§2. The World Forces New Monetary Primitives

Jason St George. "§2. The World Forces New Monetary Primitives" in Next Generation Stores of Value: Privacy, Proofs, Compute. Version v3.5. /v/3.5/read/part-i/2-world-forces-new-monetary-primitives/

The World Forces New Monetary Primitives

The forces doing the forcing run on two channels at once, and they are the same channel. In truth, the incumbent order is broadcast: one attester—state, rating agency, platform, exchange—asserts, and $N$ receivers accept. In value, the same shape: central-bank credit is one issuer to $N$ accounts, with the sovereign acting as the typesetter of uniform value, every unit identical because one authority struck it. Proofs and bearer settlement are both $N \leftrightarrow N$: anyone checks anyone’s claim; anyone settles with anyone directly. The thesis is one observation wearing two dresses—truth and value move on the same topology, and both are moving from broadcast to network. That is a way of reading what follows, not an additional premise: the chapters that follow treat macro pressure (§2: The World Forces New Monetary Primitives) and epistemic pressure (§1: The Failure of Soft Guarantees) as one phenomenon observed from two sides, and each must still carry its own evidence.

Macro Playbook: Debt → Repression → Flight to Neutrality

Total global debt (private + public) according to the IMF’s latest Global Debt Database [IMF 2025] now sits at a roughly $\sim$2.4× multiple of world GDP [Gaspar et al. 2025], with some economies (such as Japan) exceeding 400% of GDP when public and private debt are combined. Standard scenarios from multilateral institutions point to global public debt approaching $\sim$100% of GDP by decade-end [IMF 2026], after already exceeding $100 trillion in the mid-2020s [IMF 2024].

Historically, such overhangs are closed via financial repression: inflation plus regulation that imposes negative real returns, including overt capital controls and captive-balance-sheet rules, pushing savers toward neutral, censorship-resistant rails. “Voting with your feet” is harder when the exits are guarded.

The Bondholder’s Trap (How Repression Works)

Definition. Financial repression is a stealth tax on savings: cap nominal rates, let inflation run, and force captive balance sheets to hold the paper. The result is systematically negative real returns for depositors and bondholders; an engineered wealth transfer to the sovereign. In the post-WWII advanced economies, real rates were negative in roughly half the years from 1945–1980, and the “liquidation tax” (interest-expense savings on the public debt) averaged on the order of 1–5% of GDP per year [Reinhart & Sbrancia 2015].

How the cap is enforced. Overt or implicit yield-curve control (YCC), administered rate ceilings, capital/liquidity rules that force government paper into banks, insurers, pensions, and collateral boxes, plus capital controls to slow leakage. In the U.S. WWII episode, the Fed capped T-bills at 3/8% and long bonds at 2½% until the 1951 Accord [Rose 2021]; when the cap ended, long-duration holders took capital losses as yields normalized.

Why now. With global debt at several times world GDP, public debt ratios trending toward $\sim$100% of GDP by decade-end, and interest costs compounding, repression is arithmetically attractive to policymakers versus explicit default or politically costly austerity.

Two equations carry the mechanism:

1. Repression wedge (real carry):

If $i_{\text{capped}} = 2.5\%$ and inflation $\pi = 6\%$, the real return is

2. Duration burn (when the peg breaks):

For a bond with duration $D$, a yield move $\Delta y$ produces a price move

A 10-year with $D \approx 8$ takes about $-24\%$ on a +300 bps repricing.

A worked example: three years at $-3.3\%$ real/yr is $\approx -9.6\%$ real loss compounded; then a +300 bps jump deals another $\approx -24\%$ price loss, roughly $-31\%$ cumulative in purchasing-power terms. Each instrument in it is chosen by someone. What they produce together needs no motive to be real: the bondholder’s real wealth moves to the sovereign balance sheet, and the sovereign buys time.

Then and Now

The two equations have been run before, at national scale. From 1942 to 1951 the Federal Reserve pegged bills at 3/8% and capped long bonds at 2½% (§2: The World Forces New Monetary Primitives); across the advanced economies, real short rates were negative in roughly half the years from 1945 to 1980 (§2: The World Forces New Monetary Primitives). The bondholder was not defaulted on. The bondholder was held in place while the real value of the claim was taken, and when the Accord released the cap, the long holder paid the second equation as well.

