The Debt Endgame

The Incoming Fed Chair, Stealth Monetization, and Positioning for the Sovereign-Debt Crisis

TON618 Capital Research · As of June 30, 2026 · Download PDF

A point-in-time read of re-estimated indicators, not a forecast. Data: FRED, U.S. Treasury, CBO, GAO, IMF, ECB, and primary reporting — every figure as-of dated. Fact ("Data shows…") is separated from opinion ("We believe…"). This note was independently fact-checked against primary sources before release; the Source & Verification Note at the end documents that pass and the few items that remain hedged.

§0 The Argument

Every government facing an unpayable debt has only four doors. It can grow out (outpace the debt with real output). It can default (repudiate, explicitly). It can austerity its way out (tax and cut to a primary surplus). Or it can inflate and repress — hold interest rates below inflation and let time quietly transfer wealth from creditors to the borrower. The United States in 2026 has effectively closed the first three doors. It cannot grow fast enough (growth is decelerating even as the debt compounds); it will not default (the dollar's franchise depends on it); and austerity at the required scale is politically unthinkable in a two-tier, low-trust society. That leaves one door. This report is about who walks through it, how, and who pays.

Why this matters — and to whom. Stagflation is the economics of a rock and a hard place: fight the inflation and you risk breaking the debt; service the debt and you feed the inflation. There is no costless exit — and how the authorities pick their way out is not a technocratic footnote but a distributional decision that lands on three sets of stakeholders very differently. The government wants its debt to shrink in real terms without a wrenching vote. The broad public — wage-earners, savers, the young — wants its purchasing power and its shot at ownership preserved. The asset-owning wealthy want their capital shielded from both inflation and confiscation. Growth alone cannot satisfy all three at once, so someone's ox gets gored; the entire question is whose. That is why this note is worth your time even if you never trade a bond on it: the way the debt–inflation–growth trilemma is resolved decides who pays for a generation of deferred choices — and a great deal can go wrong on the way.

The machinery to walk through that door is being assembled in plain sight. A hawkish new Fed Chair, Kevin Warsh, has inherited a debt so large that his own instrument — the interest rate — has become a fiscal weapon pointed at the Treasury (§1–§2, §7). Two developments supply the political cover to ease anyway: tech- and AI-driven disinflation, which lowers headline inflation but, we argue, is captured by the sovereign as room to cut and refinance rather than delivered to households as higher real incomes (§6); and Bessent's stablecoin statecraft, a deliberate effort to manufacture a new, captive buyer for Treasury bills (§8). The cost falls on savers and the young through the oldest mechanism in public finance — the inflation tax. And here is the distinction the rest of the note turns on, argued in full in §10: the inflation tax is not a leveling redistribution but a Cantillon transfer in the wrong direction — value flows to whoever borrows and receives the new money first (the leveraged, and above all the indebted state) at the expense of the savers and wage-earners who hold it last, so it shrinks the debt while widening the gap and raising popular misery, deepening the very inequality it appears to address.

Our house view: the most probable multi-year resolution is financial repression + stealth monetization — not a dramatic default or a Volcker-style purge, but a long, quiet erosion of the real debt onto the balance sheets of anyone holding dollars and bonds. The live data already rhymes with it: the Fed has been cutting (funds rate 3.6%, down 0.7pp over the year) into an inflation impulse that is re-accelerating (inflation diffusion 80% rising), with the economy in the stagflation quadrant and debt at 123% of GDP. A central bank easing into rising inflation at record debt is not a policy error to be corrected; on our reading, it is the policy.

What we conclude, previewed. Two judgments organize what follows. First, the market's favorite escape hatch — that AI-driven deflation will quietly disinflate and grow the problem away — is being sold as a savior it has not earned: the credible productivity estimates span a tenfold range, the disinflation that does arrive is more likely captured by the Treasury as room to cut than shared with households as higher real incomes, and staking the sovereign's solvency on an unproven miracle is a hope, not a plan (§6). Second, when the real options are scored against the three outcomes a society most wants to avoid — violent upheaval, the inflation tax's misery, and worsening inequality — one dominates on the merits: common-sense progressive, wealth-transfer taxation, the only route that resolves the distribution honestly (rich→poor) rather than regressively (through inflation) or violently (through the street). We stop short of endorsing a policy platform — that is not the fund's role — but the analytical case is lopsided, and its tragedy is that the strongest answer is the one being politically and constitutionally throttled (§10), while the worst answer for the public, inflation, requires no vote at all.

The debates we carry honestly as overhangs (and do not resolve by fiat): (1) does Warsh's hawkish opening hold, or does the arithmetic force him to ease? (2) Is AI disinflation real-and-shared or real-and-captured? (3) Is the endgame the 1940s repression template or a 1980s Volcker shock? (4) Does the stablecoin bid add to dollar hegemony or merely backfill a receding foreign one? (5) Does inequality level through the ballot box or the tail-risk of the street? Each has a falsifier, stated where it lives.

§1 The Incoming Fed Chair (Kevin Warsh)

Every regime change starts with a person, and this one starts with a hawk in a trap.

Who he is. Kevin Warsh is Chair of the Federal Reserve, confirmed by the Senate 54–45 on May 13, 2026 — the narrowest margin in modern Fed history — sworn in May 22, succeeding Jerome Powell (whose chair term expired May 15; Powell remains a governor). Warsh was a Fed Governor from 2006–2011 (the youngest ever, appointed at 35, at the center of the Lehman-era Fed), resigned partly in objection to QE2, and came from Morgan Stanley M&A and the Bush National Economic Council; he has since been a fellow at Stanford's Hoover Institution.

What he wants. His stated agenda — from 2024–25 op-eds and his confirmation hearing — is to re-credibilize and "modernize" the Fed: shrink the ~$6.7T balance sheet, end QE dependence, abandon forward guidance and potentially retire the dot plot, replace flexible average-inflation targeting with a strict 2% target, and make the policy rate the primary tool. He treats independence instrumentally — a means to price stability, forfeitable if the Fed strays. At his first FOMC (June 16–17, 2026) he held the rate at ~3.6%, stripped forward-guidance language, declined to submit his own dot, and framed the mandate bluntly — the Fed has "missed for five years, and we're going to fix that." Nine of eighteen participants now signal higher rates in 2026.

Why he is trapped. We read him as a hawk walking into textbook fiscal dominance: ~123% debt/GDP, ~$39T of debt, and Treasury financing skewed to short maturities (§7), so a rate hike hits the government's interest bill almost immediately. A hawk's tools, applied here, worsen the very problem the hawk is trying to look past. The binding constraint is the arithmetic, not his convictions.

