the field, so that Iτ ∈ [0,1]; together with the positive-Eτ entry condition stated above and Tτ ≥ 0 by construction, this guarantees Iτ ≥ 0, so that Eτ·Tτ·Iτ ≥ 0 and Wτ is real-valued and non-negative over the entire accounting domain. Comparability of Iτ is accordingly intra-frame: numerical comparisons across fields presuppose a shared reference-distribution calibration and are re-established across frames only by recalibration.

4.3. Notation and Dimensional Remarks

Define the benchmark negentropic action Vτ ≡ Eτ·Tτ·Iτ, in loose terms the total “ordering work” embodied in an activity, combining its energy input, its duration of directed intervention, and its informational gain. Its dimension is J·s, the dimension of action in physics. Under the symmetric benchmark a = c = 1/2 and k = 1 the value scale is the concave transform Wτ = √Vτ, which yields diminishing returns; in the general calibrated form (1), Wτ² = k²·Eτ^2a·Tτ^2(1−a)·Iτ^2c, which coincides with Vτ only at the benchmark. We therefore treat Vτ as the benchmark ordering-work aggregate and Wτ as the calibrated value scale; the dimensional coincidence Vτ ∼ J·s remains a consistency check on the benchmark, not a theorem about the general calibrated form, since dimensional homogeneity is a necessary condition for physical grounding, not a sufficient one, and the equation is grounded axiomatically and normatively (Section 3) with empirical content to be tested (Section 5, Section 12, Section 14), never by dimension alone. Algorithmic information (Kolmogorov [28]) offers an alternative operationalization of the informational factor that we do not adopt here, since the KL-reduction form connects directly to the reference-distribution architecture of Section 11.

A remark on the unit chain closes the dimensional accounting. Wτ is a value scale of dimension √(J·s), not a currency unit, and the conversion to money runs through two declared conventions rather than through physics: the numéraire W0 (the benchmark per-capita annual negentropy output, Section 9) and the dimensionless consensus coefficient ατ set at the calibration layer (Section 8 and Section 10.2). Issuance Pτ = ατ·(Wτ/W0) is then a pure number of NVC units (one NVC unit is exactly the issuance generated by a benchmark activity: ατ = 1 with Wτ = W0 implies Pτ = 1), the dimension √(J·s) canceling at the numéraire step. The value equation thus ends exactly where the monetary unit begins, by design: the equation supplies a publicly recomputable ordering measure, the calibration layer supplies the unit of account, and no physical dimension is asked to do monetary work.

Table 1. Notation of the negentropic value equation.

Symbol	Name	Thermodynamic/informational meaning	Economic meaning
Ephysical (J)	Physical energy	Total energy input	Production energy consumption
ηex (0, 1]	Exergy efficiency	Orderly share of energy (0 < ηex ≤ 1)	Energy grade (bounded)
T0·ΔSeco (J)	Ecological entropy debt	Cumulative climate/ecosystem damage	Externality internalization
Ecognitive·ηneural (J)	Cognitive energy	Neural exergy conversion efficiency	Effective mental labor energy
||∇S|| (normalized)	Non-equilibrium gradient	Degree of deviation from equilibrium	Ordering potential
χ (0–1)	Flow index	Directed intensity of intervention	Effective labor intensity
Iτ (0–1)	Structural information gain	Entropy reduction relative to reference	Creativity and knowledge contribution
Vτ (J·s)	Negentropic action	Eτ·Tτ·Iτ (J·s; benchmark)	Aggregate value base
Wτ (√(J·s))	Negentropic value	√Vτ (benchmark); eq (1) in general	Monetary value scale

Table 1. Notation of the negentropic value equation.

Symbol	Name	Thermodynamic/informational meaning	Economic meaning
Ephysical (J)	Physical energy	Total energy input	Production energy consumption
ηex (0, 1]	Exergy efficiency	Orderly share of energy (0 < ηex ≤ 1)	Energy grade (bounded)
T0·ΔSeco (J)	Ecological entropy debt	Cumulative climate/ecosystem damage	Externality internalization
Ecognitive·ηneural (J)	Cognitive energy	Neural exergy conversion efficiency	Effective mental labor energy
||∇S|| (normalized)	Non-equilibrium gradient	Degree of deviation from equilibrium	Ordering potential
χ (0–1)	Flow index	Directed intensity of intervention	Effective labor intensity
Iτ (0–1)	Structural information gain	Entropy reduction relative to reference	Creativity and knowledge contribution
Vτ (J·s)	Negentropic action	Eτ·Tτ·Iτ (J·s; benchmark)	Aggregate value base
Wτ (√(J·s))	Negentropic value	√Vτ (benchmark); eq (1) in general	Monetary value scale

4.4. Measurement Error and the Precision Budget

The conventions of the measurement layer are auditable, but audits do not eliminate measurement error; they price it. A question the earlier text left open admits a sharp answer: what precision must the CRN deliver for the value equation to be usable at all?

(σW/W)² ≈ a²·(σE/E)² + (1−a)²·(σT/T)² + c²·(σI/I)² +
2a(1−a)·ρET·(σE/E)(σT/T) + 2ac·ρEI·(σE/E)(σI/I) + 2(1−a)c·ρTI·(σT/T)(σI/I)

(5)

Proposition 5 (error budget). Suppose the relative measurement errors of the three factors have magnitudes σE/E, σT/T, σI/I and pairwise correlation coefficients ρET, ρEI, ρTI. Then, to first order, the relative error of the computed value obeys (5). Under mutually independent measurement errors all correlations vanish and (5) reduces to the sum of squares; under the symmetric benchmark a = c = 1/2 it further reduces to σW/W = ½·√((σE/E)² + (σT/T)² + (σI/I)²). Proof. Log-linearize (1) as ln Wτ = ln k + a·ln Eτ + (1−a)·ln Tτ + c·ln Iτ, differentiate, and take the variance of the resulting sum; independence nullifies the covariance terms. □

Corollary (precision targets). Allocating the error budget equally across the three factors requires each factor’s relative error below 5.8% to keep the value error under 5%, and below 11.5% for a 10% tolerance. Two design consequences follow. First, the square-root benchmark is itself an error damper: each factor’s error passes into Wτ attenuated by its exponent, so the equation is least fragile exactly where its conventions are thickest. Second, the budget allocates audit effort numerically rather than rhetorically: since Iτ rests on the most convention-laden inputs, the implied precision targets for the information channel become binding engineering constraints on the L1–L2 layers of the CRN (Section 11.1). The measurement deficit, often invoked as a blanket objection, is hereby converted into a set of numerical tolerances that a metrology program can be engineered to meet — or can openly fail. One clarification about correlations is due. The T–I correlation ≥ 0.7 required at L1 (Section 11.1) is a correlation between true factor values, used as a cross-validation trigger, not a correlation between measurement errors; if the measurement errors themselves are correlated, the general form (5) applies with nonzero ρET, ρEI, ρTI, and the precision targets above tighten accordingly.

5. Civilization Utility Theory

5.1. The Civilization Utility Equation

ΔCUτ = λτ·ln(1 + Wτ/W0), CU = Σ β^τ−t·ΔCUτ (summed over τ = t,…,T), β = 1/(1+ρ)

(6)

The logarithmic form delivers monotonicity with diminishing marginal returns, in line with the utility-equation approach of Cheng and Cheng [10]; it also corrects the reciprocal form of earlier versions, which conflated utility level with marginal utility. The discount factor follows the Ramsey rule. With β = 0.95, a century-ahead generation would receive a weight of 0.006, contradicting the intergenerational-equity postulate, so the benchmark adopts the Stern value ρ ≈ 0.1% (β ≈ 0.999), and any policy simulation must report sensitivity over ρ ∈ [0.1%, 2%]. The choice between Stern [39] and Nordhaus [32] discounting is itself normatively loaded, a point that returns in Proposition 6.

5.2. Truth Mastery as Directional Goal

Definition 3 (truth mastery, heuristic construct). Θ(t) denotes the cumulative reduction of DKL(pbelief,t || pnature), the KL divergence between humanity’s belief distribution and the true structure of nature, understood as the gradual convergence of human knowledge toward the actual workings of the universe. Since pnature is not directly observable and scientific models are only its successive approximations, Θ(t) is a directional conceptual model rather than a measurable economic variable, and it must not be conflated with Eτ, Tτ, or Iτ. Its role is to anchor the cognitive dimension of civilization utility conceptually, not to serve as an accounting variable.