What has changed since is the enforcement. As convertibility receded, the compliance and regulatory perimeter became a larger part of what makes fiat enforceable, especially under capital controls, sanctions, and financial repression. The wartime peg needed a captive banking system. The modern version needs captive balance sheets, custody defaults, and the power to freeze, and the freezing of Russia’s reserves in early 2022 showed that the last of these extends to sovereigns. Some investors and institutions have read that episode as a reason to reassess neutral collateral and privacy-preserving settlement. When repression becomes the spread, capital rotates quietly toward neutral, bearer-like rails.

The Treasury Market as Control Panel

Modern repression does not need to announce itself as yield-curve control. It can be routed through balance-sheet architecture. A sovereign that cannot fund itself at market-clearing real yields can manufacture demand for its paper by changing the rules under which banks, insurers, pensions, money-market funds, stablecoin issuers, and custodians hold collateral.

The playbook is subtler than confiscation: lower front-end rates, pressure or tolerate higher long-end yields, steepen the curve, loosen bank leverage treatment for sovereign paper, and let regulated balance sheets absorb debt that private investors would not hold at prevailing real yields. The central bank can claim discipline because its own balance sheet is shrinking, while the banking system performs the functional equivalent of QE. In practice this shows up as redefined high-quality-liquid-asset treatment, eased leverage-ratio exemptions for Treasuries, incentives for banks to intermediate sovereign debt, privileged government collateral in payment and clearing systems, steered stablecoin reserve composition, and regulated custody as the default gateway for institutional capital.

Balance-Sheet Repression

Financial repression implemented through collateral rules, capital treatment, stablecoin reserve requirements, custody mandates, and institutional balance-sheet incentives, rather than through overt yield-curve control or capital controls.

For a post-fiat monetary stack, this matters because the adversary is not only censorship. It is substitution. Regulated wrappers, approved custody, stablecoin reserve mandates, and institutional balance-sheet rules can direct capital into price exposure while starving protocol-native privacy, proof, settlement, fee-burn, and collateral loops. The stack must therefore monitor not only whether users can access the asset, but whether they are using the monetary rail or merely holding a compliant shadow claim. §10b: Why Hold the Base Asset? formalizes this substitution risk as the Wrapper Dominance Ratio.

Who Actually Buys the Paper

A control panel is only as useful as the hands available to absorb what it issues. The composition of those hands has changed, and the change matters more than the level of any particular yield.

Foreign official reserve managers—the patient, price-insensitive creditors of the previous era—have not been dependable net buyers of U.S. government debt for some years. A growing share of marginal demand now comes from leveraged relative-value funds that hold cash Treasuries against short futures positions and finance the difference in the repo market. Federal Reserve staff research [Barth et al. 2025] estimates that Treasury holdings of Cayman-domiciled hedge funds rose by approximately $1 trillion after 2022, reaching roughly $1.85 trillion by the end of 2024, financed through bilateral and centrally cleared sponsored repo.

Two qualifications are load-bearing, and the source states both. Not all of that exposure is the cash–futures basis trade; hedge funds hold Treasuries for several repo-supported arbitrage reasons. And an estimate assembled from regulatory filings is not a market census. What survives both qualifications is structural rather than quantitative:

A meaningful share of marginal demand for the world’s reference safe asset is now supplied by leveraged, repo-dependent, balance-sheet-constrained intermediaries rather than by patient reserve managers.

This changes the failure mode rather than the debt level. Patient creditors withdraw slowly, for reasons of policy, and announce themselves. Leveraged intermediaries withdraw abruptly, for reasons of margin, and do not. The distinction yields a definition the rest of this chapter needs:

Sovereign Credit Safety vs. Collateral Stability

Credit safety is the probability that a sovereign obligation is repaid in nominal terms. Collateral stability is the reliability of that obligation’s price over the horizon at which it is actually used—as margin, as liquidity, and as the reference rate for everything else. An instrument can be entirely money-good and simultaneously an unreliable short-horizon hedge, because the institutions financing it are leverage-sensitive even when the issuer is not.

The two properties are routinely conflated, and the conflation is load-bearing for a great deal of portfolio construction. A saver who holds sovereign paper for credit safety and receives collateral instability has not been defrauded; they have been sold one property and have priced the other.

A measurement note.

