Keep the language, change the collateral.
The Calm Regime described a mechanism that has been built and benchmarked. Why We Must Exist described the market it could become. This paper is a hypothesis: what that market could mean, at scale, for the dollar.
Petrodollars, Latin American loans, and a weekend in August 1982.
On 15 August 1971 the United States closed the gold window, and the dollar’s claim on the world’s reserves lost its stated basis. The monetary basis of the postwar dollar had disappeared; within three years its global role had been refinanced instead. OPEC ran some $450 billion of surpluses between 1974 and 1981, deposited them in London and New York, and the banks lent them onward to Mexico, Brazil, and Argentina as syndicated, floating-rate dollar loans. Lending to emerging sovereigns became the principal outlet for the offshore dollar system’s swelling deposits, and the recycling renewed the dollar’s reign without a single treaty being amended.
It ended on a weekend in August 1982, when Mexico’s finance minister arrived in Washington to say the money was gone. The design had two flaws: the loans were unsecured, and the currency mismatch sat on the borrowers. When Volcker raised the price of the dollar, the mismatch detonated, and the recycling that had renewed the offshore dollar system cost Latin America a decade. The instinct (back the dollar with emerging-market yield) survived. The instrument did not.
The world is leaving the dollar’s asset while keeping its unit.
Fifty years later the pressure has returned, because this time the doubt attaches to the fiscal foundation of the risk-free asset itself. On 19 August 2026 the sovereign behind it crossed $40 trillion of debt, five months after crossing $39 trillion; annual interest passed $1 trillion in 2025 and is projected to reach $2.1 trillion within a decade. In the tradition I was raised in, forty is the age of hidayat, the age at which guidance arrives. The world’s central banks have stopped waiting for it, in a manner the de-dollarization debate has misread. In 2025 gold passed US Treasuries as a share of official reserves (a price effect in large part, but a first in the modern data), while the dollar’s share of allocated currency reserves held at 57 percent, where it has sat for years. Taken together, those two facts suggest that reserve managers are distinguishing between the dollar as an asset and the dollar as a unit.
The durable thing is the denominator: the language of the world’s obligations, in which its trade is invoiced, its debts written, and its collateral margined. None of those uses requires holding the Treasury. They require only that everyone else keep using the dollar, which is how a language survives: because its speakers keep using it, not because its country of origin stays strong.
The doubt is visible in the curve. On 15 September 2026 the US ten-year closed above 5 percent, its highest since July 2007, while the three-month bill paid 4.05, and Japan, the largest foreign holder, now earns more on its own bonds than on a hedged Treasury. The long end is charging for fiscal risk; the front end still prices confidence in the central bank over a quarter. What the system continues to price with relatively little hesitation is the short promise: ninety days rather than ten years.
The denominator needs an income. Gourinchas and Rey showed that America’s privilege was always funded by a spread: the United States earns more on its risky foreign assets than it pays on the safe liabilities the world holds. As inflationary risk erodes the real return on the safe liability, the search for another income line becomes more urgent, which is where the system stood in 1974. Only the scale has changed: the first recycling moved $450 billion across eight years, about 8 percent of one year’s world output at the time; dollar credit to non-banks outside the United States alone stood at $14.3 trillion at the end of 2025, 12 percent of world output and growing at its fastest rate since 2014.
Economists call this bind the Triffin dilemma: the world’s demand for a safe asset grows with the world, and the fiscal capacity of one issuer does not. The literature allows three ways out — rival currencies, a supranational asset, or a crisis. There is a fourth, unnamed until now: keep the unit, change the collateral.
Dollar claims on local sovereign collateral, with the currency tail kept at home.
The yield is still in the emerging markets, where it was in 1974, but it now sits in a very different obligation. Then, it was reached through a bank’s claim on a sovereign’s dollars; today it sits in the sovereigns’ own bills, at the top of each domestic capital structure. The ninety-day dollar bill pays 4.05 percent; the same ninety days pays 11.5 in Karachi, 19 in Cairo, 26.5 in Lagos, and 37 in Ankara. Neither side is free of risk; what differs is the nature of that risk and the compensation offered for bearing it. The holder of the dollar bill earns less than a point in real terms. Across the emerging markets, a dollar holder of the same ninety days can earn a premium over the dollar bill that survives the hedge, if the currency tail can be retained on a local-currency balance sheet.
