By: Tim Staheli
There is an old story, usually pinned on Bertrand Russell, about a lecture on cosmology. An elderly woman objects that, in fact, the world rests on the back of a giant turtle. And what, the lecturer asks, does the turtle stand on? “You’re very clever, young man,” she replies, “but it’s turtles all the way down.”
The stablecoin market has its own version of this problem, and it has now been written into law on both sides of the Atlantic. The global stablecoin market stood at around $316 billion at the end of June 2026, of which 97 per cent is pegged to the US dollar. Peel back any layer of this supposedly novel financial architecture and the same thing stares back. A stablecoin is a claim on a dollar, which is a claim on a Treasury bill, which is a claim on the US government. Dollars all the way down.
A statute with a sales pitch
The GENIUS Act, signed into law on 18 July 2025 after passing the Senate 68–30, is unusual among financial regulations in that its authors were candid about its purpose. Senator Kirsten Gillibrand, one of its architects, said the framework was “absolutely critical to maintaining the U.S. dollar’s dominance”, and President Trump, signing it, called it a “giant step” towards cementing American dominance of global finance. Most financial legislation buries its politics in the recitals. This one printed them on the tin.
The mechanics deserve attention as well. Permitted issuers must hold identifiable reserves on at least a one-to-one basis, with Treasury holdings limited to securities maturing within 93 days, so every dollar-pegged token minted in Lagos, São Paulo, or Karachi puts its issuer in the market for short-dated American government debt. Citi’s forecasters expect the market to reach $1.9 trillion by 2030, or $4 trillion if the bulls have it, and estimate that growth on that scale would generate close to $1 trillion in fresh demand for US government debt. And since the Act forbids issuers from paying interest to holders, the arrangement has an elegant one-sidedness: the customer holds a dollar that yields nothing, while the issuer collects the Treasury coupon on the reserves behind it. Tether, which reported net profits above $10 billion in 2025 and holds more US government debt than South Korea, has demonstrated just how lucrative minting other people’s dollars can be. Kings used to take this cut at the mint, and it says something about the durability of good rackets that the practice has survived into the age of the blockchain, with its margins improved.
America has run this experiment before
The dollar has a history of breeding beyond its keeper’s sight. The eurodollar market of the 1960s grew up in London precisely because dollars held offshore escaped the Federal Reserve’s reserve requirements, and American officials spent years fretting about a shadow dollar system they could neither see nor supervise. The GENIUS Act inverts that history. This time the offshore dollar comes with a licence, a monthly reserve disclosure and a standing order for T-bills; the escape has been organised from inside the building.
The deeper precedent is older and less flattering. The last time America allowed private firms to issue circulating money was the free banking era of 1837 to 1863, when state-chartered banks printed their own notes and a Michigan dollar might fetch ninety cents in a New York shop, or rather less, depending on who had issued it and whether anyone believed them. The rogues of the era were the ‘wildcat’ banks, named – so the story goes – for setting up in country so remote that only wildcats lived there. The scheme was simple. A banknote was a promise to hand over gold to whoever presented it at the counter, and a wildcat printed the promise generously, then put the counter somewhere nobody could reasonably get to.
Merchants defended themselves with publications such as Thompson’s Bank Note Reporter, which existed solely to list the value of paper from thousands of different banks that week – a thriving genre until a uniform national currency killed it. Kenneth Rogoff and Max Harris examined the parallel for the American Economic Association this year and concluded that the GENIUS regime resembles that chaotic period more closely than the national banking system which ended it – noting that nothing in the Act guarantees holders a dollar back, on demand, for every token. The wildcat put its counter beyond the swamp; its successors can put theirs in the small print.
The Bank for International Settlements had already reached for the same history. Stablecoin holdings, its Annual Economic Report said, are “tagged with the name of the issuer” just as the old banknotes were, and the market prices the name along with the token: a Tether dollar and a Circle dollar are both dollars in roughly the way a Michigan note and a New York note were both dollars – identical in ambition, distinguishable in a crisis. The BIS’s fuller verdict was that stablecoins fail its triple test of singleness, elasticity, and integrity – which is the central-banking way of saying they are claims on money rather than money itself. One more turtle in the stack.
Europe legislates for a currency it barely has
The EU saw all of this coming and enshrined its defence in law. Under MiCA, a token pegged to anything other than the euro hits a ceiling once it is used for payments at scale: one million transactions or €200 million in daily payment value, a cap Brussels makes no secret is there to stop the eurozone dollarising. From there the two rulebooks part company on almost everything. MiCA tells issuers to keep 30 per cent of their reserves – 60 per cent for the biggest – in bank deposits; GENIUS points issuers at short-dated Treasuries and asks for no deposits at all. An issuer seeking to serve both markets has to maintain two balance sheets that conflict with each other. And there is history behind the European choice, none of it comfortable. In March 2023, USDC slipped its dollar peg because more than $3 billion of its reserves sat in uninsured deposits at Silicon Valley Bank. The reserve mix Europe now requires by law is the one that broke the peg last time.
