The dollar holds 57 per cent of the world's official reserves, and the figure is produced as proof that the world chose it. A choice requires that the alternative be available. Ask the states that reached for one, and the answer is a list rather than an argument.
Begin in 1951, not 1974. When Iran nationalised its oil, Britain blocked Tehran's sterling accounts, boycotted the crude worldwide and withdrew the technicians from Abadan. Two years later the government was gone, removed by two intelligence services, and forty per cent of the fields had been reassigned to American companies. The mechanism was complete and tested before the dollar inherited it, which is why the 1974 oil arrangement is the wrong place to start the story.
A reserve currency is a ledger somebody else keeps. What you own is an entry, and an entry can be amended.
The ledger as it stands. More than 100 billion dollars of Iran's reserves have been frozen for forty-seven years, about a quarter of its national income, of which 24 billion is now a term in a negotiation. Thirty billion of Libya's was seized weeks before the bombing began. Between 275 and 320 billion euros of Russia's has been immobilised since 2022. And since 3 January this year more than 13 billion dollars of Venezuelan oil has been sold under American administration, with roughly 300 million released to the government of the country that owns it.
The circuit is closed and it is not complicated. Exporters accumulate dollars because trade is invoiced in them. The dollars return as purchases of Treasury securities because there is nowhere else of that size to put them. The Treasury spends the proceeds, and a large part of what it spends them on is the force that keeps the alternatives unattractive. So when central banks buy 289 tonnes of gold in a quarter, sixty-two per cent more than the year before, do not call it diversification. It is an exit, dug quietly by people who watched what happened to those who announced one.
The dollar holds 57.13 per cent of the world's allocated official reserves, and the figure is produced as a verdict: the world was free to leave and stayed. Every argument on this subject turns on whether that sentence is true. It is worth establishing what happened to the states that tried.
Begin in 1951, not 1974. Iran nationalised the Anglo-Iranian Oil Company. Britain blocked Tehran's sterling accounts, organised a worldwide boycott of the crude and withdrew the technical staff from Abadan, and by 1953 the government had been removed by the CIA and MI6. Forty per cent of the fields went to American companies. Every element of the modern instrument is present: the currency account frozen, the export market closed, the political outcome procured. The mechanism was finished before the dollar inherited it.
A reserve currency is a ledger somebody else keeps. What you own is an entry, and an entry can be amended.
Iran again in 1979, and the freeze has never ended. More than 100 billion dollars remains blocked across at least seven jurisdictions: over 20 billion in China, 7 billion in India, around 6 billion each in Iraq and Qatar, some 2 billion in the United States itself, 1.6 billion in Luxembourg, 1.5 billion in Japan. That is about a quarter of a national income, held for forty-seven years, and 24 billion of it now sits inside a negotiation as a thing to be offered.
Iraq priced its oil-for-food sales in euros from 2000, at a cost of roughly ten cents a barrel, its government calling the dollar the currency of an enemy state. It was invaded in 2003 and the sales reverted to dollars. Libya had 30 billion frozen in 2011, the largest seizure of foreign funds in American history to that date, weeks before the bombing began; it held 143.8 tonnes of gold and sold around 29 of them in its final weeks, and it had proposed a gold-backed African currency.
Then 2022, which changed the scale rather than the principle. Between 275 and 320 billion euros of Russian reserves were immobilised, around 210 billion inside the European Union, and between 185 and 190 billion of it sitting at one Belgian depository. There is now a proposal to lend 140 billion euros against that money, drafted with care so that Russia formally keeps ownership and the word confiscation never appears. When a legal formula has to be invented to describe what is being done, the thing being done is new.
And in January this year, Venezuela. Since 3 January more than 13 billion dollars of its oil has been sold under American administration. About 300 million has been released to the government of the country whose oil it is. As Michael Hudson describes the arrangement, the proceeds settle “in U.S. controlled accounts” and are disbursed “at the discretion of the U.S. government”. That is not a sanction on a transaction. That is the administration of a country's export revenue by another state.
The version of this argument that goes too far should be given up, because it is what the other side answers instead of answering us. The strong petrodollar-war thesis says that currency choice causes invasion, and it over-predicts badly. Afghanistan does not fit it. Ukraine does not fit it. Washington's quarrel with Tehran predates any question of what oil is priced in by decades. Hold the narrow claim instead, which the record does carry: currency is not what starts the wars. It is what the settlements are denominated in, what makes a war affordable to the state that starts it, and what the defeated are made to accept.
The mechanism behind that affordability is the part the mainstream will not look at. August 1971 is read as the dollar's moment of weakness. It was the opposite. Leaving gold removed the obligation to settle, which was the only real constraint there had ever been on an American deficit. What replaced it is a Treasury-bill standard: dollars paid out abroad come back as purchases of American government debt, because for a central bank holding tens of billions there is no other market of the size. In Hudson's summary, foreign countries “really had no alternative but to have their own central banks themselves finance U.S. military spending.”
Follow the sign of that carefully, because it is where this reading parts company with the other critics of the arrangement. The conservative case says the deficit is the price America pays for issuing the world's money. The evidence says the deficit is the point. Hudson found at Chase in the 1960s that the American payments deficit was entirely military, the private sector being in balance. Real goods flow in and paper flows out, the paper is not going to be redeemed, and the larger the deficit gets the more of it foreign central banks are obliged to hold.
What that costs everyone else is the exemption. Any other country running deficits of this size raises interest rates, calls in the Fund, cuts wages and pensions, and calls it adjustment. One country does not have to, and the exemption is the whole of the privilege. Its creditors cannot force the adjustment either, because selling in size would crash the value of the reserves they are still holding. They are not investors. They are hostages with a balance sheet.
The bill this pays for is not abstract. The campaign against Iran cost about 72 billion dollars in two months, better than 1.2 billion a day, with 67 billion more requested in July. Net interest on the federal debt now runs near a trillion a year against 885 billion for declared defence. The states whose reserves finance that are, in a good number of cases, the states it is aimed at.
Which is why the gold is the story and not the renminbi. Central banks took a record 289 tonnes in the second quarter, sixty-two per cent above the year before, and gold has passed both the euro and US Treasuries in official holdings for the first time since 1996. Gold pays nothing. Its entire specification is that it has no counterparty: no ledger, no depository in a friendly jurisdiction, nobody to receive an instruction. Reserve managers are not buying a return. They are buying the absence of an address at which they can be served.
The honest form of the argument admits the timetable. Writing on the same sanctions, Michael Harrison put it exactly: “Each designation demonstrates Washington's reach in the short term. Each workaround reduces that reach at the margin over the long term.” Neither half of that sentence is triumphant. The reach is real now and the erosion is slow, and anyone promising the collapse of the dollar system by a date is selling something.
But do not mistake slowness for consent. Fifty-seven per cent is what the share reads when the exits have been demonstrated one country at a time for seventy years, when the largest holders are protectorates, and when the states that did move are named on a list of what happened to them. That is not a market verdict. It is an inventory taken under supervision, and the tonnage of metal now leaving the system is what the verdict actually looks like.
