Chapter 9: Conservation Laws for Economics

9.1 From Symmetry to Conservation

Emmy Noether’s theorem, published in 1918, is among the most profound results in mathematical physics. It establishes that every continuous symmetry of a physical system corresponds to a conserved quantity. Time-translation symmetry implies conservation of energy. Spatial-translation symmetry implies conservation of momentum. Rotational symmetry implies conservation of angular momentum.

[Established Mathematics.] In Chapter 8, we established the economic Bond Invariance Principle: economic evaluations must be invariant under admissible re-descriptions — currency relabeling, unit scaling, numeraire choice. This is a symmetry of the economic decision manifold. Noether’s theorem tells us that this symmetry must correspond to a conserved quantity.

What is conserved? This chapter derives the answer: transferable attributes are conserved in bilateral exchange. Value on the transferable dimensions — monetary value (d_1), rights (d_2), and autonomy (d_4) — cannot be created or destroyed by relabeling. It can only be transferred from one party to another. The evaluative dimensions (d_3, d_5 through d_9), by contrast, are not conserved — they can increase for both parties simultaneously, which is precisely what makes voluntary exchange mutually beneficial.

9.2 Attribute Conservation in Bilateral Exchange

[Conditional Theorem.] Consider a closed bilateral exchange between agents A and B — a transaction where no third party enters and no external rules change during the exchange.

Theorem 9.1 (Attribute Conservation in Closed Exchange). In a closed bilateral economic exchange between agents A and B, the transferable attribute-vector changes are conserved:

\Delta d_k(A) + \Delta d_k(B) = 0 \quad \text{for all transferable dimensions } k

A dimension d_k is transferable if the gain to one party necessarily comes from the other: monetary value (d_1), rights (d_2 via Hohfeldian correlative structure), and autonomy (d_4, in the sense that one party’s coercion is another’s constraint).

Proof. On the monetary dimension (d_1), conservation follows from accounting identity: in a closed exchange, every dollar A pays is a dollar B receives. \Delta d_1(A) + \Delta d_1(B) = 0.

On the rights dimension (d_2), conservation follows from the Hohfeldian correlative structure developed in Geometric Ethics (Chapter 12). Every right A transfers is an obligation B assumes; every obligation A is relieved of is a claim-right B acquires. The correlative mapping preserves the total rights vector.

On the autonomy dimension (d_4), conservation holds in the sense that voluntary constraints are symmetric: if A agrees to deliver goods by Friday (constraining A’s autonomy, \Delta d_4(A) < 0), then B acquires the right to expect delivery (\Delta d_4(B) > 0 in the form of guaranteed receipt). Coercion — one party’s autonomy loss without compensating gain — violates the conservation law and is detectable as such. \square

9.3 Non-Conservation on Evaluative Dimensions

[Empirical.] The evaluative dimensions — trust (d_5), social impact (d_6), identity (d_7), legitimacy (d_8), and epistemic status (d_9) — are not conserved in general. Both parties may simultaneously perceive an exchange as:

• Trust-building (\Delta d_5 > 0 for both): the exchange reveals reliability, creating mutual confidence that did not exist before.
• Socially beneficial (\Delta d_6 > 0 for both): a commercial transaction that strengthens community ties (Maria’s coffee shop as neighborhood hub).
• Identity-affirming (\Delta d_7 > 0 for both): Elena takes pride in craft roasting; Maria takes pride in ethical sourcing. The transaction affirms both identities.
• Epistemically enriching (\Delta d_9 > 0 for both): the exchange reveals information about quality, reliability, and market conditions that both parties value.

Remark 9.1 (Mutual Value Creation). The non-conservation of evaluative dimensions is what makes voluntary exchange positive-sum. On the transferable dimensions, trade is zero-sum by conservation. On the evaluative dimensions, trade can be positive-sum. The total value created by an exchange is:

V_{\text{created}} = \sum_{k=5}^{9} \left[ \Delta d_k(A) + \Delta d_k(B) \right]

This quantity can be positive, and in healthy markets, it typically is. The insight formalizes Adam Smith’s observation that trade creates “mutual advantage” — but now we can say precisely on which dimensions the advantage accrues and why the transferable dimensions cannot contribute to it.

