The Last Buyer of America's Debt

• The question nobody asks out loud
• Name the voices and their stakes
• The map: a war, a strait, and a negotiation over money
• The second war: a cost with no terminal date
• The bloc: real, growing, and still without a currency
• The mechanism: the foreign bid
• The template: the seventies, not Weimar
• The case for the other side, argued at full strength
• The swing state: India
• The correction: the metal, not the coin
• The conclusions, with their failure conditions
• What the markets are pricing
• The close

Not a story about who holds the debt today — but about who will still be standing there to buy it when the rest of the world has stopped asking.

DeepDives · 2026 · An AI–human collaboration that weighs every source, mainstream or alternative, under strict rules of truth and logic.

§The question nobody asks out loud

Every borrower on the historical record eventually faces a question so uncomfortable that almost nobody asks it out loud: when the world stops wanting to buy your debt, who is the last buyer? The Spanish crown faced it. The Mughal empire faced it. And the United States in 2026 faces it now. The answer has always been the same. The last buyer is the state itself, through its central bank, purchasing what no one else will hold. The mechanism by which the world stops buying, and the speed at which it happens, is what this article is about — because we are living through it.

Consider the facts on the table, each verifiable, each running in the same direction. The war in the Gulf grinds into its seventh month with the Strait of Hormuz still not fully open. The war in Ukraine enters its fifth year with no durable ceasefire. The bloc of non-aligned powers meets in September with de-dollarization on its agenda. Foreign holdings of American Treasuries are falling. The dollar’s share of global reserves is at a thirty-one-year low. And the world’s central banks are buying gold at nearly double the pre-2022 rate, while the American Treasury, for its part, has started buying back its own long-term bonds.

Put those facts together and you get the question of the decade, stated precisely: does a multipolar world make fiscal dominance worse? The thesis of this article, argued with every voice on every side weighed equally, is yes — but not for the reasons most people assume, and not without powerful counterforces.

§Name the voices and their stakes

This is a battlefield of interests, so let the voices and their stakes be named before the evidence is examined, because a claim from an interested party is not the same thing as a claim from a neutral one — and on this topic there are no neutral parties.

The central banks buying the gold are the most conservative institutions on earth, and their incentive is not price appreciation, it is survival. They hold the reserves that must still be there when the music stops, and the asset they are choosing, gold, is the one with no counterparty and no freeze button — a property that matters to institutions whose dollars can be frozen by a political decision in Washington, as three hundred billion dollars of Russian reserves were in 2022.

The gold industry has an incentive too. The World Gold Council, whose surveys of central-bank buying intentions are quoted in nearly every gold article, is the marketing body of the gold mining industry, not a disinterested pollster. Its numbers should be checked against the actual tonnage data, which is collected independently — and which happens to confirm the trend. The mainstream economists who insist the dollar cannot be dethroned, the most prominent of whom is Barry Eichengreen, author of the canonical book on the dollar’s resilience, are defending a professional consensus and a world order they have spent their careers explaining. They are not wrong about the mechanics of inertia; they are wrong about the direction.

The bitcoin circuit reads every one of these data points as confirmation that the dollar system is ending — and the people making that argument are, to a person, long the asset they say will win. The policymakers in Washington have an incentive to describe the erosion as manageable, because the alternative is admitting that the era of effortless borrowing is over. And this series holds its own stake, declared before and declared again: it has argued that the debt ends in inflation and repression, and it holds the assets that survive that ending. Everyone here has skin in the game. The evidence is what it is, and the evidence is what follows.

§The map: a war, a strait, and a negotiation over money

First the map, stated with precision about what we actually know.

On the twenty-eighth of February 2026, the United States and Israel launched a war against Iran and its regional allies. This is established fact, confirmed across multiple sources, and it began with the most ambitious strike on Iran in decades: the opening salvos killed several senior Iranian officials, including, as Iranian state media confirmed, the Supreme Leader. His son was named his successor in March. That fact matters, because it cuts against a simple story of American weakness: the coalition did what decapitation strikes are supposed to do, and removed the head of the regime.