Proposition 6 (asymptotic dominance of the cognitive term, generalized). Let the instantaneous growth rate of truth mastery decay as g(t) = g0·t^−κ with κ ≥ 0, nesting geometric growth (κ = 0) and the stagnation regimes documented in the research-productivity literature [53], and let the per-period material contribution, entering through the logarithm, be bounded above by M̄ and below by m > 0 on any expanding path. Then: (i) if κ = 0, there exists a horizon T∗ beyond which the cognitive term dominates increments of CU if and only if g0 > ρ; (ii) if κ > 0, the cognitive increment g(t) tends to zero while the material increment remains bounded below, so asymptotic dominance fails for every discount rate — the cognitive share of CU is substantial only over a finite window whose length grows with g0 and shrinks with κ; (iii) under κ = 0 the discounted cognitive sum diverges whenever g0 > ρ, recovering the sharper earlier formulation. Proof sketch. (i) and (iii) compare e^(g₀−ρ)t with a bounded term and sum the geometric series directly. (ii) For κ > 0 the ratio of increments g(t)/m tends to zero, which inverts the dominance ordering beyond some T∗. □

The generalization changes the proposition’s rhetorical status, deliberately. Bloom et al. [53] document that research productivity across semiconductors, agriculture, and pharmaceuticals has fallen sharply: maintaining a constant growth rate of ideas has required ever-rising research effort, which in the present notation is a secular drift of κ away from zero. Under clause (ii), asymptotic cognitive dominance is therefore not a law of the system but a target of policy. The high-α configuration for basic research (Proposition 9) is precisely the instrument that defends a low-κ trajectory against the documented drift: by raising the share of effort allocated to fundamental research, the calibration layer works to keep g(t) from decaying. The dominance proposition and the research premium are one policy stated twice, and the window T∗ of clause (ii) measures how much time the instrument has.

Corollary (cognitivization of the monetary function). Under clause (i) of Proposition 6, the meta-mission of the monetary system is to maximize the rate of knowledge production. This is the utility-theoretic ground for assigning basic research its highest α in CPPA: its current-period Wτ is small, but the discounted sum of its truth contribution dominates the asymptotic expansion. Together with Proposition 9 (risk compensation) it forms the dual foundation of high α for basic research — and clause (ii) adds the asymmetry that this foundation must be maintained, not merely asserted.

5.3. Assessing Individual Truths

Because Θ(t) as a whole is not directly measurable, we propose a per-truth accounting path. For each cosmic truth mastered, its transmitted contribution to humanity’s negentropy acquisition is assessed by the counterfactual-worlds method: construct a parallel world without that theory and compare order stocks, ΔNj = ∫ [Wactual(t) − Wcounterfactual(t)] dt, transmitted through the chain truth → technology → efficiency, output, and survival gains.

Example A (relativity). Relativity → satellite clock correction (GPS clocks drift ≈ 38 μs/day, which if uncorrected would produce navigation errors of ≈ 10 km/day) → precision agriculture with variable-rate seeding and autonomous machinery, global logistics optimization, and timed financial networks. Counterfactual estimates place relativity’s direct economic contribution in the tens of billions of dollars annually, with precision dividends still compounding.

Example B (quantum mechanics). Quantum theory → transistors, semiconductors, lasers, MRI, atomic clocks. Estimates attribute 25–35% of advanced-economy GDP to semiconductor and laser-related industries; without quantum mechanics the order stock of the information civilization would shrink by roughly one third.

Example C (germ theory). The Pasteur–Koch theory → vaccines and antibiotics → human life expectancy rising from 40 to 73 years. Each additional year of global working life adds an effective Tτ stock of roughly 5×10⁹ people × 2,000 hours per year, a truth-level amplifier of the T dimension.

Methodological status: the per-truth ΔNj are approximate estimates, since the counterfactual world cannot be truly constructed and the method relies on inference from technology and economic history. Every link, however, carries an auditable transmission-evidence chain, which makes the path a bridge from a non-measurable Θ(t) to one that is approachable truth by truth.

5.4. Falsifiable Propositions

H4 (accountability of truth contributions). The counterfactually estimated negentropy contributions ΔNj of front-rank scientific theories correlate significantly with their subsequent patent clusters, industrial value added, and life-expectancy gains. Counterfactual thought experiments cannot be observed directly, so H4 functions as a case-assessment tool for major theories rather than a large-sample statistical test; its evidential form is deep case chains, not regression coefficients. Falsification condition: if the ΔNj estimates of most front-rank theories decouple severely from their actual technological derivatives, the per-truth accounting path fails. Test path: systematic counterfactual accounting of the ten great physical theories of the twentieth century, cross-validated against cliometric data.

Case-quality checklist. Evidence for H4 must satisfy five criteria: (1) auditable transmission chain, with independent historical or engineering evidence at every link; (2) counterfactual conservatism, crediting no contribution to substitute technologies that would plausibly have emerged anyway, with all assumptions listed and reverse-scenario-tested; (3) triangulation across at least two independent data types; (4) interval reporting, with interval width reflecting counterfactual uncertainty; (5) double-background peer review, one reviewer in the history of science and one in economics. Cases failing criteria (1)–(3) do not enter the H4 test.

A minimal illustration may fix ideas. Consider the polymerase chain reaction (PCR). Link 1 (transmission): the theoretical chain from DNA structure discovery to thermostable-polymerase engineering is documented in the Nobel citation and the laboratory record, satisfying criterion (1). Link 2 (counterfactual conservatism): had PCR not emerged, thermally cycled cloning would have remained slower and costlier, but substitute amplification methods existed by the late 1980s; the counterfactual therefore credits PCR only with the cost differential and the time advantage, an explicitly conservative treatment. Link 3 (triangulation): the value interval is cross-checked against (a) the size of the molecular-diagnostics and forensic-testing industries and (b) citations and diffusion data in clinical guidelines. Reporting takes the form of an interval, say ΔNPCR ∈ [a, b], whose width reflects the uncertainty over substitute technologies. The case then proceeds to double-background review. Larger cases (relativity, quantum mechanics) follow the same protocol with longer chains.

H5 (dynamic cognitive-weight proposition). In CPPA simulations, as material scarcity relaxes, the α-weight on basic research under civilization-utility maximization should rise continuously. Falsification condition: if the optimal basic-research α fails to rise under continuously relaxing scarcity, the dynamic version of the dominance proposition is weakened. Test path: long-horizon ABM embedding Proposition 6.

5.5. Co-Evolution of Civilization and Individuals

At the individual level, the accumulation of five-dimensional life capital (physiological, cognitive, emotional-relational, ethical-meaning, civilizational-action), forms of capital whose economic payoff is documented in the social-capital literature (Knack and Keefer [27]), is the order increment of the individual system, and the NVC system guides individuals toward high-return domains of that capital through the α-coefficients, a transformation from possessive economic man toward creative civilizational man. At the civilization level, four transitions unfold: competitive civilization (stock games), creative civilization (incremental cooperation), flourishing civilization (post-scarcity abundance), and truth civilization (systematic expansion of the cognitive frontier). Individual upgrading and civilizational transition are cause and effect of each other, a bidirectional positive feedback forming the twin engines of civilizational ascension. The CPPA loop closes the system: CRN measures Wτ and ΔΘ (where value comes from); the civilization utility equation sets the CU objective (where value goes, asymptotically toward truth mastery); CPPA couples monetary issuance through Pτ = ατ×Wτ (how to incentivize).

6. Static Microfoundations and the Conditional Decentralization Result

The economy is a continuum of agents of measure one. Agent heterogeneity is captured by creative capacity θi and time endowment L̄. There is a set of activities j = 1,…,J; agent i’s effort lji in activity j produces Wji = θi·gj(lji) with gj′ > 0 and gj″ < 0. The consensus coefficient of activity j is αj, income is yi = Σj αjWji, and the effort cost c(Σj lji) satisfies c′ > 0, c″ > 0.