9.4 The No Free Lunch Theorem

Theorem 9.2 (No Free Lunch on Transferable Dimensions). An agent cannot extract positive surplus on all transferable dimensions simultaneously from a voluntary bilateral exchange. If \Delta d_k(A) > 0 for some transferable dimension k, then \Delta d_k(B) < 0 (by Theorem 9.1). In particular, if A gains monetarily (\Delta d_1(A) > 0), then B pays (\Delta d_1(B) < 0).

Proof. Immediate from conservation: \Delta d_k(A) + \Delta d_k(B) = 0 implies that a positive \Delta d_k(A) requires a negative \Delta d_k(B). \square

This is the geometric formalization of “there is no such thing as a free lunch” — but restricted to the transferable dimensions. On the evaluative dimensions, free lunches are not merely possible but typical: a transaction that builds trust creates value for both parties at no transferable cost.

9.5 The Value Accounting Identity

The conservation law yields a powerful accounting identity that any legitimate economic analysis must satisfy.

Theorem 9.4 (Value Accounting Identity). For any closed economic system containing n agents engaged in m bilateral exchanges, the total change in transferable dimensions sums to zero:

\sum_{i=1}^{n} \sum_{k \in \{1,2,4\}} \Delta d_k^{(\text{total})}(i) = 0

while the total change in evaluative dimensions may be positive, negative, or zero:

\sum_{i=1}^{n} \sum_{k \in \{3,5,6,7,8,9\}} \Delta d_k^{(\text{total})}(i) \gtrless 0

Proof. The first identity follows by linearity from Theorem 9.1: each bilateral exchange conserves transferable attributes, so the sum over all exchanges conserves them. The second is a direct consequence of the non-conservation on evaluative dimensions established in Section 9.3. \square

The accounting identity has immediate consequences:

Corollary 9.2 (GDP Is Conservation-Consistent but Incomplete). GDP measures the sum of d_1 transactions in an economy. By Theorem 9.4, the total \Delta d_1 across all agents in a closed economy is zero — every dollar spent is a dollar earned. GDP measures the volume of these transactions (their sum of absolute values), not their net change. This is conservation-consistent: GDP does not claim to measure net value creation on d_1, only the level of activity. But GDP is silent about the evaluative surplus V_{\text{created}}, which may be the majority of value generated by the economy. An economy with high GDP but declining trust, fairness, and community is an economy with high transferable turnover and negative evaluative surplus.

Corollary 9.3 (Ponzi Schemes Violate Conservation). A Ponzi scheme claims to create transferable value (d_1 returns to investors) without corresponding transferable cost. Since \Delta d_1 is conserved, the returns must come from other investors’ capital, not from value creation. The conservation law immediately identifies the scheme as parasitic: it draws down the transferable pool while (temporarily) creating the illusion of non-conservation on d_1. The scheme collapses when the pool is exhausted — which conservation guarantees must happen.

9.6 Open vs. Closed Systems

The conservation laws established in this chapter apply to closed bilateral exchanges — transactions where no third party enters and no external conditions change. Real economies are open systems, and the distinction matters.

[Modeling Axiom.] In an open economic system — one where external agents, regulatory changes, or environmental shifts can inject or extract value — the conservation law for transferable dimensions generalizes:

\Delta d_k(A) + \Delta d_k(B) = F_k^{\text{external}}

where F_k^{\text{external}} is the net external input on dimension d_k. In a closed exchange, F_k^{\text{external}} = 0 and conservation holds. In an open system, F_k^{\text{external}} can be positive (subsidies, external investment) or negative (taxation, expropriation).

Government taxation is the most common form of external extraction on d_1: the government takes a fraction of the transaction’s monetary value, so \Delta d_1(A) + \Delta d_1(B) = -\text{tax}. The tax revenue is not destroyed — it is transferred to the government, which is a third party. In the closed system {A, B, government}, conservation holds. The conservation violation is an artifact of defining the system boundary too narrowly.