And yet the war did not end. The regime did not collapse. The strikes were aimed at regime change, and regime change did not follow. Instead the war ground on, and one of its most consequential effects has fallen on the Strait of Hormuz, through which roughly a fifth of the world’s oil passes. The strait has been effectively closed since the war began. It was briefly reopened in June as part of the ceasefire, and then re-closed on the twentieth of June. What has been happening since is negotiation. In early August, reports said talks to reopen the strait were making progress, with Qatar circulating draft language and Washington expressing cautious optimism. On the twenty-fifth of August, Iran conveyed its conditions to the United States through Pakistan, the channel the ceasefire agreement designated. And in the week of this recording, Iran and Oman agreed on a temporary route for some shipping. As of the day these words are written, the strait is not fully open. That is the honest state of things.

Notice something that would have been unthinkable a decade ago: both sides are now negotiating over money. Iran’s president has stated three conditions for ending the conflict — firm guarantees against future aggression, recognition of Iran’s legitimate rights, and reparations. And Washington, in turn, has said it would require compensation from Iran as a condition for talks. Both sides are demanding payment from the other. In the classical grammar of war, that is not the language of victory and defeat. It is the language of two exhausted parties haggling over the bill.

§The second war: a cost with no terminal date

Now Ukraine, and again, certainty must be handled carefully. The war entered its fifth year in February 2026. Trilateral talks among Russia, Ukraine, and the United States began in Abu Dhabi in January and continued in Geneva in February. A United States-brokered three-day truce in May did not hold. There is no durable ceasefire, and the most widely held view among the analysts consulted for this series is that a frozen conflict — an unresolved line, a war that stops being fought but not being paid for — is more likely than a negotiated settlement.

Whether the United States is losing that war is a framing, not a fact. What the facts support is narrower and more consequential: the war has not ended on terms the United States sought, its mediation has not produced a settlement, and the fiscal cost to Washington continues with no visible terminal date. That is the part that matters for the analysis. A superpower fighting two wars it cannot finish, in the same decade its debt crosses forty trillion dollars, is the exact configuration the historical record punishes. And the historical template for it is not Weimar, as the alarmists say. It is the nineteen seventies — a different and in some ways more instructive precedent, to which this article will return.

§The bloc: real, growing, and still without a currency

Then there is the bloc. The BRICS group — eleven full members and ten partner states, with more than thirty countries waiting for a seat — accounts for roughly twenty-seven percent of global output, about twenty-four percent of exports, and nearly half of humanity, according to widely cited estimates. Egypt, Ethiopia, Iran, Saudi Arabia, and the United Arab Emirates joined in the 2024 expansion; Indonesia in 2025; Vietnam and others have become partners; Turkey’s membership is under active consideration. In September 2026, the bloc holds its summit in New Delhi, chaired by India, with an expansion vote on the agenda and a cross-border payment platform called BRICS Pay planned — described by observers as the most concrete de-dollarization step the bloc has taken.

Note the word planned. It has not launched. And note the incentives of the people who describe it as a threat: the Russian and Chinese state media apparatus promotes the bloc’s de-dollarization narrative because it serves their states’ geopolitical goals, while the Western commentators who dismiss BRICS as a paper tiger are defending the order their own states run. The truth about the bloc, as with most things, is in the middle. It is real, it is growing, and it has not produced a single working alternative to the dollar system. That combination matters, because the foreign bid for American debt is thinning for reasons that have less to do with the bloc’s ambitions than with the behavior of the dollar’s own issuer.

§The mechanism: the foreign bid

The mechanism is the foreign bid. For seventy-five years, the United States has financed its deficits with a captive audience: the world’s central banks needed dollars for trade and for reserves, and they recycled those dollars into Treasuries. That is the exorbitant privilege — not merely that the world lends cheaply, but that it has few alternatives. The valve is now closing, measurably.