The agent solves max Ui = u(yi) − c(Σj lji), while the social planner chooses a configuration to maximize weighted social utility. One assumption, left implicit in earlier versions, must now be made explicit:

Assumption A6 (quasi-linearity over the relevant range). The utility u is linear in income (u′ ≡ 1 after normalization) for all agents, so that income effects on effort allocation are absent. The assumption is strong, and the cost of relaxing it is known. With strictly concave u and heterogeneous incomes, the marginal utility of income u′(yi) differs across agents, and no single αj can align every agent’s first-order condition with the planner’s at once; alignment then requires agent-specific transfers or the nonlinear correction α̃ij = αj/u′(yi). We leave this extension to future work and flag it again in Section 15.1.

Proposition 7a (existence). Under continuity–concavity of u, strict convexity of c, continuity–concavity of gj, and a nonempty compact feasible set, both problems have solutions. The feasible set is weakly compact and convex in a finite-dimensional Banach space, the objective is upper semicontinuous in the weak topology, and a maximum exists by the Weierstrass extreme-value theorem; individual existence follows from the Berge maximum theorem.

Proposition 7b (uniqueness). If u is strictly concave, or c strictly convex, the optimal configuration is unique almost everywhere. Suppose otherwise that two distinct optima exist; their strict convex combination yields strictly higher expected utility, a contradiction.

[Conditional reminder] The results in this section rest on two premises throughout: the lexicographic normative premise of Section 3 is accepted, and CRN measurement bias sits at its endogenous-efficiency-equilibrium level (Section 10.1). Failure of either premise weakens the conclusions accordingly.

Proposition 7c (conditional decentralization of the planner’s optimum). Let αj be the shadow price of activity j’s output in the planner’s problem. Under Assumption A6, if αj = αj∗ for all j, the decentralized equilibrium configuration coincides with the social optimum. Proof: under A6 the planner’s first-order condition reads αj·θi·gj′(lji) = c′, and the agent’s reads αj·θi·gj′·u′(yi) = αj·θi·gj′ = c′. Setting αj = αj∗ makes the two systems identical, and by Proposition 7b the solutions coincide. Q.E.D.

The scope of this proposition deserves a candid statement. It is neither a spontaneous-order theorem nor an analogue of the First Welfare Theorem. Structurally, it restates the standard result that Pigouvian shadow prices, once implemented as income coefficients, decentralize the planner’s optimum (Arrow and Debreu [1]; Debreu [15]). Whatever is novel in this paper lies upstream of that result: the CRN measurement layer (Section 11) supplies observable shadow-price signals for activities that market prices leave unmeasured, among them basic research, care work, open source production [23], and ecosystem restoration. The proposition formalizes the NVC claim that micro-incentives can be unified with the macro-mission; the measurability program, rather than the decentralization logic, is the contribution.

Proposition 7d (welfare-loss bound). Suppose CU(α) is C³ and strictly concave in a neighborhood of an interior optimum α∗, so that ∇CU(α∗) = 0. A second-order Taylor expansion gives, for the calibration error ε ≡ α − α∗,

CU(α∗ + ε) − CU(α∗) = ½·Σj εj²·(∂²CU/∂αj²)|α=α∗ + O(||ε||³)

(7)

Defining Λj ≡ −(∂²CU/∂αj²)|α=α∗ ≥ 0 by concavity, the welfare loss satisfies ΔCUloss ≤ ½·Σj εj²Λj + O(||ε||³). Calibration error is therefore locally a quadratic form: small deviations are cheap, which licenses gradual pilots; large deviations become rapidly costly, which gives the α-protection mechanism its urgency; and activities with high Λj demand the greatest calibration precision. This yields CPPA’s operational objective, namely to minimize Σj εj²Λj each period. The bound is local by construction, and its global use requires the stated smoothness and interiority conditions; in practice Λj can be estimated from the ABM simulator’s response surface.

Two reference points locate the argument. Relative to the First Welfare Theorem, the α here is not a market-clearing price but an algorithmically calibrated consensus coefficient, and negentropic creation carries intertemporal spillovers, so FWT does not apply directly. Relative to Greenwald–Stiglitz [20], the paper is a structural repair of precisely the failure they identified: CRN converts unobservable value creation into verifiable data flows and thereby compresses the information asymmetry that makes market allocations constrained-inefficient. Once measurement is repaired, α-calibration moves the economy toward the efficiency frontier. The framing suggests a third category beside market failure and government failure, namely measurement-infrastructure failure, on the hypothesis that both classical failures share a common root in the non-measurability of large classes of value creation. We offer this as a research hypothesis rather than an established doctrine.

Proposition 7e (income-effect misalignment bound). Drop Assumption A6 and let u be strictly concave, with |u′(yi) − 1| ≤ μ over the relevant income range. Implementing the planner’s shadow prices αj = αj∗ then distorts each agent’s effort by a factor bounded in proportion to μ, and the induced welfare loss satisfies ΔCUloss ≤ ½·μ²·Σj (αj∗)²·Λj + O(μ³). Proof sketch. The agent’s first-order condition αj∗θigj′(l)·u′(yi) = c′ differs from the planner’s by the factor u′(yi); implicit differentiation gives |dl/l| ≤ μ/(εc + εg), with εc, εg the local elasticities of marginal cost and marginal product; inserting the resulting effective calibration error εj ≈ μ·αj∗ into the quadratic form of Proposition 7d yields the bound. □ The content is quantitative reassurance with a boundary: quasi-linearity error enters only quadratically, and only through income dispersion. Compressing income dispersion compresses μ and hence the misalignment loss — redistribution, in this framework, is simultaneously an efficiency instrument.

7. Life-Cycle Dynamics: Learning Effects and Uncertainty

θi,middle = θi,youth·(1 + δe·le,i), δe > 0

(8)

The demographic structure is a three-period OLG model in the tradition of Diamond [16]: youth (education), middle age (negentropic creation), and old age (consumption). Proposition 8 (learning effect). g ≈ δe·(le/L̄)·(output share of high-α activities). Higher educational investment raises capacity accumulation and growth, and a high α on knowledge-intensive activities generates intergenerational growth dividends through θ-accumulation. The learning parameter δe is at present a theoretical magnitude; its micro-estimation belongs to the future agenda.

αj·E[Wj] ≥ αsafe·Wsafe + (γ/2)·αj²·Var(Wj)

(9)

Proposition 9 (risk-compensating α). Risky activities command a risk premium πj = (γ/2)·αj·Var(Wj)/E[Wj]. Derivation. With mean–variance preferences Uj = αj·E[Wj] − (γ/2)·αj²·Var(Wj), participation in risky activity j rather than a safe alternative requires Uj ≥ αsafe·Wsafe; rearranging yields the participation inequality displayed above, and dividing through by E[Wj] yields the premium. For near-binary research outcomes with success probability p, Var(Wj) ≈ p(1−p)·Wj², so the required premium scales as p(1−p)·Wj. The CPPA range α = 3.0–4.0 for basic research, where p ≈ 0.05–0.3, is therefore an endogenous requirement of risk compensation rather than a political preference — and the size of the premium is now computable from observables rather than asserted.

Proposition 10 (dynamic negentropy proposition). When αj,t equals shadow prices and θ-accumulation is endogenous, the decentralized equilibrium path attains the planner’s optimum. The argument applies Proposition 7c to each static sub-problem and adjoins the accumulation law, value-function concavity, and a transversality condition. It supplies the intergenerational-efficiency foundation of the Civilization Time Bank.

8. General Equilibrium: Endogenizing α

The full system is a fixed-point problem α = Φ(α): individual reaction functions aggregate into the labor-allocation map, quadratic voting aggregates preference distributions into Θ(t), and CPPA maps Θ(t) into α.

Proposition 11 (existence). Under continuity of the constituent maps and a compact convex range, an equilibrium α exists. Φ is an upper-hemicontinuous, convex-valued correspondence on a compact convex set, and the Kakutani fixed-point theorem applies.

Proposition 12 (uniqueness and stability). If labor-supply responses to α are strictly monotone, which strict convexity of c guarantees, and if QV aggregation satisfies the monotone likelihood-ratio property, then α is unique. Under the error-correction rule Θ(t+1) = Θ(t) + η·(shadow-price estimate − α(t)), the system is locally asymptotically stable: the Lyapunov function V = Σj(αj − α∗)²Λj decreases along the adjustment path, which follows from the quadratic structure of Proposition 7d. The Lyapunov function coincides with CPPA’s optimization objective, so the calibration rule is at once optimal and stable. Both uniqueness and stability rest on strong assumptions, and the next paragraphs trace what happens when they fail.