This observation has a methodological implication: conservation violations, when observed, signal that the system boundary is incorrect — that there are external flows the analysis is missing. A market that appears to create money from nothing (a Ponzi scheme) is violating conservation within the defined system. The violation diagnoses the presence of unobserved external flows (other investors’ capital). A company that appears to generate profit without creating value is extracting from unobserved dimensions of other agents’ manifolds.

9.7 Exploitation as Conservation Violation

Corollary 9.1 (Exploitation Is Detectable). If one party consistently extracts surplus on d_1 (monetary gain) while the counterparty consistently bears costs on d_3 (fairness), d_4 (autonomy), or d_6 (social impact), the Bond Index will be non-zero. Exploitation is not a moral judgment — it is a measurable asymmetry in the dimensional distribution of behavioral friction.

[Empirical.] This gives us a quantitative definition of exploitation that does not depend on subjective moral assessment. The measurement is: compute the attribute-vector changes for both parties across all nine dimensions. If party A’s gains are concentrated on d_1 while party B’s losses are distributed across d_3, d_4, d_5, d_6, and d_7, the transaction has the geometric signature of exploitation — regardless of whether both parties nominally “agreed” to it.

The framework explains why some voluntary transactions feel exploitative: the voluntariness (conservation on d_4) does not prevent asymmetry on the other dimensions. A worker who “voluntarily” accepts a dangerous job for high pay may be experiencing conservation on d_4 (they chose freely) but violation on d_3 (the risk is disproportionate to the pay, measured on the full manifold).

9.6 Value Conservation and the Price System

[Modeling Axiom.] The conservation law constrains the price system. A well-functioning price captures the d_1 component of the exchange — the monetary transfer. But the full exchange involves all nine dimensions. A price that reflects only d_1 systematically under-prices transactions that create evaluative value (trust, community, identity) and over-prices transactions that destroy it.

This is the geometric explanation for the persistent gap between market prices and “true” economic value. The gap is not a market failure in the traditional sense — the price system is functioning correctly on d_1. It is a dimensional projection: the price captures one dimension of a nine-dimensional transaction, and the Scalar Irrecoverability Theorem (Chapter 6) says the other eight dimensions cannot be recovered from the price alone.

Maria’s fair-trade coffee costs $2/lb more than commodity coffee. On d_1, this is a pure loss. On the full manifold, the extra cost buys fairness (d_3), community development (d_6), identity consistency (d_7), and supply chain transparency (d_9). The “premium” is not irrational — it is the monetary cost of traversing a geodesic on the full manifold rather than the d_1-only projection.

9.8 The Noether Derivation

The conservation laws established above were proved directly from the structure of bilateral exchange. But there is a deeper route: deriving them from the Bond Invariance Principle via Noether’s theorem. This derivation connects the conservation laws to the gauge invariance developed in Chapter 8 and reveals why certain dimensions are conserved and others are not.

[Conditional Theorem.] The argument proceeds in three steps.

Step 1: The Economic Lagrangian. Define the economic Lagrangian \mathcal{L} for an exchange as the difference between the total benefit (reduction in behavioral friction for both parties) and the total cost (increase in behavioral friction):

\mathcal{L}(\gamma, \dot{\gamma}) = \sum_{i \in \{A,B\}} \left[ -\text{BF}_i(\gamma_i) \right] = -\sum_{i} \left[ \Delta\mathbf{a}_i^T \Sigma_i^{-1} \Delta\mathbf{a}_i + \sum_k \beta_{k,i} \cdot \mathbf{1}[\text{boundary}] \right]

The path that extremizes this Lagrangian — the Euler-Lagrange solution — is the pair of Bond geodesics for the two parties. This is the BGE of Chapter 7, re-derived variationally.

Step 2: The Symmetry. The economic BIP (Chapter 8) requires that \mathcal{L} be invariant under admissible re-descriptions. For the transferable dimensions, the relevant symmetry is relabeling invariance: renaming who holds the money, who holds the rights, who exercises the autonomy cannot change the value of the exchange. Formally, for any transferable dimension d_k, the transformation \Delta d_k(A) \to \Delta d_k(A) + \epsilon, \Delta d_k(B) \to \Delta d_k(B) - \epsilon (transfer of \epsilon units from B to A) leaves \mathcal{L} invariant only if the total \Delta d_k(A) + \Delta d_k(B) enters \mathcal{L} as a constant. This is the symmetry.