The tension that runs through the whole report. As a candidate, Warsh leaned on AI as "a significant disinflationary force" to argue for lower rates (§6). As Chair, his opening act is hawkish credibility. We read the AI-disinflation argument as his eventual face-saving off-ramp — the mechanism by which a self-styled hawk can cut and monetize while declaring victory over inflation. Whether he takes that ramp, and when, is the single largest swing factor in this note.

How markets test new chairs (the historical pattern). A shock reliably arrives early and defines the tenure:

Volcker and Greenspan held or eased decisively; Powell's pivot minted a "put." Warsh's test has not yet come — and which of these he becomes is, for markets, the whole ballgame.

§2 The Dual Mandate at the Extremes

Warsh's problem is not his resolve; it is that his instrument now breaks the thing it is meant to protect.

The mandate. The 1977 Federal Reserve Act charges the Fed with maximum employment and stable prices (since 2012, a 2% PCE target). At the extremes, the policy rate collides with two objectives that are not in the mandate but bind in practice: financial stability and fiscal sustainability.

Do hikes break things? Yes — three failure modes. (i) SVB and the US regionals (March 2023) — ~500bp of hikes impaired long-duration bond books, an uninsured-deposit run crystallized the losses, and SVB, Signature, and then First Republic failed. (ii) The UK gilt/LDI crisis (Sept–Oct 2022) — 30-year yields rose >100bp in four days; pension LDI levered ~3:1 faced margin calls and forced selling until the Bank of England intervened (up to £65bn pledged, ~£19.3bn bought). (iii) The repo spike (Sept 17 2019) — SOFR jumped 2.43%→5.25% (10% intraday) as reserves fell below ~$1.4T, requiring $75bn/day of NY Fed injections. Rate regimes break the most-levered, least-hedged plumbing first — and the more debt outstanding, the more plumbing there is to break.

Do hikes worsen the deficit? Fiscal dominance. With debt near $39T and roughly a third repricing each year (§7), higher rates raise interest expense faster than they cool a supply-driven inflation. This is fiscal dominance (Sargent–Wallace, 1981): debt service so large that tightening is self-defeating, pressuring the central bank to tolerate inflation or cap rates. That the US is now fiscally dominant is our judgment, not settled fact — but it is the judgment the rest of the report rests on.

"Monetizing savers" — financial repression. The precedent is the 1945–1951 Fed–Treasury peg (a ~2.5% long-bond ceiling, ~⅜% bills) that held down WWII debt service and produced deeply negative real rates until the 1951 Accord. Reinhart & Sbrancia's "liquidation effect" formalizes what followed: negative real rates plus captive buyers plus capital controls liquidated government debt at ~3–4% of GDP per year in the US and UK, with real rates negative roughly half the time across advanced economies from 1945–1980. That is monetizing savers — a tax on bondholders and cash-holders levied through the price level rather than the tax code. Note precisely what it does and does not do: it repairs the sovereign's balance sheet by destroying the real value of the saver's, and it immiserates anyone on wages or fixed income — but it moves no wealth toward the poor. It is a stealth transfer from the public to the state, not a leveler. Today the lever is not yet pulled (real fed funds are modestly positive, the real 10Y is ~2.16%), but it is the lever the arithmetic points to.

§3 The Two-Tiered Economy & the Cantillon Effect

Repression does not fall on an undivided nation. It lands on a society already split into those who own the assets and those who earn the wages — and that split determines who can absorb the coming inflation tax and who cannot.

The split, quantified (Fed Distributional Financial Accounts, Q1 2026). The top 1% hold 31.6% of US net worth (up from 22.8% in 1989); the top 10% ~67.9%; the bottom 50% just 2.5% (down from 3.5%). The top-1%-to-bottom-50% ratio has widened from ~6.5:1 to ~12.6:1 in one generation.

The mechanism — Cantillon. Richard Cantillon observed (~1730) that new money is not neutral: those nearest the injection — the financial sector and asset holders — transact at old prices before the new money bids prices up, capturing an arbitrage the wage-earner, who receives the money last, never gets. Quantitative easing was the textbook modern channel: the Fed bought bonds from banks, lifting asset prices held disproportionately at the top long before (if ever) the money reached wages.

The single number that anchors this report. Since 2008 the Fed balance sheet went from $0.91T to a ~$8.97T peak in 2022 (now $6.74T); the S&P 500 rose roughly 11× (676 in March 2009 → 7,499 on June 30, 2026); yet real median weekly earnings rose just +12% in 47 years ($335 in 1979 → $376 in Q1 2026). Assets up ~11×; real wages up 12%. The two-tier economy is not a talking point — it is in the data, and it is the reason the inflation tax and the instability tail (§10) are the same story told from two ends.

§4 The Four Quadrants & What Stagflation Is

Before prescribing, locate the patient. The regime the Fed is acting in is not a soft landing — it is the one regime where its two goals pull against each other.

The live read (our four-quadrant nowcast, June 2026). The economy sits in Quad 3 — Stagflation: growth decelerating (growth-z −0.11) while inflation accelerates (inflation-z +1.33). Translated through Dalio's All-Weather lens: **gold, commodities, and TIPS carry; nominal stocks and bonds struggle together.**

Reconciling the two readings honestly. Our indicator diffusion shows activity levels still broadly expanding (growth 75% rising, inflation 80% rising), while the quadrant model reads the second derivative — the rate of change — where growth is losing momentum and inflation is re-accelerating. Both are true and not contradictory: the economy is still growing but decelerating as price pressure builds. That is the texture of early stagflation, not late.

What stagflation is. The simultaneous combination of stalling growth (or rising unemployment) and rising inflation — the breakdown of the Phillips-curve tradeoff. It is the one regime in which the Fed's two mandate goals point in opposite directions, so there is no costless policy path. Which is precisely why a constrained hawk reaches for cover (§6) rather than a clean choice.

§5 Stagflation: The Historical Record

The 1970s are the rhyme everyone reaches for. The 1940s are the one that actually fits.

What it was like. The 1973 OPEC embargo drove CPI from ~3.4% (1972) to 12.3% (1974) amid the deepest recession since the 1930s. The 1979 oil shock pushed CPI to a 14.8% peak in March 1980; Volcker took the funds rate to ~20–21%, producing a double-dip recession and 10.8% unemployment in late 1982; the misery index crested near 20–22%. The cure worked — CPI fell to ~3% by 1983 — but at the cost of two recessions and a decade of lived pain.