What Europe is defending remains, for now, mostly hypothetical. The ECB’s own Macroprudential Bulletin puts euro-denominated stablecoins at roughly €450 million as of January 2026 – a ninefold rise in two years, and still a rounding error against the dollar’s $300 billion. The cavalry is assembling: Qivalis, a euro stablecoin venture backed by 37 European banks across 15 countries, including BNP Paribas, is pursuing a Dutch e-money licence and targeting launch in the second half of 2026, while the digital euro, on the ECB’s own most optimistic reading, will not be issued before 2029 – and only if a contested law clears the Parliament. ECB board member Piero Cipollone said in an interview with El País in January that geopolitical tensions and “the weaponisation of every conceivable tool” reinforce the case for a payment system built on European technology and under European control. He may be right. But a sovereignty shield that arrives in 2029 protects a battlefield that will have been mapped, fenced and paved by then, largely in dollars. And Frankfurt itself sounds unconvinced by its private-sector deputies: Christine Lagarde argued in May that the case for promoting euro stablecoins is “far weaker than it appears”, and that a currency earns international standing through deep capital markets rather than tokens. Europe, in other words, is fielding a champion its own central bank president does not believe in.
Meanwhile, the fortifications have begun functioning as designed, with the side effects fortifications tend to have. MiCA’s transitional period expired on 1 July 2026, and the immediate consequence has been subtraction rather than substitution. Tether declined to seek European authorisation, objecting to the deposit rules, and USDT – the world’s most traded stablecoin – is now coming off regulated European venues one by one. Revolut told its users on 3 July that purchases would stop within days and that any USDT still sitting in an account on 31 August will be converted to ordinary money at whatever the rate happens to be that morning. The United Kingdom, watching from mid-Channel, is assembling a third regime of its own, with the Bank of England reconsidering its holding limits after industry criticism that its proposals would leave the UK diverging from everyone. Three reserve philosophies, three supervisory perimeters, three definitions of the same instrument.
The treasurer inherits the mess
For corporates, the appeal was always simplicity. Paying a supplier on another continent still means routing money through a chain of banks, each one adding time and taking a margin, and it means keeping cash parked in foreign accounts in advance so that payments don’t stall – money that sits idle by design, doing nothing except waiting. A stablecoin promised to collapse all of that into a single transfer that clears in seconds and observes no weekends. The demand is beyond argument: $33 trillion moved through stablecoins in 2025. And in frontier markets, where international banks have spent a decade cutting ties, the appeal is close to existential. The Reserve Bank of India observed in its June Financial Stability Report that stablecoin demand has been strongest in economies with weaker institutions and limited access to dollars, which is a diplomatic way of saying the token goes where the bank will not.
New rails, old borders
What is emerging, though, is a landscape with the familiar contours of the system it was meant to retire. More than 60 per cent of global stablecoin liquidity sits in just five trading pairs; venture beyond them and the costs climb, exactly as they always did in the currencies of smaller countries. A treasurer shopping for a token today finds the American one barred from scaling as a payment instrument inside the EU, the European one backed by a fraction of the liquidity, and a British one whose rulebook is still being written – with the compliance map redrawing itself at every border in between. Even the technology has borders now: a company holding its tokens on Ethereum can find its counterparty settling on a chain it has never used, and moving value between the two is its own small adventure in fees and risk.
The old headache – systems that cannot talk to one another – has simply moved house, from correspondent banking to the blockchain. The treasurer who once kept idle cash in bank accounts around the world may come to keep a wall of wallets instead: tokens that answer to different laws and settle on different chains, while resting, at the bottom, on the same few governments’ promises. The BIS, in this year’s report, added the unkindest projection of all: even if the market grows to between $1 trillion and $3 trillion, the net effect on economic output would be modest, and in some scenarios slightly negative. The plumbing gets faster; the water is the same.
None of which means the dollar wins by default forever. Its share of global FX reserves has drifted from 71 per cent in 2001 to around 57 per cent at the end of 2025, and yen, Singapore dollar and euro tokens are multiplying at the margins. The EU has built a wall; the US has built a market – and walls keep things out, while markets, inconveniently, are where everyone has to shop.
But the old woman in the lecture hall got the last word, and she gets it here too. Ask what a stablecoin rests on, then what that rests on, and keep going as long as you like. For the foreseeable future, the answer at every level is the same: dollars all the way down