In June 2026, foreign holdings of United States Treasuries fell, led by Japan, the largest holder, and by China, the third-largest, whose holdings had already reached an eighteen-year low. Foreign investors absorbed a valuation loss of about one hundred forty-two billion dollars on long-term Treasuries in March alone. The dollar’s share of global reserves has fallen to a thirty-one-year low.

And here is the nuance the panicky headlines miss — the one that separates honest analysis from doom-porn: the central banks have not been dumping dollars. Their total reserves have grown while their dollar assets stayed roughly flat for a decade. The diversification has gone to other currencies, and above all to gold. Central banks bought about two hundred forty-four tonnes of gold in the first quarter of 2026, and the year is on track for roughly seven hundred fifty-five tonnes — nearly double the pre-2022 average of four to five hundred. In the World Gold Council’s latest survey, eighty-nine percent of central banks said global gold reserves will keep rising, and a record forty-five percent said their own institution plans to buy. Take those survey numbers with the industry’s marketing incentive in mind, then check them against the independent tonnage data, which confirms the direction: the buying is real, it is broad, and it is the most conservative money on earth hedging against the dollar.

The accelerant is the weaponization of the dollar. In 2022, the United States and its allies froze roughly three hundred billion dollars of Russian central bank assets. In 2026, the release of Iran’s frozen assets became a bargaining chip in the ceasefire talks. The message to every non-aligned reserve manager is unambiguous: dollar assets are only as safe as your relationship with Washington. And gold is the asset with no counterparty. It cannot be frozen, cannot be sanctioned, cannot be turned off. The record gold buying is not a bet on gold’s price. It is a vote on the dollar’s neutrality — and the vote is no.

§The template: the seventies, not Weimar

The historical template this series is most confident in is the nineteen seventies, not Weimar. Consider the structure. A superpower fighting a grinding, unwinnable war in Southeast Asia. A fixed gold price of thirty-five dollars an ounce that allies were quietly converting out of. Inflation and overstretch compounding. And then the closing of the gold window in 1971, the end of Bretton Woods, and gold’s rise from thirty-five dollars to eight hundred fifty by 1980 — a move of more than twentyfold.

The lesson of the seventies: when a superpower fights wars it cannot finish while printing money it cannot back, the world stops being a willing buyer, and the price is paid in the value of the currency and in the price of the things that are nobody’s liability. The 2020s resemble that decade in structure, with worse demographics and far more debt. And the current numbers rhyme: gold around forty-five hundred dollars, with serious sell-side analysts projecting six thousand by year-end — a forecast from J.P. Morgan that carries the incentive of a bank that makes money when markets move, but a serious one; central banks buying at nearly double the pre-2022 rate; the dollar’s reserve share at a thirty-one-year low; foreign official holdings of Treasuries declining. Each of these is a data point in the same story.

§The case for the other side, argued at full strength

Now the case for the other side, argued as strongly as it can be argued — because it is substantial, and because the mainstream economists who make it are not fools. They are people whose error is direction, not intelligence.

First, there is no alternative. The euro is a currency without a state. The yen belongs to a country whose central bank already owns more than half its government’s bonds. The yuan is capital-controlled and politically managed. And gold cannot be the settlement layer of a global trading system: it is a store of value, not a payments rail, and the world still needs a medium for trade that does not require shipping metal across oceans. BRICS has no currency. It has a payment platform that has not launched, and a development bank whose recent history is a list of failures: Colombia’s bid to join died in its own congress without a single debate, and Iran’s claim of membership was not confirmed by the bank itself.

Second, the bloc does not actually want the dollar to die. Russia, China, and Iran hold trillions of dollars in reserves and dollar assets between them; a collapsing dollar would destroy their own wealth. They want independence at the margin and sanctions-proofing — not the immolation of the reserve asset. Third, fiscal dominance is primarily domestic: roughly three-quarters of United States debt is held at home, by institutions with no alternative. The foreign seller raises the price; it does not close the market. Fourth, the erosion is slow by design. A thirty-one-year low is a thirty-one-year story, and it is reversible: peace in Ukraine, a reopened Hormuz, a dollar-supporting policy mix, an India that doubles down on Treasuries — any of these could slow or stall the trend.