What happened next — two templates. The Volcker disinflation (a credibility shock, real pain, resolution) versus the 1940s repression (inflate it away quietly, no purge). With record debt/GDP and a fiscally constrained Fed, we judge today rhymes more with the 1940s than with 1979 — inflation more likely tolerated and eroded than crushed. The steelman against us is exactly Warsh's stated intent: a credible, independent Fed could still choose the Volcker road, and says it wants to.

What won (real returns, ~1973–82; the window is endpoint-sensitive):

The pattern — real assets and pricing power win, nominal duration and the 60/40 lose — is the empirical backbone of the positioning in §14.

§6 Tech/AI Deflation: Savior or Cover?

Here is the hinge of the whole thesis. If AI disinflation is real and shared, the debt can be grown out of and we are wrong. If it is real but captured, it becomes the alibi for monetization. The evidence points to captured.

(a) Historical tech deflation was real and measurable. Quality-adjusted US durable-goods prices fell ~2.8%/yr from 1995 to Covid. The Boskin Commission (1996) found the CPI overstated inflation by ~1.1pp/yr, roughly half from unmeasured quality and new goods — the tech channel. Oliner–Sichel (2000) attributed ~two-thirds of the late-1990s productivity acceleration to IT, with computers and semiconductors ≈ half of economy-wide TFP growth in 1996–99. But the payoff lagged the technology by ~20 years (Solow's paradox). A defensible synthesis: tech shaved on the order of ~0.5pp/yr off core inflation in 1995–2005 — a real supporting actor, alongside globalization and credible policy, not the whole story.

(b) Is AI big enough? Size it; don't guess. The debate spans an order of magnitude. Goldman (2023) sees ~+1.5pp/yr US productivity and +7% global GDP; Acemoglu (NBER 2024) sees ~0.66% TFP total over a decade (~0.07pp/yr), because only ~20% of tasks are exposed, ~23% profitably automatable, at ~27% cost saving. Bracketing these gives a ~0.1–0.7pp/yr drag on core inflation, and the upper bound requires Goldman-scale adoption and competitive pass-through. Steelman the skeptic: rebound/Jevons effects and new-task creation offset price declines; the CPI is ~60%+ services (shelter alone ~34%), the categories least exposed to AI; and the AI capex build-out (~$4T globally by 2030) is inflationary for chips, copper, and power near-term — Powell himself said data centers are "probably driving inflation" in the short run. Present the poles; never average them — the gap between them is the debate.

(c) Will the public even see it? The capture thesis. This is our central contention — and, unusually for a forward-looking claim, it is well-supported by the current record. The live evidence is on the record: there is an open split inside the Fed — Warsh frames AI as "a significant disinflationary force" arguing for lower rates, while Goolsbee warns of stagflation from the build-out. Our argument: if AI lowers headline inflation, it hands the Fed cover to cut nominal rates even as the capex boom runs hot — and lower rates lower the government's real debt-service burden on ~$39T. The disinflation is thereby captured by the sovereign as a smaller real debt, not delivered to households as higher real incomes. Steelman the counter: in competitive markets, cost savings pass to consumers as lower prices, so real wages rise even at flat nominal wages — as they did in the late 1990s. The capture thesis is strongest under concentrated market power and a fiscally dominated Fed; weakest under intense competition and genuine Fed independence. It is a risk to flag, not a certainty to assert — but it is the risk this entire report is organized around.

The bottom line on AI: do not depend on it. As a policy or investment premise, "AI will grow us out of the debt" fails on three counts, each established above. It is unproven — the credible estimates of its productivity punch differ by an order of magnitude (Goldman ~1.5pp/yr vs. Acemoglu ~0.07pp/yr), and the bulk of the CPI sits in AI-unexposed services. It is near-term inflationary — the ~$4T build-out bids up chips, power, and copper before it disinflates anything. And even in the optimistic case, the disinflation is more likely captured by the Treasury than shared with households. The prudent posture is to treat AI deflation as an upside option, not a load-bearing assumption; building a nation's solvency on it is the fiscal equivalent of writing naked options on a miracle. And AI is only the nearest miracle: the same faith extends to a broader techno-abundance stack — humanoid robotics, orbital data centers, asteroid-mined resource abundance — any of which would, if it arrived at scale, supply the missing growth engine (§9). But the logic only hardens as the horizon lengthens: the more speculative and distant the breakthrough, the less it can be a load-bearing assumption under a debt that compounds now. Hope is not a maturity schedule. That is precisely why the resolution worth arguing for is one that does not require the miracle to show up (§13).

§7 The National Debt: Interest Expense & the Rollover Wall

This is the section that forces every other hand. It is the "growth-sucking sound" made explicit.

The scale. Total debt is $39.35T (June 29 2026) — $31.62T held by the public — against ~$31.9T GDP, i.e. ~123% debt/GDP. FY2025 net interest was $970.4B, up 10.1% year-over-year.

A precision the thesis must get right. Net interest now exceeds national-defense outlays (~$917B, Treasury FY2025) — and CBO projects it does so every year through 2035 — but it is the third-largest line item, still behind Social Security ($1,580.7B) and Medicare (~$997B). The correct claim is therefore "interest already exceeds defense and is closing on the largest entitlements," not "interest exceeds all entitlements." (Gross federal interest runs ~$1.2T; the $970B is the CBO net figure. We report net and disclose gross.)

The rollover wall. Weighted-average maturity is ~6 years — and has actually receded from its ~75-month (20-year) high in 2023, because the Treasury has leaned on short bills rather than terming out — yet ~33% of marketable debt matures within 12 months, bills are ~22% of the stock, ~$9.1T was refinanced in FY2025, and ~$9–10T is projected for FY2026. Because maturing debt reprices from sub-2% legacy coupons to the current ~3.4% average, each rollover ratchets interest expense higher — the mechanical core of fiscal dominance.

The trajectory. Net interest was ~3.2% of GDP in FY2025 (a record), projected toward ~4.1% by 2035, and ~18.5% of federal revenues now (past the 1991 peak), heading toward ~28% by 2055. In our view, the self-reinforcing loop — higher rates → higher interest → wider deficit → more issuance → yield pressure → higher rates — is why the resolution runs through repression, not restraint. This is the driver that also anchors the instability test in §10.

§8 Exporting the Inflation Tax

Here is why stablecoins belong in a note about debt and inflation — and why they are more than plumbing. Like foreign-held Treasuries, dollar stablecoins let the United States offload part of its inflation tax onto people who are not American.