The same research [Barth et al. 2025] finds these positions are largely invisible in the Treasury International Capital data, which understate Cayman holdings by roughly $1.4 trillion at end-2024 because repo collateral transfers break the reported chain of ownership. Because the TIC data feed the Financial Accounts of the United States, and because the household sector’s Treasury holdings are computed there as a residual, the error propagates: published household holdings are overstated, and the measured personal saving rate was overstated by roughly 2.1 percentage points during 2023–24.

This is worth dwelling on, because it is the exact pathology this thesis asks protocols to avoid. A structurally important flow grew to trillion-dollar scale inside the most heavily instrumented financial system in the world, and the official accounts went on publishing confident figures about household balance sheets the entire time. §23: Extended Telemetry argues that telemetry must be built so unmeasured dependencies surface rather than disappear into a residual. Here the residual was called “households.”

Automatic Demand Is Not an Anchor

The leveraged residual is only half of the buyer-quality problem. The other half is that a large, persistent bid can still fail to warehouse interest-rate risk.

Green’s later work on the long end of the Treasury market [Green 2026b] distinguishes two kinds of buyer. The automatic buyer follows a rule—default enrollment, a target-date glide path, a market-value-weighted bond index—and continues to purchase regardless of whether the offered yield compensates the risk. The marginal buyer absorbs whatever duration remains after automatic and other captive allocations, and therefore sets the clearing price. Substantial gross demand can coexist with an unstable market if most of that demand is short-maturity, leveraged, one-time, or insensitive to an increase in risk.

Automatic vs. Marginal Buyer

The automatic buyer supplies demand by formula. The marginal buyer sets the price by absorbing the residual. Quantity of bids and quality of risk absorption are different objects, and they are routinely reported as if they were the same.

The replacement of Japanese life insurers and quantitative-easing central banks—institutions with enduring liabilities, or with a policy mandate to hold duration—by American retirement defaults is therefore not a like-for-like substitution. The old buyer bought duration to match a promise. The new buyer follows an index weight.

The insurer had liabilities. The index has weights.

That sentence is the same transition this thesis already names in verification and in retirement architecture more generally: a rule can execute faithfully while producing the wrong aggregate result. Verifiability guarantees that a rule was followed. It does not guarantee that the rule was wise.

A Fixed Dollar Inflow Is Not a Fixed Risk Bid

Market-value-weighted bond indexes allocate each new contribution according to current market value. When a long bond’s price falls, its index weight falls, and a smaller share of every subsequent inflow is directed toward it. Existing holdings are not sold by the rule; incremental stabilizing capacity is. The automatic system therefore absorbs less interest-rate risk per dollar precisely when long bonds have cheapened and the residual is largest.

The relevant unit is not dollars of demand but DV01: the dollar change in value produced by a one-basis-point change in yield. Automatic inflows can remain large in cash terms while DV01 absorption shrinks. The residual then falls more heavily on the leveraged and discretionary buyers already identified above.

This is a compositional analogue of the Market Realization Plane (§10b: Why Hold the Base Asset?). In equities, market-cap weighting directs more dollars toward names that have already risen. In bonds, market-value weighting directs less duration risk toward names that have already fallen. In both cases the governing rule is price-insensitive in the ordinary sense and procyclical in risk terms. Target-date rebalancing, which is often cited as a stabilizer, is itself a relative-performance rule: it fires when stocks and bonds diverge, and it contributes little when they fall together. Appendix I: Scenario Analysis: The Collateral Loop Under Stress traces that case as a scenario.

None of this requires attributing any particular yield move to the mechanism. The structural claim is narrower and is the one used here: the quality, horizon, funding stability, and risk-bearing capacity of demand for long-duration sovereign paper have deteriorated, even where the quantity of automatic buying has not.

The same distinction splits credit risk from duration risk on the corporate side of the AI complex. Investors may still believe a hyperscaler will repay and still refuse to lend for thirty years at the offered yield. The first crack in the collateral loop (§2: The World Forces New Monetary Primitives) need not be an equity collapse. It can be the loss of quasi-sovereign status for long corporate paper—a maturity-specific scarcity of buyers, not a scarcity of nominally safe claims.