Scenario 1 (QV monotone likelihood ratio fails). Multimodal preference clustering turns the aggregation map Γ set-valued, and the fixed-point system may admit multiple equilibria sustained by coalition expectations. The welfare-loss bound survives, being a local property, but global optimality is no longer guaranteed. The institutional remedy is the mechanical randomness of sortition citizens’ assemblies, which breaks preference clusters by construction (the full institutional design is the subject of the companion governance paper [61]).

Scenario 2 (non-monotone labor supply). If learning effects bend an activity’s labor-supply curve backward, the reaction map φ ceases to be monotone and stable manifolds may split into periodic orbits, so α-adjustment oscillates. Proposition 12’s asymptotic stability degrades to bounded oscillation, the oscillation amplitude becomes an additional variance source for εj, and the quadratic welfare-loss structure makes the cost grow with the square of the amplitude. Conservative adjustment steps η are the quantitative response.

Scenario 3 (capture parameter-domain violation). If lobbying costs collapse or supervision fails, the capture equilibrium re-emerges and εj becomes systematic bias rather than zero-mean noise. Losses then accumulate as Σ εj²Λj, and persistent divergence between λτ and α serves as the early-warning signal for emergency review by the Audit and Emergency Adjustment Committee (AEAC; the committee’s appointment, powers, and capture-resistance are specified in the companion governance paper [61]).

Across the three scenarios, assumption failure degrades the system into a diagnosable, warnable, and intervenable condition rather than destroying it, which is the practical meaning of the scenario analysis.

9. Application Examples: The Value-Discriminating Function of the Equation

Let the benchmark per-capita annual negentropy output be W0 = 5×10⁷ √(J·s). All examples are computed from the benchmark equation Wτ = √(Eτ·Tτ·Iτ) with the parameters stated in each example, so that every figure can be reproduced from the stated inputs. [Conditional reminder] The discriminating structures below are corollaries of the lexicographic premise: a “value vacuum” (Wτ → 0) means that the monetary incentive system offers no reward, not that the activity should be prohibited.

Example 1: fundamental theoretical breakthrough. Eτ = 4.3×10⁹ J, Tτ = 1.2×10⁷ s, Iτ = 0.95. Then Wτ = √(4.3×10⁹ × 1.2×10⁷ × 0.95) ≈ 2.2×10⁸ √(J·s) ≈ 4.4W0. The extraordinary value is carried by Iτ; creative value is a direct manifestation of the informational dimension, and its discounted truth contribution exceeds the current-period Wτ by Proposition 6.

Example 2: care labor. Eτ = 5×10⁷ J, Tτ = 4×10⁶ s (χ ≈ 0.5 with ||∇S|| persistently significant), Iτ = 0.25. Then Wτ ≈ 7.1×10⁶ √(J·s) ≈ 0.14W0. Labor that market GDP renders invisible receives a definite accounting measure, and the implied care-to-physicist value ratio, roughly 1:31, follows from the stated physical quantities rather than from any political assignment.

Example 3: open-source database core module. Eτ = 4×10⁸ J, Tτ = 2×10⁶ s, Iτ = 0.80, giving Wτ ≈ 2.5×10⁷ √(J·s) ≈ 0.51W0. The free-rider dilemma of knowledge public goods is structurally mitigated once the informational dimension enters the accounting.

Example 4: solar versus coal generation (10⁹ J each). Under Section 4.2 both sources obey 0 < ηex ≤ 1, and the discrimination enters through the ecological entropy debt. Solar: ηex = 0.30, the second-law efficiency of photovoltaics, with T0ΔSeco ≈ 0, so Eτ = 0.30×10⁹ J. Coal: ηex = 0.55, a high conversion efficiency, but with a cumulative ecological entropy debt T0ΔSeco = 0.35×10⁹ J, so Eτ = 0.20×10⁹ J. With Tτ and Iτ equal across the two, Wτ is proportional to √Eτ, and Wτ,solar/Wτ,coal = √(0.30/0.20) ≈ 1.22; equivalently, coal’s value output runs about 18% below solar’s at equal generation (1/1.22 ≈ 0.82). Carbon costs are internalized ex ante through the audited entropy debt, complementing rather than replacing carbon taxation.

Example 5: high-frequency trading. Eτ = 8×10⁸ J, but χ ≈ 0, ||∇S|| ≈ 0, and Iτ ≈ 0, so Wτ → 0. The criterion discriminates rather than negates: financial activities that supply price-discovery information structures or risk management earn nonzero Iτ and remain in the accounting.

Example 6: wetland ecological restoration. Eτ = 3.3×10⁸ J, Tτ = 5×10⁶ s, Iτ = 0.85 (diversity index rising from 0.3 to 0.75), giving Wτ ≈ 3.7×10⁷ √(J·s) ≈ 0.75W0. Ecosystem services become measurable as structural entropy reduction.

Example 7: machine-tool manufacturing. Eτ = 5.6×10⁸ J, Tτ = 1×10⁶ s, Iτ = 0.45, giving Wτ ≈ 1.6×10⁷ √(J·s) ≈ 0.32W0. Manufacturing value is contributed by all three factors in balance, though at the stated scales it sits below the per-capita benchmark. The result illustrates the equation’s discrimination among production modes rather than any privilege of manufacturing by default.

Three features of Table 2 deserve notice. Value sources are highly differentiated, which answers the classical problem of heterogeneous-labor commensuration on the basis of public measurement conventions. The blind spots of traditional metrics (care, open source, ecology) come into view, and together they exceed manufacturing in aggregate. Value vacua are identified by the accounting rule rather than politically decreed. Cross-example ratios depend on the stated parameter settings; the claim concerns the discriminating structure, not any specific ratio. All ratios in Table 2 are computed under the symmetric benchmark Wτ = √(Eτ·Tτ·Iτ); by Proposition 4, general calibrated exponents rescale these ratios but preserve every ranking whose value gap satisfies 2|Δa| + |Δc| < ρ/L, and rankings between near-tied rows are correspondingly sensitive to calibration.

10. Mechanism Design: Incentive Compatibility, Measurement Cost, and Defense

10.1. Endogenizing Measurement Cost

Cm′(K) = |∂CU/∂σ²|·|dσ²/dK|

(10)

Optimal audit intensity equates the marginal audit cost to the marginal welfare benefit of reduced measurement error. Measurement is not free, and the theorem’s validity condition must be restated accordingly: what is required is CRN bias at the endogenous-efficiency-equilibrium level, not the stronger and unrealistic requirement of zero bias. The same condition prices CRN’s budget.

10.2. The CPPA Iterative Algorithm

αj(t+1) = αj(t) − η(t)·[αj(t) − α̂j(t)], Ση(t)=∞, Ση²(t)<∞

(11)

Under the Robbins–Monro [34] conditions the algorithm converges to a neighborhood of α∗ with probability one in the presence of measurement noise. In implementation the ABM simulator generates shadow-price estimates α̂ each period, while real-time λτ and Wτ data trigger calibration quarterly; decoupling adjustment frequency from data-update frequency guards against noise-driven overreaction.

10.3. Manipulation Cost

An attacker inflating Wτ at manipulation intensity q faces a success probability P(q,K) that decreases convexly in audit intensity K and a cost C(q) increasing convexly. Expected profit is E[π] = P(q,K)·q·W·α − C(q) − r·(1−P(q,K))·M. When the stake M ≥ W·α/r, deterrence covers the gains even at high P(0,K). In the CRN baseline parameter domain (three-level audit pass-rate product ≤ 10⁻³, 100% staking, r = 1), expected manipulation profit is strictly negative.

10.4. Elite-Capture Game

s = argmax [ρ(d)·R(s) − s], d = argmax [(1−ρ(s))·ΔW − d]

(12)

Multiple independent checks (staggered AEAC terms, random mutual audit, citizen petition, and QV) raise capture costs and flatten ∂ρ/∂s until the optimal lobbying effort s approaches zero, making capture uneconomic within a declared parameter domain. Out-of-domain risks are listed in Section 15.1.