The mechanism: dollar hegemony exports debasement. When the US inflates to shrink the real value of its debt, the loss falls on everyone holding dollar claims — a foreign central bank with Treasuries and an Argentine holding a stablecoin both experience the debasement exactly as a US saver does. This is the modern face of the "exorbitant privilege": the wider the world's holdings of dollars and Treasuries, the larger the foreign share of the American inflation tax, and the less of it falls on US households. Broadening that foreign base is, quietly, a debt-management strategy — and it introduces a fourth stakeholder the domestic frame omits: the world's dollar holders, who absorb what Americans are spared.

The plan, in Bessent's words — and now in law. Treasury Secretary Scott Bessent has made the stablecoin–Treasury link explicit: "we are going to keep the U.S. the dominant reserve currency in the world, and we will use stablecoins to do that" (Mar 2025); a thriving stablecoin ecosystem, he argues, will "drive demand… for US Treasuries, which back stablecoins." The GENIUS Act (signed July 18, 2025) turned intent into structure: payment stablecoins must be 100% reserve-backed in cash and short Treasuries and may not pay interest — converting every token into a captive, structural bid for T-bills, a digital echo of financial repression's captive audience. (It also disintermediates banks — Fed staff modeled ~$190–408B of reduced lending, and officials have warned of deposits in the trillions at risk — but the debt story is the point here.)

The twist: the sophisticated foreign bid is already leaving. The captive audience the US needs is exiting. Foreign investors still hold ~$9.35T of Treasuries (Mar 2026), but only ~28–30% of the marketable stock, down from the mid-30s a decade ago — and the official sector has gone flat-to-negative as reserve managers diversify. China has cut from a ~$1.32T peak (2013) to ~$652B (Mar 2026), an 18-year low; the dollar is ~57% of allocated FX reserves, down from ~71% in 2000; and, tellingly, gold overtook US Treasuries as the largest global reserve asset in 2025 (first time since 1996). The central banks have wised up to the debasement and are walking.

So the real function of stablecoins: recruit a replacement base. Credible projections put stablecoin-driven demand at ~$0.9–1.0T of incremental T-bills by 2028 (TBAC, Standard Chartered); Tether alone already holds ~$141B in US Treasuries, the 17th-largest holder of US government debt. On the funding question we judge, with moderate confidence, that this is a front-end backfill, not an offset — it eases bill plumbing but does not fund the coupon curve, and the Kansas City Fed finds it mostly recycles existing dollars rather than creating net-new demand. Its deeper role is distributional: stablecoins swap fleeing central banks for captive emerging-market retail — the Argentine, Nigerian, or Turkish saver dollarizing-by-app precisely because their own currency is worse, becoming a willing new holder of the American inflation tax because the alternative at home is a worse one. Verdict: stablecoins neither fund the long end nor restore the reserve franchise — but they broaden the foreign base that absorbs US debasement, which is exactly why they matter to the endgame and exactly why they are being pushed. They buy the repression runway more room; they do not buy an exit.

§9 The 1946 Precedent: Same Lever, Missing Engine

The US has faced a debt this large once before — after World War II — and got out of it. That escape is the template every "don't worry, we've done this before" argument leans on. The numbers say the comfort is misplaced.

We are back above the 1946 peak. Federal debt held by the public crested near 106–109% of GDP in 1946 (gross debt ~119–122%; the figures come from CBO/OMB historical tables, which predate the FRED series). It then fell for three decades to ~23% by the mid-1970s. Today debt held by the public is ~123% of GDP — already past the postwar high.

How the last escape actually worked — and it was not mostly inflation. The popular story ("they inflated it away") is wrong. Hall & Sargent's decomposition of the ~80-point 1945→1974 decline attributes it to three cylinders: real GDP growth ~40%, primary surpluses ~43%, and negative real returns via inflation/repression only ~20%. Growth and fiscal discipline each did more than twice the work of inflation — repression was the smallest cylinder of the three. Financial repression was real — the Reinhart-Sbrancia "liquidation effect" eroded ~3–4% of GDP a year — but it was the supporting actor, not the star.

The scaffolding that made even that repression work. From 1942 the Fed pegged bill yields at 0.375% and capped the long bond at ~2.5%, holding them below an inflation rate that peaked near 19.7% in 1947 — deeply negative real yields by design. Bretton Woods capital controls created a captive domestic buyer with nowhere to run, and the peg held until the 1951 Treasury–Fed Accord. Above all, it rode a genuine engine: real GDP compounded ~4.2% a year through the 1950s–60s, powered by baby-boom demographics and postwar productivity.

Why the template does not transfer — the missing engine. Of the three cylinders, only the smallest is available now:

• Growth is gone as a deleveraging force. CBO projects real potential growth of just ~1.7%/yr for 2025–2055 (vs. ~2.4% prior and ~4.3% in the postwar boom), dragged by an aging population and ~0.3%/yr labor-force growth. The cylinder that did the most work in 1946 is the one we most lack.
• Primary surpluses are politically absent — the US runs structural deficits, the opposite of the 1946–74 contribution.
• That leaves repression — the smallest lever — to carry the whole load, and its 1940s scaffolding is gone. Capital is mobile (no controls), so no captive buyer can be manufactured; ~31% of the debt is foreign-held and can walk (§8); ~33% matures within a year, so higher rates hit the budget fast rather than being locked in 7-year paper; and the Fed is independent and un-pegged — there is no standing 2.5% cap to enforce.

The punchline. The postwar escape was a three-cylinder engine in which repression was the smallest cylinder — and today the two larger cylinders, growth and surpluses, are missing while the plumbing is leakier. This is precisely why the market is casting AI as the modern substitute for the postwar productivity boom: the engine that would let repression stay mild and tolerable. §6 is our answer to that hope — it is unproven, and a nation cannot stake its solvency on it. We are trying to run the 1940s playbook on its weakest cylinder alone.

§10 Wealth Inequality: How It Resolves

Extreme wealth inequality resolves, historically, through one of two channels — the redistributive (wealth-transfer taxation, rich→poor) or the violent (upheaval). A third kind of "tax" runs through this report — the inflation tax of the debt endgame — but it is crucial to see that it is not a resolution of inequality at all: it is regressive, it widens the gap, and so it intensifies the choice between the other two. Separating the two taxes is the whole point of this section.