The Rule Manufactures the Residual

The two halves of the buyer-quality problem have so far been stated separately: a leveraged residual that withdraws for reasons of margin (§2: The World Forces New Monetary Primitives), and an automatic bid that follows a rule regardless of the yield offered (§2: The World Forces New Monetary Primitives). The Board’s own staff have shown that they are one chain. Barth, Kahn, Monin, and Sokolinskiy [Barth et al. 2024] find that variation in mutual funds’ Treasury-futures positions drives much of the time-series variation in aggregate Treasury-futures open interest—more than sixty percent of the rise they examine—and that those positions are largely associated with funds reaching for duration: tracking a benchmark’s duration while holding cash portfolios tilted toward higher-return, lower-duration assets such as mortgage-backed securities and equities. The duration the benchmark demands and the cash portfolio does not hold is bought synthetically, in futures. Someone must take the other side, and the paper shows who does: leveraged funds, short the future, long the cash bond, financed in repo.

The index does not hold the duration. It orders it.

That is a stronger statement than the one §2: The World Forces New Monetary Primitives made. The automatic buyer does not merely fail to warehouse interest-rate risk; it commissions the leverage that warehouses it in its place. The chain runs benchmark duration target $\to$ cash portfolio tilted elsewhere $\to$ long futures $\to$ leveraged short against the cash bond $\to$ repo. Its two ends have opposite withdrawal properties. The rule-follower at the head cannot withdraw, because it is following a rule; the intermediary at the tail can withdraw in an afternoon, because it is following a margin call. The demand that appears most stable is therefore the demand financed most fragilely, and the two withdrawals that §2: The World Forces New Monetary Primitives distinguished—slow and announced, abrupt and silent—are not alternatives but stages of one structure.

The structure has been stress-tested once. Kruttli, Monin, Petrasek, and Watugala [Kruttli et al. 2021] document that hedge-fund gross Treasury exposures doubled between 2018 and February 2020, to roughly $2.4 trillion, and that the funds concentrated in the cash–futures basis faced the greatest margin pressure in March 2020 and cut Treasury exposure and repo borrowing the most. A position that is nearly riskless in duration terms—long the bond, short the future—is not riskless in funding terms, and when it is unwound the thing sold is the cash bond. The reading is the one this chapter already gives for collateral stability, now with its mechanism attached: in the ordinary regime the chain absorbs duration and damps its price; in the stressed regime the same chain sells duration and amplifies it. It is a shock absorber that works until its mounting fails, and the mounting is the repo market.

Description becomes instruction.

Green’s later formulation of the same mechanism [Green 2026g] supplies the sentence that generalizes it. A market-value-weighted index answers one question—how much eligible debt is outstanding, at what market value—and answers it correctly. It does not choose how much duration an investor should hold, what curve exposure, what convexity, what share of mortgage optionality; those emerge from what issuers sell, how borrowers refinance, and where prices move. The index describes the market. Once capital is managed relative to the index, the description becomes the instruction, and, in the essay’s phrase, the seller of credit increasingly determines the portfolio of the buyer of credit. Nobody sat down and chose the benchmark’s duration in any year; it is whatever issuance, refinancing, and prices have made it. What the thesis takes from it is structural: a rule that enough capital follows stops describing a market and starts being one.

That is the human-institutional form of the concentration the threat model names for agents—holders that vary in their rules against holders that are instances of one (§4: Threat Model)—and it is why the Native Policy Concentration Ratio of §23: Extended Telemetry is read as a benchmark’s native analogue: the question it asks of a reserve cohort is the question this paragraph asks of the bond market.

Why the macro model has states and not a threshold.

The unit this section has insisted on is DV01, and the mechanism above says where the binding constraint lives. It is not the stock of debt against output. It is the incremental interest-rate risk that must be absorbed in a period, set against the capacity of the absorbers: unlevered demand from institutions with liabilities to match, automatic demand that will buy at any yield but less DV01 per dollar as prices fall (§2: The World Forces New Monetary Primitives), and the levered intermediation that services the gap—itself bounded by repo availability and cost, dealer balance sheet, margin, and volatility. While that ratio is comfortable, quantities of issuance that a discretionary-buyer model would call unabsorbable clear without incident. As it approaches capacity the system becomes nonlinear, and it does so without warning in the yield, because the absorption that has been working is the leveraged kind. This is why §26: Adoption Curve & Ecosystem Dynamics specifies states with observables rather than a debt ratio with a trigger, and why the plumbing series in Appendix I: Scenario Analysis: The Collateral Loop Under Stress are read in DV01 and by counterparty class rather than in dollars issued. Central clearing of cash and repo transactions, mandated to phase in through 2026 and 2027 [SEC 2023], changes the denominator: netting reduces the balance sheet each unit of intermediation consumes, which moves the capacity wall outward. It does not remove it, and the thesis tracks where it is rather than assuming it has gone.