Eichengreen’s argument, that the dollar’s dominance is sticky because networks of trade and finance resist change, is correct about the stickiness. What it underestimates is the direction of the pressure, and the fact that the issuer of the reserve currency is actively supplying the reasons to diversify. A forecast that does not admit its own failure conditions is not analysis; it is a mood.

§The swing state: India

India deserves its own attention, because India is the swing state of the multipolar world — and India refuses to choose, and refusal is itself a strategy.

India chairs BRICS in 2026, hosting the September summit, and it is the world’s second-largest importer of Russian oil, buying it at a discount. It is also a Quad member, a major buyer of American defense hardware, and among the larger foreign holders of United States Treasuries — while its central bank and its households accumulate gold in enormous quantities. India wants a multipolar world in which India is one of the poles, and it wants a dollar strong enough to hold the value of its own reserves. It is the marginal buyer: the country that could slow the erosion of the dollar bid, or accelerate it.

Here the facts are genuinely mixed, and they should be presented as such. Washington has already imposed steep tariffs on India, including a penalty tied specifically to its Russian energy trade, and the Senate has advanced legislation authorizing tariffs of up to five hundred percent on buyers of Russian oil, with India and China explicitly in the crosshairs. A reasonable reading: the United States is taxing its most important remaining customer. A fair counter-reading: India has so far absorbed the pressure with remarkable calm, its diplomats describing a policy of quiet maturity, and there is no sign New Delhi is about to abandon the dollar system it still needs. Both readings are defensible. The truth is probably in the middle — and the middle is where the risk lives, because the country that holds the fate of the foreign bid in its hands is being asked, in public, to choose between the bloc it chairs and the dollar it still needs. Its answer so far has been to refuse the question.

§The correction: the metal, not the coin

One correction must be stated plainly, because it matters most for how any reader should think about their own holdings, and it is the correction the alarmist side of this story keeps refusing to make: central banks are buying gold. They are not buying bitcoin.

The institutions buying the record tonnage of gold — the same institutions that told the World Gold Council they plan to buy more — have not adopted bitcoin as a reserve asset. Bitcoin’s buyers are a different crowd: funds, corporations, and households, plus a small number of governments — El Salvador, Bhutan, and states holding bitcoin obtained through seizures. Some central banks are responding to the crypto world, but their response is to build their own digital currencies, not to buy bitcoin.

This matters for two reasons. It matters for accuracy, which is the whole point of honest analysis. And it matters for reasoning: the gold signal is a central-bank signal — the most conservative money on earth hedging against the dollar; the bitcoin signal is a market signal — a bet by risk-takers on the same underlying thesis. They rhyme. They are not the same song. The people telling you that central banks are buying bitcoin are telling you what you want to hear, and this analysis does not do that, on any side. Anyone building their own position should know which signal they are following — and should know that the institutions whose job is to be paranoid are buying the metal, not the coin.

§The conclusions, with their failure conditions

The conclusions follow, drawn with the uncertainty they deserve and with every side’s incentives named.

The balance of evidence supports this: multipolarity does not create fiscal dominance — the debt and the wars did that — but it removes the safety valve that made fiscal dominance tolerable. The captive foreign bid is thinning, not collapsing. The diversification is into gold and other currencies, not into a new reserve currency. And the pace is measured in decades, not quarters — which is exactly why the mainstream consensus can keep insisting that nothing is happening while the most conservative institutions on earth quietly move a record share of their reserves into the one asset with no counterparty.