10.5. Wealth Influence in Quadratic Voting

The standard criticism of quadratic voting (Lalley and Weyl [30]; Weyl [42]) is that quadratic costs admit wealth into voting power. The mitigation distributes voting credits equally per capita, makes them non-transferable, and grants a free base allowance, with quadratic pricing applying only to purchases beyond the base and a public fund subsidizing expression allowances for low-income citizens. The mechanism reverts to one base credit per person plus voluntary top-ups, confining wealth influence to the marginal purchase layer.

10.6. Expression Layer, Navigation Layer, and the Knowledge Problem

A familiar objection runs as follows: humanity treasures many things that add little systemic order (entertainment, symbolic luxury, pure pastime), and the willingness to pay for them is real. The two-layer structure is designed to absorb this objection. Subjective preferences are fully respected at the expression layer: willingness to pay for entertainment enters through λτ, and CPPA’s consensus generation includes the public’s weighting of diverse ways of life, so the entertainment industry’s α need not be zero and is set democratically. Preferences are calibrated, not overruled, at the navigation layer: the monetary system’s meta-mission of civilizational survival and truth mastery (Section 3 and Section 5) depends on net order increments, and this is a response to the collective-action problem rather than a denial of individual preference, since the aggregate of individually rational choices can be civilizational-scale irrationality (the fallacy of composition). Precedent exists: subjective value theory never held that whatever people want should receive equivalent monetary-issuance rewards, and both central-bank asset-bubble suppression and tobacco taxation respect preferences while declining to reward them through the monetary system. The proposal codifies this intuition by separating the channel of preference expression (the market) from the channel of civilizational navigation (α): the former free, the latter calibrated.

A deeper objection must now be faced rather than deferred. Hayek’s [48] knowledge problem does not disappear when the accounting unit becomes physical: the information that coordinates production is dispersed, tacit, and in large part generated only through the price process itself. If the navigation layer sets α for broad activity classes while the expression layer clears concrete exchanges, then the everyday work of allocation still runs through market prices — the framework never claimed otherwise. What the negentropic anchor removes is discretionary issuance; what it cannot remove is the epistemic burden of knowing what to produce. Three design features keep that burden within bounds. First, α operates on activity classes, not individual commodities, so the calibration problem is coarser by several orders of magnitude than central planning. Second, the expression layer’s realized exchange data feed back into CRN as one audit channel: persistent divergence between measured Wτ and revealed willingness to pay is itself a review trigger under the L2 rules of Section 11. Third, the welfare-loss bound of Proposition 7d quantifies exactly how expensive residual calibration error is, and converts “how much do we need to know?” into an estimable quantity. None of this dissolves Hayek’s objection; it relocates the objection from the issuance layer, where we claim it is answered, to the calibration layer, where we claim only that it is bounded, priced, and auditable. That is the honest division of labor between the two layers, and the residual gap is recorded in Section 15.1.

11. The CRN Blueprint: From Research Program to Engineering Design

CRN is a forward-looking research program for technological infrastructure. Its measurement layer (large-scale flow-state sensing, cognitive-exergy metering, cross-domain knowledge graphs) depends on sensing and AI technologies that are not yet mature, and we make no claim of present-day deployability. What we claim is that the architecture is engineering-deducible and that every measurement convention it requires is publicly documentable and auditable in principle.

11.1. Four-Layer Closed Loop

L1 (perception): Eτ through the device–compute–energy IoT metering model plus information-complexity energy-equivalence; Tτ through multimodal biosensing with behavioral-log cross-validation and zero-knowledge proofs (raw data remain local and only existence proofs are uploaded, data usable but not visible); Iτ through domain-specific indicator systems with Cronbach α ≥ 0.8 internal consistency and historical-benchmark external cross-validation, where a 15% deviation triggers review; and post-quantification synergy validation (T–I correlation ≥ 0.7) against single-factor manipulation. L2 (audit): rule-based first-pass, cross-source validation, and interpretive-community review by peers, stakeholders, and ethicists, reinforced by stake-and-penalty game mechanics. L3 (α computation): a five-module algorithm (attribute extraction, dimension mapping, weight fusion, dynamic calibration, verification and audit), confirmed by at least five independent node clusters under Byzantine fault tolerance and written to the chain. L4 (incentive): smart contracts execute NVC = W×α; the macro-smoothing coefficient ks (0.95–0.98) buffers issuance; and feedback drives periodic parameter optimization, making the system a learning system.

11.2. End-to-End Demonstration: The Global Brain Science Program

In the demonstration, L1 collects green-power server energy and itemized metering of fMRI, EEG, and optogenetics rigs, computes flow-time via trusted-execution-environment privacy computing, and quantifies Iτ for theoretical models. L2 cross-validates energy reports against laboratory logs, with interpretive-community review of narrative consistency for high-controversy results such as theories of consciousness. L3 maps attributes to S = 0.95, E⁺ = 0.97, C = 0.90, U = 0.85 and reaches a consensus α = 0.95; basic research enters the above-0.9 regime because its output flows across regions and generations and creates sustained civilizational-scale ordering effects whose consensus base transcends market-interest bargaining. L4 disburses NVC = W×α to institutions, researchers, and infrastructure providers by weight (theoretical breakthroughs by Iτ, technical solutions by Tτ, hardware by Eτ), closing a negentropy creation–value return–re-creation cycle. The illustrative values of S, E⁺, C, and U are calibration placeholders pending the estimation procedure of Section 10.2 rather than empirically established magnitudes.

12. Empirical Evidence

12.1. Cross-Country Regression (Component-Level Evidence)

The data are the World Development Indicators [62] for 40 major economies over 2010–2019. The variables are R&D expenditure as a share of GDP (a proxy for the I-dimension input; Coe and Helpman [11]), GDP per capita, resident patent applications, and population; ten-year country means form the cross-section, and the dependent variable is log patent density (patents per million people; Griliches [21]). t-statistics are heteroskedasticity-robust, so for single-regressor models R² need not equal t²/(t² + df) exactly; the table should be read with this reconciliation note. Whatever the results, this evidence alone can neither confirm nor falsify the value posits, and interpretation is confined to the component-correlation level. We add one further discipline: the knowledge production function literature has documented the R&D-to-patent link for four decades, so the regressions below are offered as a consistency check on our proxy construction and sample, not as independent discovery. Their value to the program is negative as much as positive — had the I-dimension proxy failed to line up with innovation output even at this crude level, the operationalization of Section 4.2 would already be in trouble.

Table 3. Cross-country regressions (dependent variable: log patent density).

Model	Regressors	R²	Key coefficients (t, p)
M1	ln GDP per capita	0.699	1.527 (10.7, 2e-12)
M2	NE composite score	0.223	0.650 (3.3, 0.002)
M3	NE + ln GDP per capita	0.724	NE: 0.270 (1.6, 0.12); GDP: 1.216 (5.2, 6e-6)
M4	ln R&D intensity	0.790	2.135 (13.4, 3e-15)
M5	ln R&D intensity + ln GDP per capita	0.810	R&D: 1.969 (9.9, 1e-11); GDP: 0.169 (0.7, 0.51)

Table 3. Cross-country regressions (dependent variable: log patent density).

Model	Regressors	R²	Key coefficients (t, p)
M1	ln GDP per capita	0.699	1.527 (10.7, 2e-12)
M2	NE composite score	0.223	0.650 (3.3, 0.002)
M3	NE + ln GDP per capita	0.724	NE: 0.270 (1.6, 0.12); GDP: 1.216 (5.2, 6e-6)
M4	ln R&D intensity	0.790	2.135 (13.4, 3e-15)
M5	ln R&D intensity + ln GDP per capita	0.810	R&D: 1.969 (9.9, 1e-11); GDP: 0.169 (0.7, 0.51)

Three findings stand out. The I-dimension proxy alone explains more variance (R² = 0.790) than GDP per capita (0.699). In the joint model M5, R&D intensity remains highly significant (t = 9.9, p ≈ 10⁻¹¹) while GDP per capita turns insignificant (p = 0.51): once the knowledge input to value creation is measured directly, the explanatory power of the traditional income variable is absorbed. Robustness checks exclude the patent-density outlier (Israel), whereupon the R&D coefficient is 1.849 (p = 1.1×10⁻⁷), and the NE composite shows a significant quartile gradient with a top-to-bottom median ratio of about 19.