What this does not establish.

Neither Board paper shows that benchmark mechanics set the level of the ten-year yield, or explain any given day’s move, and Green’s essay says so before proposing the decomposition that would settle it: net demand for Treasury interest-rate risk by tenor, in DV01, separated into benchmark-sensitive, leveraged, foreign, domestic real-money, and central-bank demand against issuance, with a published residual. Until that is run, the mechanism is established and its dominance is not. The claim that mandate-followers and their arbitrageurs are the market’s dominant voice is quarantined in Appendix I: Scenario Analysis: The Collateral Loop Under Stress as a conjecture awaiting its own test, and no move in yields, including those contemporaneous with this writing, is attributed to it here. The thesis adopts the chain and the question. It does not adopt the answer.

The Collateral Loop

The preceding sections describe a sovereign managing the demand side of its own debt market. That description is incomplete in one important way. It treats asset prices as something the sovereign observes and occasionally influences. Increasingly they are something it depends on [Gromen 2025–26].

Collateralized Sovereign Stack

An arrangement in which a state’s effective fiscal capacity depends not only on taxation and monetary authority but also on the market value of the asset complexes it regulates—because those values determine taxable financial income, consumption through the wealth effect, foreign capital inflows, private credit creation, and the collateral against which the private sector borrows.

This Is Not an Eighth Layer

The collateralized sovereign stack is a description of the incumbent order. The seven layers of §5: Layers of the Cypherpunk Stack are a specification for something to be built. They are different kinds of object and are not to be merged: no layer of the cypherpunk stack produces equity valuations or repo collateral, and no amount of Layer 0 capacity substitutes for a functioning sovereign balance sheet. The duration warehouse of §2: The World Forces New Monetary Primitives belongs in the same diagnostic column. The parallel drawn in this section is diagnostic, not architectural. The same caution applies here that §10b: Why Hold the Base Asset? applies to market realization: some things are properly modelled around the stack rather than within it.

The Circuit

Four dependencies, each individually unremarkable, close into a loop.

1. Asset values support receipts. Capital-gains realizations, taxes on stock-based compensation, corporate profits, and consumption financed by the wealth effect are all functions of market levels rather than of output alone.

2. Receipts support fiscal capacity. A weaker receipt base widens the deficit and increases the quantity of debt that must be placed, independent of any change in policy.

3. Fiscal capacity and central-bank support sustain Treasury market functioning. The reference asset stays reference-grade only while it can be financed, cleared, and repo’d without incident.

4. Treasury yields sustain asset values. The long yield is the discount rate against which every other claim is priced, and Treasuries are the collateral beneath most private credit.

Step 4 returns to step 1. The system is reflexive, and the reflexivity is not a pathology introduced by bad policy—it is the ordinary structure of a financialized economy with a large sovereign balance sheet.

The state has become partially dependent on collateral values it is nominally free to let the market determine.

Why This Explains Balance-Sheet Repression

§2: The World Forces New Monetary Primitives describes repression routed through collateral rules, capital treatment, reserve mandates, and custody defaults, and asks why a sovereign would choose that route over the blunter instruments. The loop answers the question.

Every classical adjustment mechanism attacks the loop somewhere. Austerity cuts receipts by cutting activity. Higher policy rates raise the discount rate and depress collateral values directly. Explicit yield-curve control announces the constraint and invites a test of it. Outright default is the one instrument a sovereign that borrows in its own unit never needs, which is why it is the one it reaches for last.

Balance-sheet repression is attractive precisely because it is the only lever that manufactures demand for sovereign paper without requiring collateral values to fall. It operates on who is obliged to hold the debt rather than on what the debt is worth. That is not a marginal preference; it is close to a dominant strategy, and it is the reason to expect this specific form of repression rather than its historical predecessors.