Two taxes, two problems — do not conflate them. In our reading, the inflation tax (§2) is not a leveling redistribution — it is a Cantillon transfer in the wrong direction. Value flows to whoever receives the newly-created money first — risk-taking borrowers, the leveraged, and above all the indebted state — and away from whoever holds nominal claims last: savers, wage-earners, the cash-holding majority. To the saver it is destruction; to the first-receiver, a windfall. Either way it moves wealth toward those already nearest money and assets (Cantillon, §3), which is the opposite of leveling — so it deepens inequality even as it shrinks the debt, and its dominant felt effect on the majority is misery.

One objection deserves a straight answer: the state is ultimately the public, so isn't the inflation tax just the public taxing itself? In aggregate, largely yes — but the transfer is real between the public's cohorts and classes: from savers, creditors, and future taxpayers to borrowers, debtors, and today's beneficiaries of state spending. "Public to sovereign" is shorthand for "one part of the public to another, routed through the state's balance sheet" — the aggregate nets, the distribution does not. Wealth-transfer taxation is the opposite kind of tax: it deliberately moves wealth rich→poor. Only that second tax can resolve inequality; the inflation tax can only deepen it. So the genuine inequality question is narrower and starker than the debt question: redistribute, or rupture.

History's menu. Walter Scheidel's The Great Leveler (2017) finds that extreme inequality has rarely unwound peacefully — historically it took one of four violent "horsemen" (mass-mobilization war, transformative revolution, state collapse, catastrophic pandemic). Piketty supplies the engine: when r (~3–4%) > g (~1.5%), wealth concentrates mechanically. The peaceful counter to the horsemen is redistributive taxation — the channel being actively attempted today (Path A).

Path A — Redistributive taxation (the ballot box)

The true peaceful branch: move wealth rich→poor openly through the tax code. It is live — and its limits are precise.

• The state level is genuinely live and, for income, it works. A coordinated multi-state push (CA, NY, WA, IL, CT, MD, HI, MN, OR, and others) has produced real enactments. Massachusetts' 4% "Fair Share" surtax on income over $1M raised ~$5.7B since it took effect in FY2023 (three-plus fiscal years), beating projections, with no evidence of a millionaire exodus (the >$1M population rose ~39%). Washington enacted a 9.9% capital-gains surtax and, in March 2026, a 9.9% millionaires' income tax — its first broad income tax in nearly a century. California's one-time 5% billionaire net-worth tax qualified for the November 3, 2026 ballot (the proponents estimate ~$100B over five years; the state's Legislative Analyst projects a materially lower, offset-adjusted figure). The Democratic reform impulse is not rhetorical; it is on statutes and ballots.
• **But it taxes the flow, not the stock. What passed were income and capital-gains surtaxes; the wealth / mark-to-market** bills largely stalled — Washington's 0.5% tax on financial assets >$50M failed two years running. Taxing unrealized wealth is where the channel hits its ceiling.
• The federal ceiling is hard — political and constitutional. Federal policy is moving the other way: the One Big Beautiful Bill Act (July 2025) cut top-end taxes and made the $15M/$30M estate exemption permanent (the top 10% capturing ~80% of the benefit). And _Moore v. United States_ (2024) — while upholding a specific tax — saw at least **four justices go on record that the 16th Amendment's "income" requires realization,** leaving a federal wealth/unrealized-gains tax exposed to an apportionment challenge before a hostile Court.
• Capital is mobile; the international record is discouraging. OECD countries levying a net wealth tax fell from 12 (1990) to 4 (2017) — most repealed as unworkable; France scrapped its ISF in 2018 citing flight; Norway's 2022 hike prompted a wave of departures and a contested net revenue loss (though receipts still rose on the broad base). (Steelman for the bull: Massachusetts shows income surtaxes can raise real money with tolerable migration.)
• History cuts the reformer's way only partly. In the mainstream economic-history reading, the 1940s–70s "Great Compression" was war-driven first (total war, wage/price controls, New Deal unionization); progressive taxation sustained the compression more than it created it. Taxation was necessary but not sufficient — a caution against expecting peacetime taxation alone to re-level.

Calibrated read on Path A: the redistributive channel is live and effective for the income flow (state surtaxes), but capped and blocked for the wealth stock (federal, constitutional, mobile capital). It can trim the top of the income distribution — and it makes the transfer visible and voted-on, which is its great virtue over the stealth inflation tax. But it cannot, on current law and politics, reach the wealth stock at scale — which is why the wealth stock keeps concentrating, and why the debt's stock adjustment instead defaults to the regressive inflation tax examined below.

Path B — Violent upheaval

Rather than gesture at "revolution," we test it with a measurable framework: Peter Turchin's structural-demographic theory, whose Political Stress Indicator is the product of three drivers — popular immiseration, elite overproduction, and state fiscal distress. In 2010 Turchin forecast a ~2020 US instability peak.

• Scoring the drivers today: State fiscal distress is the strongest and hardest-data leg — and it is simply our §7 (net interest ~$970B, 3.2% of GDP, deficit ~5.9%). Immiseration is directional (real wages +12% in 47 years vs assets ~11×, §3). Elite overproduction is the softest (credential inflation; ~2× bar-passing law graduates vs openings).
• Warning signs, current: trust in the federal government is 17% (Pew, Dec 2025), near a 70-year low; targeted political violence rose >30% (2024→2025), with Capitol Police threats up 58%.
• Base rates cut both ways. Scheidel raises the tail (peaceful reform is historically rare) yet also implies that, absent one of his four specific shocks, extreme inequality persists rather than resolves — so persistence, not upheaval, is the modal path.
• Steelman the "no revolution" case (weighted heavily): wealthy consolidated democracies essentially never undergo actual revolution — institutional resilience, elections as a pressure valve, aging demographics, and welfare transfers blunt the pressure; Turchin's own model implies violence converts radicals into moderates, predicting elevated unrest, not overthrow; and the theory faces real falsifiability and determinism critiques.

Calibrated read on Path B. For an institutional book we judge the violent path to be a modest, non-negligible tail — not a fat tail, and not "revolution." The investable object is a rising political-instability-risk premium (contested elections, targeted violence, governance dysfunction) that prices into term premia, the rule-of-law premium, and left-tail hedges — not systemic collapse. Turchin's drivers are usefully diagnostic; his determinism should be discounted.