What must change is the instrument. The 1974 claim was a dollar loan, unsecured, with the currency risk pushed onto the borrower. The claim now needed is the reverse: a dollar-denominated claim secured on the sovereign’s own local-currency bill, with the currency risk kept by a local balance sheet that is paid to hold it. That instrument, together with the mechanism that retains the tail locally, is what Tranchco is building. Capital arrives in local currency and buys the sovereign’s own bill, the claim a government is least likely to default on because it can be paid in the currency it prints. The dominant risk is the currency, and that is the risk the structure bounds: the tail is retained by a domestic institution that elects to hold it, for a premium, in a loss order agreed before entry. The remaining risks, sovereign, convertibility, liquidity and execution, remain, and must be governed separately through structure, counterparties and operating controls. The dollar appears exactly once: as the denominator of the senior claim. Dollar instruments have drawn their safety from the top of the hierarchy of money, from the Fed and its swap lines; this one enters sideways, its collateral in another monetary system even as its settlement remains within this one.
The structure buys the local-currency treasury bill, not the US Treasury; it makes no dollar loan and leaves no currency mismatch on the borrower. The cost of the conventional alternative, a continuously hedged position, can be measured against it. In Turkey from January 2022 to April 2026, through the 2023 devaluation, a continuously hedged position returned minus 10.4 percent a year over the dollar risk-free rate, while governed exposure returned plus 6.0. The 16-point annual gap is the observed performance difference between continuous forward protection and governed exposure to a currency that moves in regimes. This extends the distinction already visible in reserve portfolios, rotation out of the asset while retaining the unit, into yield-bearing form: de-dollarization of the asset with retention of the language.
| First recycling, 1974–82 | Second recycling | |
|---|---|---|
| The claim | Unsecured; a bank’s signature | Secured on local sovereign bills, held onshore, bankruptcy-remote |
| The currency | Dollar loans; mismatch on the borrower | Local currency; the dollar as denominator only |
| The tail | Unpriced, left on the sovereign borrower | Priced, held domestically by election, loss order agreed at entry |
| The control | Money-center bank underwriting; no systematic control of sovereign currency mismatch | Rules-based, local balance-sheet risk sharing; parks at pre-committed thresholds; independent attestation |
| What it produces | Dollar claims on emerging-market borrowers; systemic currency mismatch | Dollar claims on local sovereign collateral; the currency tail retained locally |
The dollar can keep its throne. What stands beneath it changes.
Whether this renews dollar hegemony or replaces it with something else is the right question, and the answer runs on two layers. At the layer of the unit, it is renewal: secured spread restores a return worth staying for. The pound sterling is the serious objection: a denominator does eventually follow its asset down. It does so at a lag measured in generations (the pound’s ran half a century), and what the system builds during the lag decides what stands at the end of it. The test is public: the day dollar invoicing follows the reserve rotation downward, the lag is closing and this thesis is wrong. That series has not moved.
At the layer of balance sheets, it is the opposite of the 1974 arrangement: emerging markets acquire external funding in their own currency, the exit from what economists call original sin, and the tail stays home by consent rather than leaving by default. The deeper change is to what stands beneath the denominator. The first Eurodollar rested every claim, in the end, on one sovereign’s paper. The secured version spreads the yield across many sovereigns’ promises while keeping the one unit, and a denominator anchored to a portfolio of promises could be sturdier than one anchored to a single issuer’s macro scaffolding. The intuition has an old pedigree. Keynes arrived at Bretton Woods seeking a monetary architecture less dependent on the balance sheet of a single sovereign; White’s dollar won. Eighty years later, markets may be reopening part of that question in the currency that defeated him.
The first recycling took the world’s offshore dollars and placed claims against emerging-market borrowers. The second places claims against emerging-market sovereign collateral, in Karachi, Istanbul, Cairo, and Lagos, while keeping the dollar as denominator, with the structure built and benchmarked before the capital arrives. That solves the currency mismatch differently, and it creates a new question. If a secured dollar claim no longer requires its collateral to come from the sovereign that issues the denominator, what qualifies as collateral? The first recycling asked emerging markets to borrow the world’s currency. The second asks whether the world’s currency can borrow their collateral.