For a monetary stack the consequence is the one already stated in §2: The World Forces New Monetary Primitives: the adversary is substitution rather than prohibition. The loop shows why substitution is the instrument of choice.

A Branch in the Playbook

§2: The World Forces New Monetary Primitives runs from debt through repression to a flight toward neutral, verifiable rails. Nothing in this section changes that direction. It adds a branch at the transition.

The playbook implicitly assumes the standard crisis sequence, in which risk-off lowers long yields and buys the sovereign time. §2: The World Forces New Monetary Primitives gives a reason that sequence can invert: when marginal demand is leveraged, a volatility shock converts buyers into forced sellers, and the flight to safety can be followed by a rise in long yields rather than a fall. Conventionally safe and conventionally risky assets then fall together.

The branch matters for savers rather than for forecasters. In the standard sequence, duration is the hedge. In the inverted one, duration is a second position in the same risk, and the only assets that behave differently are those whose value does not derive from someone else’s balance sheet. That is an argument about the structure of neutral collateral, and it is the same argument §3: First Principles: What a SoV Must Survive makes on other grounds. Appendix I: Scenario Analysis: The Collateral Loop Under Stress traces the mechanism in full; consistent with the epistemic policy of this document, it is treated there as a scenario rather than a projection.

Why AI Sits at the Center of the Loop

The loop would be an abstraction if the assets inside it were diversified. They are not. One complex currently touches every step, which is what makes a sectoral repricing a sovereign question rather than an investor’s question. The roles below are landing zones within the incumbent macro order—where the consequences of the AI complex show up—and not layers of the architecture this document specifies. Several carry names that resemble layers of the cypherpunk stack because the same physical constraints bind in both places, which is a fact about electricity and silicon rather than a claim about architecture.

One complex, acting simultaneously at every step of the loop. Concentration of this kind is what converts a sectoral repricing into a fiscal event. The last row is the one this section had not named: the same physical assets that load the collateral loop must be financed across decades, and the buyer architecture of §2: The World Forces New Monetary Primitives is becoming less able to hold that interval.

Two conclusions follow, and they point in opposite directions.

The first is defensive. A significant repricing of this complex would not be contained to technology equities. It would weaken collateral, receipts, consumption, and foreign inflows simultaneously, and it would do so while the mechanism in §2: The World Forces New Monetary Primitives was making sovereign paper a less reliable hedge. Appendix I: Scenario Analysis: The Collateral Loop Under Stress traces that chain.

The second matters more for this thesis. If the complex is fiscally load-bearing, the state acquires a strong interest in preventing its repricing—and the instruments available for that purpose convert private infrastructure into protected public infrastructure. §4: Threat Model treats this as an adversary class, because from the standpoint of an open stack it is one. The danger to a neutral, verifiable monetary rail is not that the incumbent AI complex fails. It is that the incumbent AI complex is rescued, and the terms of the rescue determine who is permitted to compute.

Sovereign maturity transformation.

If private buyers will not absorb long duration at politically acceptable yields, the least overtly coercive immediate option is to issue bills and accept rollover risk. That does not extinguish duration. It transfers it from investors onto the sovereign’s refinancing calendar. A state that finances long-lived infrastructure, defense, or industrial investment with short-term public liabilities has performed, at civilizational scale, the maturity mismatch that periodically destabilizes banks. Gromen’s fiscal-dominance argument [Gromen 2025–26] describes where that path ends. The buyer-quality argument of §2: The World Forces New Monetary Primitives describes why it begins: the private duration warehouse is inadequate. Neither claim is load-bearing for the red lines. Both belong in the diagnosis of the incumbent order.

Two Kinds of Duration

Duration-neutrality is a first principle of the monetary object (§3: First Principles: What a SoV Must Survive). It is not a complete account of time.

There are two different objects, and they have been sharing one word.

Duration of the Claim vs. Duration of the Project

Duration of the claim is the interest-rate exposure of a promised cash flow—a coupon the state can pin, a bond whose real return repression can drive negative. A repression-resistant store of value must not be this instrument.

Duration of the project is the years between committing present resources and receiving the output of a plant, grid, fab, reactor, or data center. Someone must warehouse that interval. A rule that allocates by index weight is not that someone.