Step 3: Noether’s theorem. By Noether’s theorem, this continuous symmetry implies a conserved current. The conserved quantity is:

J_k = \Delta d_k(A) + \Delta d_k(B) = \text{const} = 0

for each transferable dimension k. The initial condition J_k = 0 (before the exchange, neither party has gained or lost on dimension k due to this exchange) determines the constant. Conservation of transferable attributes follows.

Why evaluative dimensions are not conserved. The evaluative dimensions are not subject to the relabeling symmetry. There is no transformation that transfers trust from one party to another while leaving the Lagrangian invariant, because trust is not a transferable quantity — it is a property of the relationship. The absence of the symmetry means the absence of the conservation law. Noether’s theorem works in both directions: symmetry implies conservation, and the absence of symmetry implies the absence of conservation.

This is the deepest reason for the transferable/evaluative distinction. It is not a classification imposed by the modeler. It is a consequence of which symmetries the economic Lagrangian possesses. The conservation laws are not axioms — they are derived from the BIP, just as conservation of energy is derived from time-translation symmetry in physics.

9.9 Connection to Physical Conservation Laws

[Speculation/Extension.] The structural parallel between economic conservation and physical conservation is precise:

We do not claim these are the same conservation laws. We claim they have the same mathematical form — continuous symmetry → conserved quantity — and that this shared form is not coincidental but reflects the power of the geometric framework to capture conservation phenomena across domains.

The parallel extends to the failure modes of conservation. In physics, conservation of energy can appear to be violated in open systems — a ball rolling to a stop appears to “lose” kinetic energy. The resolution: the energy was not destroyed; it was transferred to the environment as heat (a dimension the initial accounting neglected). In economics, conservation of monetary value can appear to be violated when value seems to appear “from nowhere” — a stock doubles in price with no change in fundamentals. The resolution: the monetary gain came from other investors (conservation on d_1 within the closed system of market participants), or the “gain” is nominal rather than real (a gauge violation, Chapter 8).

The second law of thermodynamics — entropy of a closed system never decreases — has an economic shadow in Theorem 9.3: trust, once destroyed, cannot be fully recovered by the same sequence of interactions that built it. The time-asymmetry of trust creation and destruction mirrors the time-asymmetry of entropy production. Both arise from the same geometric feature: the space of states accessible by “constructive” paths is smaller than the space accessible by “destructive” paths. Building trust requires a specific sequence of reliable interactions; destroying it requires only one betrayal.

Worked Example: Maria’s Supply Chain Transaction

Maria places her monthly order with Elena: 80 pounds of single-origin Colombian beans at $25/lb = $2,000.

Transferable dimensions (conserved):

Evaluative dimensions (not conserved):

Total value created: 0 on transferable dimensions (conservation holds) + 2.4 units on evaluative dimensions (positive-sum). The transaction creates value even though no new goods are produced — the value is in the relational, identity, and epistemic dimensions that the price cannot capture.

Now compare with a transaction where Maria buys commodity beans from an anonymous broker at $18/lb:

Evaluative dimensions: All near zero. No trust built (\Delta d_5 \approx 0), no community (d_6 \approx 0), no identity affirmation (d_7 \approx 0). The monetary savings ($7/lb × 80 = $560) come at a cost of 2.4 units of evaluative value — a cost invisible to d_1-only accounting.

The Bond geodesic for Maria — the minimum-cost path on the full manifold — depends on the Mahalanobis metric \Sigma. If Maria’s \Sigma weights evaluative dimensions heavily (as the behavioral data suggests for morally loaded transactions), the fair-trade geodesic has lower total cost despite higher monetary cost.

Implications for Economic Theory

Technical Appendix

Definition 9.1 (Transferable Dimension). A dimension d_k of the economic decision complex is transferable if for every closed bilateral exchange, \Delta d_k(A) + \Delta d_k(B) = 0. Otherwise it is evaluative.