Where the inflation tax fits — it tilts the board, it does not resolve it. In our view, the debt endgame's inflation tax (§2, §7, §11) is not a third resolution of inequality; it is the mechanism by which solving the debt problem worsens the inequality problem. Because inflation destroys real incomes rather than redistributing them (§3: assets ~11× vs wages +12%), draining the ~$39T on savers' backs widens the gap and directly raises the popular immiseration that is Turchin's first driver — feeding Path B while raising the pressure for Path A. So the real dynamic is not "reform vs repression vs revolution." It is this: the inflation tax deepens inequality, and society is then forced to the older fork — the redistributive path (Path A, live but capped) or drift toward the violent one (Path B, thin but fattening). The honest choice is visible redistribution over the stealth, regressive inflation tax — exactly the §13 prescription, and exactly what we doubt the country will choose (§15). The debt is drained by inflation; inequality is not — it is aggravated by the same act, and left for the ballot box or the street to settle.

§11 Synthesis: Stealth Monetization as the Resolution

Pull the threads: a trapped hawk, a backfiring instrument, a divided society, a stagflationary regime, captured disinflation, a compounding debt, a funding scheme that buys time but not an exit, and — the sting in the tail — a social pressure that repression does not relieve but aggravates, because the tax that drains the debt is the same regressive tax that widens the gap (§10).

The escape that requires the least default, the least austerity, and the least legislation is financial repression: hold real rates low, let inflation do the work, and route the loss to savers (§2, §10). The honest alternative — explicit taxation — is live at the state level but capped and constitutionally blocked for the wealth stock federally (§10, Path A), so it trims the flow without draining the debt. Captured AI-disinflation supplies the cover to cut (§6); the stablecoin bid recruits a fresh foreign base to absorb the debasement (§8); a hawkish chair supplies the credibility that makes the eventual easing look like victory rather than surrender (§1). Repression wins by default — not because it is best, but because every honest alternative is harder.

Scenario tree (the probabilities below are our judgment, not a measured output):

• Base (~55%) — Repression / muddle-through. Real rates held low, inflation runs a persistent 3–4%, real debt/GDP grinds sideways-to-down over years. Savers pay; real assets win; the wealth gap widens, nudging the §10 instability premium higher even as it stays a tail — the debt shrinks while the social pressure builds.
• Bull (~20%) — Genuine productivity boom. AI delivers Goldman-scale, shared gains; real wages rise; growth outpaces interest; the debt is partly grown out of. We are wrong in the best way.
• Tail (~25%) — Disorderly adjustment, up to hyperinflation. A failed rollover, a term-premium spike, or a dollar-confidence break forces a hard fiscal adjustment or overt monetization — with the §10 instability premium repricing sharply inside this branch. This is where Warsh's inevitable "new-chair test" (§1) arrives as the shock rather than a market wobble: a hawk facing a buyers' strike in his own bond market is the scenario that defines a tenure.

A word on this tail, because it is easy to under-weight. Repression and hyperinflation are not different monsters; they are the same lever at different settings. Moderate financial repression — a persistent 3–4% inflation eroding the debt quietly — is the base case. But if fiscal dominance fully wins, or confidence in the currency breaks, the dial turns, and quiet erosion becomes the disorderly, self-feeding kind. Crucially, a genuine productivity boom does not guarantee escape: abundant output does not, by itself, stop a government from printing — only the political will to stop does. Hyperinflation is repression that has lost its brakes, and the brakes are political, not economic — which is why we carry this tail at a full quarter of the distribution rather than dismissing it.

Triggers to watch: real-wage growth turning decisively positive (→ Bull); net-interest/GDP breaking above CBO's path or a weak long-bond auction (→ Tail); the curve re-steepening on term premium rather than cuts (→ Tail).

§12 Winners & Losers (in the base case)

If repression is the most likely path, distributional honesty demands naming who it rewards and who it taxes.

Winners.

• The sovereign borrower — the primary beneficiary; the real debt is eroded without a vote.
• Leveraged real-asset owners — homeowners with fixed-rate mortgages, holders of scarce assets, equity in pricing-power businesses; nominal debts shrink while assets reprice up (§3, §5).
• Bitcoin and gold holders — the debasement hedges; gold's reserve-share gain (§9) is the institutional version of the same trade.
• Commodity and energy producers — pricing power in a sticky-inflation regime.

Losers.

• Savers and bondholders — the explicit payers of the inflation tax; negative real returns on cash and nominal Treasuries (§2).
• Wage-earners without asset ownership — the bottom 50% (2.5% of net worth) who receive money last (Cantillon, §3); the group whose immiseration feeds the §10 tail.
• The young and future first-time buyers — priced out by asset inflation, inheriting the debt and its service.
• Pensions and long-duration liability holders — the 2022 gilt/LDI episode (§2) is the warning shot.

The uncomfortable through-line: the winners are those who already own the assets, and the losers are those who only earn wages — which is to say repression widens the very two-tier split (§3) that generates the instability risk (§10). The policy that manages the debt aggravates the social pressure underneath it. That is the report's central tension, stated plainly.

§13 The Least-Bad Path: Why Redistribution Beats the Alternatives

A note that only forecasts, without saying which resolution is actually better, ducks the question its own analysis raises. So we answer it — on the merits — while stopping short of endorsing anyone's policy platform, which is not our role.

Score the options against what a society actually wants to avoid. There are three ways the debt-and-inequality knot ultimately gets cut, and three outcomes nearly everyone claims to want to avoid: violent upheaval (§10), the inflation tax's broad misery (§2), and a still-wider wealth gap (§3). Line them up:

On the merits the answer is not close: explicit, progressive, wealth-transfer taxation is the only resolution that avoids all three bad outcomes at once. It is also the most honest — a transfer that is voted on and visible rather than inflicted by the price level — and it is not utopian: the states are already proving the flow can be taxed and survive (Massachusetts' millionaires' surtax raised ~$5.7B since FY2023 with no measurable exodus; §10). This is not an ideological preference; it is what falls out of scoring the options against the outcomes the country itself says it wants. (Honest fiscal housekeeping helps at the margin — terming out the §7 rollover wall, pro-investment policy to chase the §6 upside — but those ease the debt; they do not decide the distribution. The distribution is the political choice, and taxation is the honest way to make it.)

We should be candid about why the ballot box resists, because it sharpens rather than softens the case. Broad asset ownership — ~62% of households, through home equity and retirement accounts — means a wide electorate fears it could be the target, even when it would not be. That slippery-slope fear is exactly why the stealth inflation tax is politically easier: it is the levy no one has to admit to imposing.