The current reserve-asset system uses one instrument, the long Treasury, for both functions: politically usable collateral and the conversion of present savings into long-lived public and private capacity. §2: The World Forces New Monetary Primitives shows the second function weakening while the first is still treated as if it were intact. Gold may be superior as politically neutral collateral because it is not another sovereign’s liability and has no refinancing requirement. It does not, by itself, finance a reactor. Bitcoin is in the same position on a digital rail. Proofs can make construction progress, energy receipts, covenants, and restructuring states checkable. They cannot delete time risk, inflation risk, construction risk, or political risk.

Neutral settlement is not a substitute for capital formation. A post-fiat system that cannot finance duration is a savings technology, not a complete political economy.

The design instruction is therefore an unbundling, not an expansion of the token:

• Reserve collateral should remain duration-neutral, politically hard to pin, and bearer-capable.

• Duration finance should remain labeled credit: underwriting, covenants, maturity transformation, loss-bearing equity, and institutions that can survive marks.

• Proofs audit the credit. They do not become the bond.

A post-fiat order that cannot tell collateral from credit will recreate opaque leverage. A post-fiat order that refuses credit altogether will not build Layer 0. The completeness objection is stated and answered in §30: Objections & Responses. The closed-stack comparison in §29: The Closed Sovereign Stack names the existing answers: commanded balance sheets in one system, retirement defaults and leveraged intermediaries in the other. Neither is a cypherpunk layer.

This Is Not a Tenth SoV Requirement, and It Is Not a Layer 7

Time is a function of the political economy. Making it a property of the monetary object would undo duration-neutrality. The collateralized sovereign stack remains a diagnosis; Layers 0–6 remain a specification. The duration warehouse sits in the diagnosis, beside the collateral loop, not inside the stack.

Where the Triad Enters

The diagnosis is now complete enough to say what it asks of a store of value. If “safe” nominal assets become de facto taxes on savings, capital does not protest. It routes to assets whose issuance cannot be decreed and whose verification is cheap and public. In that environment, a credible store of value must:

• Avoid being a duration instrument whose real return can be pinned negative by policy; and

• Live on rails that cannot easily be gated by a single jurisdiction.

This is where the triad enters:

• Privacy as an anti-seizure hull that makes savings bearer-like again.

• Proofs as cheap public verification that replaces platform vouching.

• Compute as a non-coupon revenue base whose clearing price floats with nominal budgets rather than being pegged by decree.

Each primitive is designed to remain neutral and verifiable even when traditional financial infrastructure is weaponized.

Social: The Web’s Trust Default Has Flipped

The same technological forces reshaping money are also reshaping trust.

The web used to borrow its epistemic norms from broadcast: seeing was believing, and the job of editors was to gate what got seen. Deepfakes and generative models invert that. AI systems can now synthesize faces, voices, and scenes that are indistinguishable from genuine footage to lay observers. Exposure itself does not inoculate people; survey data across multiple countries [Ahmed et al. 2024] suggests prior exposure to deepfakes can actually increase susceptibility to misinformation.

In parallel comes the liar’s dividend [Chesney & Citron 2019]: once everyone knows deepfakes are possible, any inconvenient real video can be dismissed as fake. The result is a crisis of knowing [UNESCO 2025], not just a rise in error rates.

This is not entirely new (authoritarians have been editing history since Stalin airbrushed Trotsky out of photos), but the cost curve and scale have changed. Deepfake tools now let anyone with a laptop fabricate a leader conceding an election, a riot that never happened, or “footage” of a historical event that subtly rewrites who was present and who was not. The fear is not just short-term hoaxes; it is revisionist history at machine speed, where archives, livestreams, and “receipts” themselves become contestable. In that world, the substrate of trust shifts from memory (“I saw the clip”) to mechanism (“I can verify how this artifact came to be”).

At the same time, the feeds themselves are no longer obviously “neutral pipes.” Investigations and hearings around government–platform coordination [Twitter Files 2022–23] [House Judiciary 2023] [State Dept. 2021–25] have documented extensive relationships between agencies, NGOs, and major platforms for content moderation in the name of combating misinformation and foreign influence. Whether one views this as necessary harm reduction or problematic overreach, the empirical point stands: the old story that your feed is simply “what’s popular” no longer survives contact with the evidence. What you see (and do not see) is the result of political, institutional, and algorithmic priors you do not control.