12.2. What These Regressions Cannot Establish: Identification, Proxy Distance, and the Limits of Cross-Country IV

Two caveats limit the exercise, and this revision states them more sharply than earlier versions did, because they define the boundary between what is claimed and what is not.

The first caveat is the proxy distance. The theory’s Iτ is a KL-divergence reduction relative to a field’s reference knowledge distribution (eq (4)); the regression’s proxy is R&D expenditure intensity, an input measure that is at best monotonically related to the true construct. A significant coefficient therefore validates the weakest link of the proxy chain — that the input side of the I-dimension covaries with innovation output — and says nothing directly about the KL form, the reference distribution, or the necessity structure of eq (1). Upgrading the proxy toward the construct is a more productive investment than upgrading the estimator, and it defines the first task of the extension protocol below: measures of knowledge recombination computed from patent citation networks (the cross-class dispersion of a country’s cited technology classes, a Herfindahl-type concentration index over IPC classes in PATSTAT or equivalent microdata) are structurally closer to Iτ than expenditure shares, because recombinant novelty is precisely a distributional shift relative to an existing knowledge base. A proxy that mirrors the functional form of the construct would convert the consistency check into something closer to a measurement validation.

The second caveat is identification. R&D intensity is endogenous: reverse causation (innovative economies can afford and demand more research), omitted variables (institutional quality, human capital, agglomeration), and common trends all bias the cross-sectional coefficient in unknown directions. The standard remedy — instrumental variables — is available in principle but fragile at the cross-country level in ways that deserve explicit statement rather than routine deployment. Candidate instruments for national R&D intensity (legal-origin dummies, historical university density, distance-to-frontier interactions, tax-credit generosity) face exclusion-restriction objections that are difficult to defeat, since each plausibly affects patenting through channels other than current R&D expenditure; and with 40–60 observations, weak-instrument bias is first-order rather than a technicality. We therefore deliberately do not report an IV specification here: a weakly identified IV estimate would confer a spurious causal gloss on what is, and is honestly labeled as, a consistency check. The companion empirical paper will instead pursue identification through within-country variation where the exclusion restriction is defensible — changes in R&D tax-credit schedules as natural experiments, in the tradition of the recent tax-incentive literature, and lag structures in a panel setting — and will report the executed H2 experiment alongside.

12.3. A Pre-Registered Extension Protocol

To make the consistency check improvable rather than rhetorical, we commit here to the following extension, to be executed and reported in the companion empirical paper whatever its results. (i) Sample: all economies with non-missing WDI data on R&D intensity, resident patents, GDP per capita, and population over 2010–2019 — we project 60–80 economies rather than 40 — with the estimation sample and exclusion rules fixed before any coefficient is examined. (ii) Proxies: the baseline I-dimension proxy (R&D intensity) plus two construct-nearer proxies — scientific and technical journal articles per million inhabitants (WDI), and the knowledge-recombination concentration index from patent microdata described in Section 12.1a; the primary specification is fixed ex ante as M5 run jointly on all three proxies, with no selective reporting across proxies. (iii) Robustness: exclusion of the patent-density outlier, winsorization at the 1st/99th percentiles, population-weighted variants, and a panel variant with country and year fixed effects on the full time series rather than ten-year means. (iv) Decision rule: the consistency check is considered to survive if the I-dimension proxies retain sign and significance jointly under (iii); if they do not, the operationalization of Section 4.2 — not the data — is the first object of revision. (v) Identification: causal claims, if any, will rest solely on the within-country natural-experiment designs of the companion paper, never on the cross-section reported here. The protocol is registered in this text so that the outcome cannot be curated post hoc.

12.4. Calibration of Accounting Blind Spots

Taken together, the three items are of the same order of magnitude as global GDP (≈US$105 trillion): scarcity-exchange accounting omits real value creation of market-comparable scale, which is directional evidence for the discrimination hypothesis.

Table 4. Independent scale estimates of accounting blind spots.

Activity	Value in market accounting	Independent scale estimate	Source
Unpaid care work	Zero in SNA	≈US$11 trillion/yr (≈9% of global GDP)	UN Women [41]
Open-source software	Price ≈ 0	≈US$8.8 trillion (demand-side)	Hoffmann et al. [23]
Ecosystem services	Zero	≈US$125–145 trillion/yr	Costanza et al. [12]

Table 4. Independent scale estimates of accounting blind spots.

Activity	Value in market accounting	Independent scale estimate	Source
Unpaid care work	Zero in SNA	≈US$11 trillion/yr (≈9% of global GDP)	UN Women [41]
Open-source software	Price ≈ 0	≈US$8.8 trillion (demand-side)	Hoffmann et al. [23]
Ecosystem services	Zero	≈US$125–145 trillion/yr	Costanza et al. [12]

12.5. Test Designs for H2 and H3

H2 (incentive effect) is a 2×2 between-subjects online experiment (α subsidy 1.0 versus 2.5 crossed with a basic-research task versus an applied task), with power analysis d = 0.5, power = 0.8, n ≈ 64 per cell, and pre-registration on OSF. H3 (anchor stability) uses the JST Macrohistory Database (1870–2020, 18 countries; Jordà et al. [25]) with three-anchor-regime Bai–Perron [3] break tests, war and crisis dummies, and a synthetic-control alternative. Execution of H2 is inexpensive and is a precondition for any revision claiming empirical support for the incentive layer; we commit to reporting it in the companion paper.

13. Macrostability: The Central Open Problem

A sketch of the stabilization mechanics illustrates the intended logic. Under an issuance anchor, ΔM is proportional to ΣWτ and the price level anchors to the ratio of order increments to money supply, with inflation redefined as the system entropy-increase rate (false negentropy exceeding true negentropy) to be treated by CRN audit and consensus repair. Under exchange-rate dissolution, a unified value scale removes the basis for exchange rates and cross-border settlement becomes clearing. Countercyclical buffers combine the endogenous metabolism rate δ, countercyclical GPF investment, and Time-Bank savings smoothing, with ks buffering technology-jump shocks. Crisis response contracts the T-horizon under extreme shocks, triggering emergency α-recalibration with an AEAC emergency brake and ex-post ratification by the Global Value Parliament (GVP).

Earlier versions of this program relegated these mechanisms to a closing sketch. The present version refuses that comfort, because this is the point at which a scarcity-free anchor confronts the hardest evidence of monetary history: anchor drift, speculative attacks on pegs, and the stabilization failures of non-scarcity monies. Whether an issuance rule proportional to audited order increments can deliver a stable unit of account, and under what feedback and fiscal-backing conditions, is the decisive question on which the entire program stands or falls. The companion paper [54] takes up exactly this problem; we note here only that its answer is a construction identity for the accounting unit — the money price of a unit of measured negentropic value, held constant by an issuance–extinction mirror — together with an explicit agenda for the experienced price level, which the companion treats as a separate object governed by a dual-price band rather than as stabilized by decree. We refrain from claiming that this paper has solved monetary stability. The distinction drawn at the end of Section 4.1 bears repeating here: a self-consistent measuring rod is not yet a stable currency, and the conditional coherence of the value equation neither implies nor substitutes for the macrostability of the anchor built upon it.

14. Simulation and Transition Political Economy

Following the agent-based computational economics tradition (Tesfatsion [40]; Farmer and Foley [17]), the ABM (implemented in NetLogo [45]; 100 periods ≈ one century; 1,000 heterogeneous households; 50 firms; the TMS arm includes QE and macroprudential rules) reports means ± SD, 95% confidence intervals, effect sizes, and Sobol [38] global sensitivity, with code in an online appendix. Because every parameter is author-set rather than empirically calibrated, the simulation answers only a conditional question: if the mechanism holds, what follows macroscopically? It cannot establish that the mechanism holds in reality, and we position it as mechanism illustration rather than evidence. The full results appear in Appendix A; here we report only that, under the stated parameterization, the NVC arm dominates the TMS arm on all four metrics by large effect sizes (Cohen’s d from 2.8 to 4.6), with the Sobol indices identifying the α-calibration rule as the dominant sensitivity parameter. The magnitude of these effects is a property of the parameterization, not a finding about the world.