Which points to the design that answers the objection: focus the tax at the very top, where the gains actually went. The rewards of the past two cycles did not accrue evenly across the top decile — they concentrated at the apex. Within the top 1% (31.6% of wealth, §3) the gains are denser still at the top 0.1% and the roughly 990 US billionaires (a record high, per the 2026 Forbes list), whose fortunes are overwhelmingly unrealized founder equity that has never been taxed at all. That is precisely why the live proposals aim at the apex, not the middle: California's measure is billionaires only; the stalled federal proposals set the floor at $50–100M of net worth. A tax drawn there touches a fraction of a percent of households — not the 62% — which dissolves most of the fairness objection on the merits, even if the fear survives it in politics. Two honest caveats cut the other way, and we state them plainly: a top-focused wealth tax raises modest revenue against a $39T debt (it is an inequality-and-legitimacy tool, not a debt-eraser — which is why it must run alongside the growth lever above, not instead of it), and it faces the sharpest operational obstacles precisely because the ultra-rich are the most mobile (California has already seen departures) and their wealth the least liquid and hardest to value — the very unrealized-gains character that runs straight into the realization question Moore left open (§10). The logic of taxing where the money actually is holds; the honest qualifier is larger than usually admitted. Redistribution fixes who owns the country's wealth; it does not, by itself, fix what the country owes or what its money is worth. No tax on the rich, however just, erases a $39T debt or the price-level adjustment that most likely works it off — the debt is still most probably inflated away regardless, a Cantillon transfer to borrowers (§10) that no wealth tax can repeal. Progressive taxation is the right instrument for the distribution; the debt falls to growth (if we are lucky) or repression (if we are not), and the inflation it leaves largely untouched.

The tragedy is that the strongest answer is the throttled one. Redistributive taxation is capped and blocked exactly where it would matter — federally, by politics (the 2025 tax law moved the other way) and by a Court with four votes on record for a realization requirement (§10). Meanwhile the worst answer for the public, inflation, requires no vote, survives no legal challenge, and is already the base case. So the country drifts toward the regressive solution because it is the easiest, not because it is the best — which may be the single most important thing to take from this note.

The other honest lever: actively unlock growth. Redistribution fixes who pays; it does not grow the pie. But §9's lesson is that growth was the largest cylinder of the postwar escape and is the one most missing today — so the second honest response is to manufacture growth through structural reform rather than pray for it from AI (§6). The most underused tool here is vigorous antitrust — breaking up entrenched monopolies to restore the competition that turns productivity into lower prices and higher real wages. The precedent is concrete: the 1984 break-up of AT&T — the Bell System divested into seven regional "Baby Bells" plus a competitive long-distance business, ending a telephone monopoly that had stood since 1913 — unleashed exactly that. Long-distance prices fell sharply — roughly 80% over the following decade — as the divestiture's equal-access rules cemented the competitive footing MCI and Sprint had already won in court; equipment innovation accelerated once Western Electric's lock was broken; and the cheap-bandwidth, competitive landscape that resulted is widely credited with helping seed the internet and mobile booms that followed. That is the mechanism the debt actually needs: not a one-off inflation, but a structural lift to the growth denominator. And it does double duty against the §6 capture problem — competition is precisely what forces a productivity gain (from AI or anything else) to be shared with the public as lower prices rather than captured as monopoly rents. Today's concentration — in cloud, search, payments, and the AI stack itself — is the reason to expect capture; antitrust is the standing answer to it. (Steelman: break-ups are disruptive and can dismantle valuable R&D scale — Bell Labs is the cautionary case, and the academic literature on breakups-versus-innovation is genuinely mixed — and divestiture's precise causal debt to the internet is contested; but the consumer-price and new-entry effects are not.)

What this leaves for the Fed — and its limits. None of this is Chair Warsh's to fix: a distributional problem cannot be solved with an interest rate, and neither can a growth problem. The most useful thing a credible hawk can do is refuse to become the silent instrument of the regressive path:

• Don't let AI-disinflation become the alibi. If you ease, ease because employment or financial stability demands it — and say so — not because a convenient disinflation print lets the sovereign refinance on the quiet (§6).
• Guard the plumbing. The real risk in your term is the next SVB/gilt/repo break (§2), not 25bp in either direction.
• Make the arithmetic legible. A hawk's highest use under fiscal dominance is to refuse to monetize silently — forcing the fiscal authority, and the voters, to own the choice the Fed is otherwise left to make for them by stealth.

We expect little of the redistributive path to be taken and much of the inflationary one — which is exactly why our positioning (§14) is built for the world as it will most likely be, not as it should be.

§14 Tactical & Strategic Positioning

We invest for what will happen (§11 base case), while hoping for what should (§13). The regime read and the historical record point the same way.

Our principle: own the numerator of debasement, not the nominal claim on the debtor. The live Quad-3 stagflation read (§4) and the 1970s return table (§5) agree.

• Strategic (multi-year): overweight scarce and real assets — gold (whose reserve-share gain in §9 is the world's central banks making this trade first), commodities and energy, real estate with pricing power; Bitcoin as the fund's primary debasement hedge, sized via a bounded Kelly tilt against the Global-M2 liquidity lead and the regime composite (never an absolute μ); and equities with genuine pricing power and short payback. Underweight long nominal duration and the naïve 60/40 — the two things stagflation and repression punish together.
• Tactical (0–12m): respect Warsh's hawkish opening (§1) — a real hold can lift the front end and pressure risk before any pivot. Favor floating-rate and short duration and TIPS over long nominals; hold dry powder for the "new-chair test," historically the best entry into the easing that follows.
• Tail hedges (§10, §11): carry cheap convexity against the disorderly-adjustment branch and the rising political-instability premium — long-vol, gold, and rule-of-law-diversified exposure.
• Scenario mapping: real assets and BTC win in Base and Tail; the productivity Bull favors quality equity and hurts gold and long-vol — so size the hedge, do not bet the book on the thesis. Conviction and humility are not opposites here.

(The precise BTC sizing and liquidity overlay are to be run through btc-global-m2 / four-quadrant-nowcast / btc-kelly-sizer as a companion sizing note; this section sets the direction, not the position.)

§15 House View

The call: we are in early stagflation with a debt load that makes sustained hawkishness fiscally self-defeating. The most probable multi-year resolution is financial repression + stealth monetization, with captured AI-disinflation and a stablecoin front-end backfill as the enabling cover. Confidence: moderate — the regime read, the debt math, and the policy setup agree, but the Warsh reaction function and the true magnitude of AI productivity are genuinely uncertain, and we hold the base case with humility.

What would change our view (observable, with thresholds): real median weekly earnings turning positive year-over-year for two-plus quarters (→ Bull); net-interest/GDP stabilizing below CBO's ~4.1% path; the Fed sustaining a hawkish hold through an equity drawdown >10%; core PCE returning to 2% on a 3-month annualized basis without cuts.