Content authentication standards such as C2PA [C2PA spec] and watermarking prototypes exist on paper, but real-world implementations are voluntary, inconsistent, and often invisible to end-users. Tests across major platforms such as the Washington Post’s 2025 C2PA test [Harwell & Patton 2025] show provenance metadata being stripped in transit, and the few disclosures that do survive are buried in UI that most users never touch.

In other words, the trust default has flipped. The practical baseline for online media is now “untrusted unless proven,” and the “proven” part cannot safely be left to platform labels or government–platform partnerships. Platform UX is inconsistent; cryptographic provenance and computation proofs are the scalable, cross-platform backstops.

In a SoV context, this matters because monetary systems sit on top of communication systems: if claims about ownership, origin, and behavior are cheap to fake and expensive to audit, then both money and memory become soft targets. Anchoring value in receipts that anyone can verify, rather than narratives that someone must vouch for, is the one equilibrium that does not depend on who is vouching.

Recent work constructs PoUW for arbitrary matrix multiplication [Komargodski et al. 2025] with verification that is asymptotically cheaper than production ($O(n^2)$ verify vs. $O(n^3)$ produce, via Freivalds-type randomized checking [Freivalds 1977]; the miner’s $1+o(1)$ overhead relative to naive multiplication vanishes relatively as problem size grows — a constant-factor gap remains at finite $n$, so “asymptotically” is doing real work here and the finite-$n$ constants are what VerifyPrice measures). Matrix multiplication is the operation that bottlenecks modern AI. This is the asymmetry PoW always needed: hard to produce, cheap to check.

When verification is very cheap relative to production, honesty becomes a market equilibrium and commodities emerge (standardized proofs, verified FLOPs) that can be priced, saved, and eventually used as collateral and stores of value.

Political: The Participation Line Has Broken

Financial repression becomes politically durable when enough citizens lose the capacity to participate. The relevant threshold is not only poverty, but participation: the level of redundancy at which a household can fail, retry, move, learn, transact, save, form families, and refuse coercive terms.

Participation Line

The household or organizational threshold below which a person or firm lacks the redundancy to act freely across time—to fail, retry, transact, move, learn, refuse coercive terms, or survive shocks.

Below that line, legal rights remain on paper while agency disappears in practice. Citizens become managed objects: eligibility files, credit scores, benefit cliffs, tax audits, healthcare forms, student-debt balances, real-name accounts, and platform identities. The state does not need to ban exit if every practical route requires permission.

The line can also be crossed from above. A person who retains formal title over agents that act for them, but cannot revoke, cap, or re-key those agents through the protocol, holds standing on paper and agency in nobody’s hands. §4: Threat Model names that condition and treats it as a failure kind; §5b: The Delegation Economy: Who Buys Verification says what enforcing standing through the stack would require.

The line is also approached gradually, and the gradient is where the political danger concentrates. A household can remain formally inside the system — banked, employed, housed, consuming — while becoming progressively less able to accumulate: employed but not in a viable career, housed but unable to save, consuming through debt, technically solvent but one shock from default. Aggregate statistics register none of this until households cross an actual delinquency or unemployment threshold, because the aggregates are dollar-weighted and the deterioration is person-weighted (“The Vibecession Was Real” [Green 2026e]). Call the approach path the exhaustion gradient: a household above the participation line today can be on a trajectory that reaches it, and the aggregates will look resilient the entire way down. This is why participation telemetry must report buffers and trajectory, not just stocks.

In that world, lawful privacy is not romantic secrecy. It is the technical form of the fresh start. Selective disclosure, private settlement, receipt-based reputation, and identity without doxxing are the digital equivalents of bankruptcy, homesteading, and due process: they preserve the option to act without turning the person into a permanent dossier.

Agency-Preserving Infrastructure

Infrastructure that expands a user’s capacity to act without converting the user into a dossier, dependency object, or platform account.

This is why Privacy, Proofs, and Compute are not merely institutional tools. They are agency-preserving infrastructure. The stack must equip citizens and firms to act; if it merely manages them more efficiently, it has reproduced the failure it was built to escape. §33: Conclusion: A Bell Labs for Privacy, Proofs, and Compute returns to this as the thesis’s closing civic frame, and §3: First Principles: What a SoV Must Survive makes it a formal SoV requirement.

New here? Start with the one-minute version.

Tip: hover a heading to reveal its permalink symbol for copying.