ρ∗ > ρ ≡ Br / (Br + ΔCUtransition)

(13)

In the transition game, the blocking coalition’s expected return E[block] = (1−ρ)·Br − ρ·ΔCUtransition turns negative once the pilot success probability ρ exceeds the critical threshold ρ∗. Political feasibility thereby reduces to a calibration-speed problem, which inherits the convergence logic of Proposition 12. The companion macrostability paper [54] formalizes the complementary coverage condition for the transition window — adoption of the new unit must pass a computable tipping point before the structural stabilizers bind — and the two results should be read together: feasibility is a race on two fronts, calibration speed and coverage.

15. Limitations and Conclusion

15.1. Limitations

The value equation is an ontological posit awaiting empirical adjudication, not a physical theorem. The normative foundation is a philosophical argument, and readers who reject the survival-first premise cannot be persuaded within the framework. The measurement layer embeds conventions: exergy efficiency, ecological entropy debt, the flow-state composite, and the reference distribution of Iτ are publicly recomputable but convention-dependent, and what is eliminated is unilateral discretion rather than interpretation. Appendix B specifies the institutional protocol — statistical independence, an issuance firewall against measurement error, and incentive-compatible auditing — under which these conventions operate.

Several technical qualifications deserve separate mention. Proposition 7c requires quasi-linear utility (Assumption A6); with income effects, no single αj aligns all agents simultaneously, and agent-specific corrections or transfers are needed, a genuine gap rather than a footnote — though Proposition 7e now bounds the resulting misalignment to second order in the income-dispersion parameter μ, converting the gap from open-ended to quantitative. The welfare-loss bound of Proposition 7d is local and presumes interiority and C³ smoothness. General-equilibrium uniqueness and stability rely on strong assumptions (the QV monotone likelihood-ratio property), and the conclusion that capture is uneconomic holds only within a declared parameter domain. The consensus layer is ineliminable: reference distributions and α-generation contain finite consensus and require recalibration across civilizations. Cross-corpus sensitivity analysis bounds but does not eliminate the partition dependence of Iτ. Measurement cost places validity at an endogenous-efficiency equilibrium constrained by K. GitHub and bibliometric proxies for Iτ are noisy, and empirical identification remains the hardest open problem — Section 12.1a states precisely why the cross-section cannot carry causal weight and why we decline to dress it in weak instruments. The parameters δe, Var(W), Br, and the exponents a, c (with b = 1−a) are theoretical settings awaiting estimation. Truth mastery Θ(t) depends on the unobservable pnature and remains a heuristic construct.

Three further limitations, added in this revision, belong on the record. Proposition 6, in its generalized form, makes the dependence on knowledge-growth regimes explicit: dominance holds only if the decay exponent κ of research productivity stays at zero, and the empirical literature on declining research productivity (Bloom et al. [53]) puts exactly that condition under pressure; the result’s usefulness is directional. The physical discipline of Definition 1 now forbids negative joint entropy production outright, which means the theory can never claim that value creation reduces the entropy of the universe; it claims only local order purchased at an audited environmental price, and any future formulation that drifts from this constraint should be treated as an internal inconsistency. And the response to the knowledge problem in Section 10.6 bounds the epistemic burden of calibration without eliminating it: the feedback loop from expression-layer exchange data into CRN audit is designed but untested, and a hostile reading can still argue that we have compressed Hayek’s [48] problem rather than answered it.

Two limitations connect the paper to its sister traditions and its companions. The incommensurability critique of Section 2.1a is answered at the architectural level, not dissolved: if the calibration layer’s democratic procedure fails to carry plural values in practice — if the α-procedure degenerates into technocratic scalar imposition — then the reply to Martínez-Alier, Munda and O’Neill [57] collapses and the framework reverts to exactly the commensuration the critics diagnose. Whether the governance design of the companion paper [61] can carry that weight is therefore a load-bearing condition of this paper’s value-theoretic defense. And the largest limitation remains the one we place last for emphasis rather than for modesty: the macrostability of a scarcity-free anchor is unsolved in this paper. The companion paper [54] addresses it directly; strictly speaking, no claim of this paper survives its failure.

The premise structure of the paper can now be made explicit, because not every result stands on the same ground. Table 5 maps each result to the premise layer it requires: L0 denotes physical law (the second law of thermodynamics); L1 the measurement conventions of Section 4.2; L2 the axioms and screening criteria of Section 4.1; L3 the normative survival-first premise of Section 3.4; and L4 parameter settings (ρ, g0, κ, the exponent benchmark).

The map makes visible a fact the earlier text left implicit. The entire microeconomic and mechanism-design apparatus — existence, uniqueness, conditional decentralization, the welfare bound, risk compensation, dynamic consistency, and general equilibrium — requires only the axiom layer and survives intact if the survival-first premise is rejected: it is a technology for steering incentives toward whatever objective a society adopts. Only the choice of objective — civilization-utility maximization, and with it Proposition 6 and the α-structure — draws on the normative layer. The premise carries the direction of the program, not its machinery.

15.2. Conclusion

Granted the premise that civilizational survival is a lexicographically prior meta-constraint, this paper has constructed a two-step theoretical program. The first step is the negentropic theory of value: value as the net increment of systemic order, with its concept terminologically disciplined, its connotation stratified into three layers, its boundary explicitly drawn, its normative foundation defended through a lexicographic hierarchy and systematic replies to five counter-positions, and its commensuration claim qualified against the incommensurability critique of ecological economics. The second step is civilization utility theory: civilization combats entropy through continuous negentropic creation, with cosmic-truth mastery as the ultimate directional goal, supported by the conditional asymptotic-dominance proposition (Proposition 6) and the per-truth counterfactual accounting path worked through relativity, quantum mechanics, germ theory, and, at the level of protocol, PCR. Civilization and individuals co-upgrade through the five-dimensional-capital mechanism.

The methodological stance behind the whole construction can be stated in one sentence: we do not claim to have found the truth, but to have built the apparatus that leads toward it under explicit premises. The value posits are derivable, the propositions provable under stated assumptions, the calibration targets computable, the algorithms runnable, the blueprint deducible, and the tests executable, each annotated with its premises and failure boundaries. Where earlier versions of this program overclaimed, whether about physical freedom from convention, about theorem status, about efficiencies above unity, or about negative joint entropy production, the present version withdraws the claim and keeps the apparatus. If the premises are accepted and the tests support the theory, monetary thought may remember this migration from the reification of social relations toward the socialization of physical order. If the premises are rejected or the tests fail, the apparatus remains as a criticizable and improvable benchmark for the theory of monetary value.

B.1. Statistical Independence: The Measurement Institution

Definition B1 (statistical independence). A measurement process for (Eτ, Tτ, Iτ) is statistically independent if (i) no agent whose payoff depends on issuance outcomes can alter, delay, or suppress estimates; (ii) the full estimate-generating pipeline is reproducible by any third party from published inputs and code; and (iii) the measuring institution holds no mandate over α, which remains in the CPPA/QV domain (Section 8, Section 10.2, and Section 10.5). The measuring institution prices nothing; the pricing institution measures nothing.

Four provisions implement Definition B1 within the CRN architecture.

• Tenure. Measurement-office seats follow a staggered, non-renewable term structure (the design that the companion governance paper [61] assigns to its audit committee), with term length exceeding the political cycle and one cohort rotating per electoral period; removal for cause requires a supermajority of the parliament with ex-post public justification.
• Budget. The measurement budget is fixed multi-year as a declared share of the ks smoothing reserve (Section 11.1, layer L4), removing annual discretionary leverage over the measuring institution.
• Mandate. The office publishes estimates of Eτ, Tτ, Iτ and their precisions only. It neither sets nor recommends α; shadow-price formation remains with the CPPA iteration (eq (11)) and its Delphi–simulation–ratification stages.
• Reproducibility. Anonymized inputs, transformation code, and the complete revision history are published under the L1 zero-knowledge architecture — data usable but not visible (Section 11.1) — so that every published number can be recomputed and every revision traced.

Proposition B1 (staggering multiplies capture cost). Let the measurement committee have n seats staggered across m rotating cohorts under the elite-capture game of Section 10.4 (eq (12)), and suppose distorting published estimates requires a sitting majority. Then capture requires controlling at least ⌈n/2⌉ seats appointed in at least ⌈m/2⌉ distinct cohort rounds; non-renewable terms eliminate incumbency lock-in, so the coalition must be re-assembled at every rotation; and with random mutual audit at per-seat detection probability ρ per period, a biased estimate sustained over H periods evades detection with probability at most (1−ρ)^H·⌈n/2⌉. The optimal lobbying effort in eq (12) is therefore driven to zero over a strictly larger parameter domain than in the single-cohort case, and the domain expands multiplicatively in m. The argument is arithmetic: staggering serializes what capture must simultaneously buy, rotation removes the discount on future bribes, and independent per-seat detection compounds geometrically.