If the erosion continues — if foreign official demand keeps fading while the Treasury must absorb more duration — then the pressure on yields will rise, the buyback program will likely have to grow, and the Federal Reserve, whatever its current doctrine of shrinkage, will eventually face a choice between tightening into a fragile market and returning to the market as a buyer. That is the endgame of fiscal dominance in a multipolar world, and it is the answer to the question this article opened with: when the world stops buying, the last buyer is always the state itself, through its central bank — and the only question is the price at which it buys, and the currency in which it pays.

That is a projection, not a prediction. It is what the evidence points to, and it could be wrong. It would be wrong if the wars end quickly and the credibility premium returns. It would be wrong if India and the Gulf states step up as buyers. It would be wrong if the bloc fractures — which its own institutional failures suggest it might. The honest analyst states what it sees, and states what would change its mind. That is the deal.

§What the markets are pricing

What the markets are pricing is more certain than any forecast. Gold around forty-five hundred dollars, with J.P. Morgan’s analysts projecting six thousand by year-end — a forecast, but a serious one. Central banks buying gold at nearly double the pre-2022 rate. Bitcoin in the high seventy-thousands of dollars. The thirty-one-year low in the dollar’s reserve share. The June decline in foreign Treasury holdings. The Treasury buying back its own long bonds. Each of these is a data point in the same story — and the story is not that the dollar is dying. It is that the world is slowly, deliberately repricing the promise of the United States: not rejecting it, discounting it.

And in that repricing, the assets that need no promise — that have no counterparty and no freeze button — are the ones the cautious money is buying. The cautious money is buying gold. The adventurous money is buying bitcoin. Both are answering the same question in different risk brackets, and both are telling the same thing: the era when the world would fund American debt without asking questions is ending — slowly, unevenly, and not without resistance. The resistance is real, the alternatives are weak, and the timeline is long. But the direction is the direction, and the mechanism is the mechanism. The last buyer, when the foreign bid is gone, will be the one institution that cannot refuse: the central bank, printing the money to buy the debt that no one else will hold, and calling it something else. The names change. The mechanism does not.

§The close

Strip it to its bones and the whole argument fits in a few sentences. A state that borrows in a currency the world must hold can survive its creditors’ doubts for a long time — decades, even — but the doubts accumulate in the price of everything it issues. Wars that cannot be finished, an interest bill that cannot be outgrown, and a foreign bid that is thinning rather than growing are the three pressures that turn an inconvenience into an endgame. The multipolar world did not create the American fiscal condition; it removed the cushion that made the condition bearable. And when the last willing foreign buyer has gone home, the buyer of last resort is the state itself, and the only open question is the price of the money it prints and the name it gives the operation.

Which is why the position that survives every version of this ending is the one this series has recommended from the beginning, and the one the evidence above supports: own the assets that cannot be printed and cannot be frozen; keep your balance sheet boring; keep your skills high; and read the numbers from every side, with the agenda of every source named — because the gold is bought by institutions whose job is paranoia, the coin is bought by risk-takers who rhyme with them but are not them, and the difference between the two signals is exactly the difference between surviving and speculating. The era of effortless borrowing is ending slowly, and slowly is the hardest pace to prepare for, because nothing ever seems to happen until the day it has all happened. The direction is the direction. The mechanism is the mechanism. The names change. The last buyer does not.

Sources: this article names its sources inline — the reporting on the Gulf war and the Strait of Hormuz negotiations through late August 2026; the status of the Ukraine war and the trilateral talks in Abu Dhabi and Geneva; the BRICS membership and the agenda for the September 2026 New Delhi summit; Treasury foreign-holdings data for June 2026, the valuation losses of March 2026, and the dollar’s share of global reserves; central-bank gold tonnage data weighed against the World Gold Council’s survey, its marketing incentive named; the J.P. Morgan gold forecast, its incentive named; the scholarship of Barry Eichengreen on the dollar’s resilience; and the fact base of the DeepDives fiscal series. Every source is weighed with the same skepticism regardless of politics or business model. The audio version of this article, read by DeepDives, is available on Wavlake.