Proposition 9.1 (Transferability Test). [Conditional Theorem.] Dimension d_k is transferable if and only if the gain of d_k-value by one party requires the surrender of d_k-value by the other — i.e., the dimension admits no “creation from nothing.” In the Hohfeldian framework, this corresponds to dimensions whose changes are governed by correlative pairs (right↔︎duty, privilege↔︎no-right).

Proposition 9.2 (Open System Conservation). [Conditional Theorem.] In an open economic system with m agents and external flows F_k^{\text{ext}}, the generalized conservation law is:

\sum_{i=1}^{m} \Delta d_k(i) = F_k^{\text{ext}} \quad \text{for all transferable } k

When F_k^{\text{ext}} = 0 (closed system), conservation holds exactly. When F_k^{\text{ext}} \neq 0 (open system), the total internal change equals the external flow. A positive F_k^{\text{ext}} indicates external value injection (subsidy, investment); a negative F_k^{\text{ext}} indicates extraction (taxation, expropriation). Conservation violations within the system diagnose unobserved external flows.

Proposition 9.3 (Exploitation Index). [Modeling Axiom.] For a series of n transactions between parties A and B, the exploitation index is:

E(A \to B) = \frac{1}{n} \sum_{t=1}^{n} \left[ \Delta d_1^{(t)}(A) \cdot \sum_{k \neq 1} |\Delta d_k^{(t)}(B)| \right]

A positive exploitation index indicates that A’s monetary gains are systematically accompanied by B’s non-monetary losses. The index is zero for fair exchanges and positive for exploitative ones.

Implications for Institutional Design

Notes on Sources

The conservation laws developed in this chapter have antecedents in accounting identity theory (the double-entry bookkeeping system, invented by Luca Pacioli in 1494, is the oldest conservation law in economics) and in Walras’s law (total excess demand across all markets sums to zero — a conservation statement about monetary flows). The contribution here is the extension to the full nine-dimensional manifold and the distinction between transferable and evaluative dimensions, which is original.

The Noether derivation (Section 9.8) follows the general structure of Noether (1918) as applied to field theories, adapted to the economic Lagrangian. The analogy between conservation laws in physics and economics has been noted informally by many authors (Samuelson’s Foundations of Economic Analysis, 1947, explicitly drew on variational methods from physics) but has not, to my knowledge, been developed with the full BIP-Noether machinery applied here.

The distinction between extensive (conserved, transferable) and intensive (non-conserved, evaluative) economic quantities parallels the distinction in thermodynamics (Callen, 1985, Thermodynamics and an Introduction to Thermostatistics). The economic analogue of the second law of thermodynamics — asymmetric trust destruction — appears to be new.

The evaluative surplus concept connects to the social capital literature (Putnam, 2000; Coleman, 1988) and the relational contracting literature (Baker, Gibbons, and Murphy, 2002). The conservation framework provides the mathematical structure that these literatures have described qualitatively: social capital is the accumulated evaluative surplus of repeated economic interactions, and it can be created (positive-sum) or destroyed (asymmetrically) but not transferred (not conserved).

References

Bond, A. H. (2026a). Geometric Methods in Computational Modeling. SJSU. Bond, A. H. (2026b). Geometric Ethics: The Mathematical Structure of Moral Reasoning. SJSU. Cecchetti, S. G., and Kharroubi, E. (2012). “Reassessing the Impact of Finance on Growth.” BIS Working Paper No. 381. Coleman, J. S. (1988). “Social Capital in the Creation of Human Capital.” American Journal of Sociology, 94, S95–S120. Noether, E. (1918). “Invariante Variationsprobleme.” Nachrichten der Königlichen Gesellschaft der Wissenschaften zu Göttingen, 235–257. Polanyi, K. (1944). The Great Transformation. Rinehart. Putnam, R. D. (2000). Bowling Alone. Simon & Schuster. Smith, A. (1776). An Inquiry into the Nature and Causes of the Wealth of Nations.