B.2. The Issuance Firewall: Measure Fast, Issue Slow, Correct Ex Post

Proposition 5 prices measurement error; the firewall decides who bears it. Issuing on current estimates would transmit σW/W into the accounting unit one-for-one. The protocol therefore separates estimation from confirmation: estimates are produced at the measurement frequency, but issuance responds only to confirmed value.

Definition B2 (confirmed value). The issuance-relevant value of activity τ is the log-domain error-corrected estimate

ln W̃τ(t) = θ·ln Ŵτ(t) + (1−θ)·ln W̃τ(t−1), 0 < θ ≤ 1

(B1)

The log-domain form matches the multiplicative structure of eq (1): ln Wτ = ln k + a·ln Eτ + (1−a)·ln Tτ + c·ln Iτ is linear in the exponents, so the correction remains valid for the general calibrated form, not only at the benchmark Wτ = √Vτ. Ex-post revisions are posted to an extinction-gap ledger rather than to current issuance. (For self-containedness: the extinction-gap ledger, developed in the companion macrostability paper [54], is the stock account that records money which should have been extinguished under the issuance–extinction mirror but was not; a positive entry metabolizes at the intrinsic rate δ, and a negative entry credits the issuance reserve for release under the normal ks schedule. Only the accounting role of the ledger is needed here.)

ΔX(t+1) = (1−δ)·ΔX(t) + ρW(t), ρW(t) = Στ [W̃τ(t) − Wτfinal(t)]·ατ

(B2)

A positive ρW — confirmed value exceeded final value — adds to the extinction gap and metabolizes at rate δ; a negative ρW credits the issuance reserve and is released through the normal ks schedule. Measurement error moves the stock ledger, never the flow anchor: ex post, the money stock is neutral to measurement error.

Proposition B2 (bounded error transmission). Write ln Ŵτ(t) = ln Wτ(t) + ε(t) with ε(t) zero-mean, serially uncorrelated, Var(ε) = σ². Then:

Vartrans = [θ/(2−θ)]·σ²

(B3)

(i)

the relative variance transmitted to issuance is (B.3): the confirmation filter is an AR(1) whose noise gain is θ/(2−θ), so θ = 1/4 transmits about 14% of measurement variance, roughly 38% of the standard deviation;

(ii)

smoothing is one-sided in the safe direction: W̃τ never exceeds the recent maximum of Ŵτ, and under trend growth g the confirmation lag biases issuance downward by ≈ (1−θ)g/θ — an under-issuance drift that the ks buffer absorbs by construction, never an over-issuance impulse; and

(iii)

if revisions are unbiased, E[ρW] = 0, and the geometric decay of ΔX gives E[πerr] = 0 asymptotically, with correction half-life ln 2/δ. The variance of the flow is compressed by θ; the bias of the stock is discharged by δ.

A numerical closure ties B.2 to the precision budget. At the equal-allocation tolerance of Section 4.4 (each factor at 5.8%, σW/W ≈ 5%), θ = 1/4 transmits a 1.9% relative standard deviation to issuance; under δ = 0.01 the stock correction of any residual bias has a half-life of about 69 periods. The first line of defense is therefore the filter and the ledger is the backstop — the same ordering of defenses, and the same acknowledgment that the backstop is thin, as in the macrostability analysis of the companion paper [54].

B.3. Incentive-Compatible Audit

Definition B3 (audit game). The audit game has three players. A reporter chooses falsification intensity m ≥ 0 at convex cost C(m) (Section 10.3). K auditor nodes, drawn randomly each round under anonymous assignment, choose effort e at cost c(e). The audit committee commits ex ante to penalty F and bounty B, the latter funded from collected penalties and slashed stakes. The detection probability q(e, K) is increasing and concave in e and increasing in K.

Proposition B3 (truth-telling equilibrium). Truthful reporting (m = 0) supported by audit effort ē is a Bayesian Nash equilibrium whenever (i) deterrence: q(ē, K)·(F + M) ≥ m·Wτ·α for all m ≤ m̄, where M is the reporter’s posted stake (Section 10.3); (ii) participation: q(ē, K)·B ≥ c(ē); and (iii) collusion resistance: bounties vest only on independent corroboration by at least two randomly drawn nodes. The conditions are jointly satisfiable whenever the falsification gain is bounded relative to the stake-penalty pool, and condition (iii) keeps the required bounty independent of the number of compromised nodes below the corroboration threshold. The companion paper’s computational implementation ran with per-period detection hazard qA = 0.25 — cumulative detection of about 68% within four periods — and recovered from accounting-falsification attacks in a mean of 10.4 periods; Definition B3 is the institutional source from which that hazard is drawn.

Proposition B4 (Goodhart resistance of the geometric form). Manipulating any single factor by a multiplicative factor γ, with the other two honestly measured, inflates reported value by γ^a, γ^1−a, or γ^c according to the manipulated factor — each strictly less than γ for 0 < a, c < 1, and equal to √γ at the symmetric benchmark, so that a 10% inflation of Wτ requires a 21% distortion of one factor. The L1 cross-validation triggers (T–I correlation ≥ 0.7; 15% deviation review, Section 11.1) bind feasible γ before the audit game begins, so single-factor manipulation is dominated; factor-proportional manipulation requires simultaneous collusion across three technologically disjoint pipelines — energy metering, multimodal biosensing, and knowledge graphs — raising the collusion dimension from one to three. Because a + (1−a) = 1 forces max{a, 1−a} ≥ 1/2, the symmetric benchmark minimizes the largest per-factor exponent: its square root is the strongest uniform damping guarantee within the family (1), and the exponent that damps honest error (Section 4.4) damps dishonesty identically.

B.4. Threat–Mechanism Map and Residual Limits

Table B1. Threats to measurement integrity and the protocol layer answering each.

Threat	Layer	Mechanism	Formal result
Political capture of the measurement institution	B.1	Staggered non-renewable terms; mandate separation from α; budget autonomy	Proposition B1
Random measurement error in E, T, I	B.2	Log-domain confirmation filter	Proposition B2(i), eq (B.3)
Systematic measurement bias	B.2	Unbiased revisions posted to ΔX and the reserve	Proposition B2(iii)
Smoothing-induced over-issuance	B.2	One-sided lag bias, absorbed by ks	Proposition B2(ii)
Reporter falsification (accounting attack)	B.3	Stake–penalty–bounty game with corroborated bounties	Proposition B3
Single-factor Goodhart gaming	L1 + B.3	Square-root damping; cross-factor correlation triggers	Proposition B4

Table B1. Threats to measurement integrity and the protocol layer answering each.

Threat	Layer	Mechanism	Formal result
Political capture of the measurement institution	B.1	Staggered non-renewable terms; mandate separation from α; budget autonomy	Proposition B1
Random measurement error in E, T, I	B.2	Log-domain confirmation filter	Proposition B2(i), eq (B.3)
Systematic measurement bias	B.2	Unbiased revisions posted to ΔX and the reserve	Proposition B2(iii)
Smoothing-induced over-issuance	B.2	One-sided lag bias, absorbed by ks	Proposition B2(ii)
Reporter falsification (accounting attack)	B.3	Stake–penalty–bounty game with corroborated bounties	Proposition B3
Single-factor Goodhart gaming	L1 + B.3	Square-root damping; cross-factor correlation triggers	Proposition B4

Three residuals remain on the record. First, the reference distribution of Iτ is a convention, and conventions drift; the protocol guarantees that drift is public and recomputable, not that it is impossible. Second, simultaneous three-factor manipulation by a captured measurement institution — capture of layer B.1 by the very coalition layer B.3 polices — is bounded only by Proposition B1’s cost multiplication and by the λτ–α divergence early-warning signal (Section 8), not eliminated. Third and deepest: the protocol secures the honesty and isolation of measurement, not the truth of the measured. Whether E·T·I tracks systemic order is the ontological wager of Section 3, and no institutional design can discharge it.