The steelman against us: an independent, credible Warsh Fed genuinely fixes inflation; AI productivity is Goldman-scale and shared, lifting real wages and honestly growing the debt down; stablecoins deepen dollar demand and the rollover clears smoothly. In that world the debt is a manageable legacy, not an endgame, and repression is avoided. We hold our base case — but a thesis worth publishing is a thesis worth trying to break, and this is where it breaks.

Forecast versus prescription — hold the two apart. Our forecast is repression: it is the path of least resistance, and the most probable. Our reading of the merits points the other way: measured against upheaval, the inflation tax, and a widening gap, progressive wealth-transfer taxation is the least-bad resolution — the only one that spares the public the stealth tax, spares the country the instability tail, and actually narrows the divide (§13) — paired with the harder structural work of unlocking growth (antitrust chief among it) to lift the denominator the honest way rather than praying AI will. The distance between what we expect and what would be better is not a hedge; it is the point of the paper. We position for the world as it is likely to be (§14) while stating plainly that it does not have to be that way — and that the reason it probably will be is not economics but the fact that the worst answer is the one that needs no one's permission.

A final honesty, which this note refuses to dodge. Even the ideal — redistribute the wealth, unlock the growth, tell the truth about the tax — does not make the $39T or the inflation disappear. There is no clean exit: progressive taxation resolves the distribution; the debt is still most likely inflated away, a Cantillon transfer to borrowers that no wealth tax can repeal (§13). So the real choice is not between a solution and a problem — it is between resolutions that are visible and just and one that is hidden and regressive. Our fear is simply that a country offered that choice keeps choosing the hidden one, because it is the only one that asks for no vote — and that the quiet 3–4% erosion of the base case is the good outcome, with a fatter-than-comfortable tail (§11) where the brakes come off entirely.

§16 Disclosures

Regulatory status — publisher's exclusion. This report is published solely as general, impersonal information of regular circulation. It is not tailored to the investment objectives or circumstances of any individual and is not issued in connection with compensation from any client. TON618 Capital has no clients and distributes all research free of charge. On that basis the Fund publishes in reliance on the publisher's exclusion from the definition of "investment adviser" under the Investment Advisers Act of 1940 (§202(a)(11)(D); cf. Lowe v. SEC, 472 U.S. 181 (1985)). Nothing herein is personalized investment advice.

Conflicts, positions, compensation & registration. Conflicts of interest: TON618 Capital is a Bitcoin fund and may hold or transact in securities it discusses; material conflicts are disclosed where they exist. Ownership: the Fund holds no position in any single security named in this report; it maintains a strategic Bitcoin allocation consistent with the debasement thesis advanced herein — a material conflict the reader should weigh. Compensation: the Fund received no compensation from any party in connection with this report and charges nothing for it. Registration: TON618 Capital is not registered as an investment adviser or broker-dealer in any capacity.

CFA. This report was prepared to align with CFA Institute analytical standards (methodology only). CFA® and Chartered Financial Analyst® are registered trademarks owned by CFA Institute. That reference describes the analytical framework applied; it does not imply the report was prepared, reviewed, or authored by a CFA charterholder, and the report is not issued, reviewed, endorsed, certified, or approved by — nor affiliated with — CFA Institute.

Use of AI & feedback. Artificial intelligence is used in the creation of this research; all methodology and data integrity are reviewed and approved by the Fund's Chief Investment Officer, Keyth Beck. Direct feedback to him, keyth@ton618capital.com. Past performance is not indicative of future results. Version 1.4 · analyst: TON618 Macro Research.

Source & Verification Note

This note was independently fact-checked in three adversarial claim-verification passes (the third scoped to the reworked §8, §9, and §13) against primary sources — FRED, the U.S. Treasury Monthly Treasury Statement and Debt-to-the-Penny, CBO, GAO, IMF COFER, the ECB, the Moore v. United States (2024) opinion (parsed directly), the Hall & Sargent (2011) postwar-debt decomposition, Pew Research, and primary reporting. The great majority confirmed precisely against the primary record, including the highest-stakes items: the Warsh confirmation vote (54–45) and June-2026 FOMC outcome, the GENIUS Act signing date, the ECB's gold-overtakes-Treasuries finding, the Fed DFA wealth-concentration figures, the China/Japan Treasury-holdings trend, the Moore realization split (four justices, verified against the opinion text), the Massachusetts/Washington/California tax facts, Pew's 17% trust reading, and the postwar debt/GDP decomposition (growth ~39% / surpluses ~38% / repression ~23%).

Corrections applied against the primary record (reflected in the text): national-defense outlays at ~$917B and Medicare at ~$997B (Treasury FY2025 MTS, replacing an NDAA-authorization figure and an overstated Medicare figure); the debt's weighted-average maturity as ~6 years receding from its 2023 high (removing an unsupported "26-year high"); central-bank gold buying ~850t in 2025 (not a flat ~1,000t/yr); the Massachusetts surtax revenue as ~$5.7B since FY2023 (not "in two years"); the net-interest-to-revenue ~28% figure dated to 2055; and the USD reserve-share baseline as ~71% (2000). None alters the report's conclusions. The third (scoped) pass corrected four further items in the new sections: the Hall & Sargent postwar decomposition to its published ~40% growth / ~43% surpluses / ~20% repression (an earlier draft had spliced in a narrower footnoted estimate — the correction strengthens the "repression was the smallest cylinder" point); the US billionaire count to ~990 (2026 Forbes, a record); Tether's Treasury holdings to ~$141B (Q1-2026 attestation, 17th-largest holder); and the 1984 AT&T long-distance sequencing (competitive entry preceded the divestiture; the break-up's contribution was equal-access, now stated as such).

A few items remain appropriately hedged in-text rather than asserted as hard fact: Bessent's June-2025 stablecoin remark (substance corroborated, exact wording pending a primary-page confirm); the §8 bank-disintermediation context (the ~$190–408B Fed-staff reduced-lending estimate is primary-sourced; the "deposits in the trillions at risk" characterization is from reporting — the specific $6.6T figure was removed as untraceable to a Treasury primary document); the ~−2.8%/yr durable-goods deflation figure; and the 1970s asset-return table (endpoint-sensitive; approximate). The four-quadrant regime read is a proprietary model output, labeled as such, and is directionally corroborated by public data (core PCE at 3.4% YoY in May 2026, its highest since October 2023).

Full manifest (status: partial — two non-required feeds recorded as degraded, not substituted) and the ten research-sweep source files live with the Phase 0 snapshot at data/debt-endgame-2026/2026-06-30-185936/.