Every five weeks, Washington has to persuade the world to lend it another trillion dollars. The war with Iran makes each of these conversations more expensive and the emergency exit through the printing press more dangerous.
On October 13, 2026, Ray Dalio, founder of Bridgewater Associates, one of the world's largest hedge funds, published a calculation that reads like a superpower's credit card statement. The next day, the U.S. Bureau of Labor Statistics reported that consumer prices had risen 3.7% over the year and gasoline 34.2%. A day after that, Brent crude closed at nearly $106 a barrel, even though on October 15 the International Energy Agency had announced the release of the last 100 million barrels from emergency stockpiles. The entire drama of American public finance in the fall of 2026 fit between those dates.
The paradox looks like this. The country that prints the planet's main reserve currency behaves like a borrower who has to renew a credit line every week. It cannot go bankrupt in its own dollars. But it may discover that other people's money costs more than it is willing to pay.
The Millionaire Who Can't Pay
Dalio begins with an everyday analogy. A stockholder may count as a millionaire at market prices, but on the day a payment is due he needs a buyer for those shares, or his wealth remains a number on a screen. A paper fortune cannot be carried into a bank and used to pay off a debt. In the U.S. Treasury market, the same danger arises at the moment of refinancing: maturing securities must be replaced with new ones, and for that to happen investors must once again agree to hand their money to Washington.
The scale of this operation usually escapes a public that watches the deficit. According to the figures Dalio relies on, about $10 trillion of U.S. debt matures in 2026, rising to $11.1 trillion in 2027 and $12.5 trillion in 2028. Divide $10 trillion by 52 weeks, and you get roughly $190 billion a week.
A trillion dollars every five weeks or so.
The Bridgewater founder states the key point dryly: if investors become even slightly less willing to reinvest, the effect is the same as a sharp increase in the deficit. A financial hole opens without a single new spending bill, simply because some holders prefer to take their money out rather than roll it into fresh Treasury securities.
To be fair, Dalio has been warning about debt crises for a long time. In 2018 he published a book on big debt crises, and in June 2025 a book on how countries go broke. Critics point out, not without reason, that the dollar has not collapsed in the meantime and that U.S. bonds have kept selling. I treat his timing forecasts with restraint: Dalio himself puts the horizon of a possible crisis at roughly three years, give or take two, and that range is honest but guarantees little. The strength of his latest piece lies elsewhere. The key figures come from the Treasury Department and the Congressional Budget Office, not from the author's own model, and they are harder to refute than his tone.
37 Cents on Top: A Budget That Ignores Its Own Credit Line
By the end of September 2026, about $32 trillion in marketable Treasury securities was outstanding, 8% more than a year earlier. In the first nine months of the calendar year, the Treasury issued roughly $1.5 trillion in net new debt, and for the full year, according to the projections Dalio cites, the figure will reach $2.1 trillion.
The Congressional Budget Office has closed the books on fiscal 2026, which ended September 30: the deficit came to $2.0 trillion. Spending reached $7.4 trillion against revenue of $5.4 trillion. The gap was $218 billion larger than a year earlier and $140 billion above the agency's own February projection. As a share of the economy, the deficit rose to 6.2% of GDP, up from 5.8% in fiscal 2025.
Translated into plain English, it is simple. For every dollar collected in taxes, Washington spends a dollar plus roughly 37 cents of borrowed money. A sudden halt to borrowing would require cutting federal spending by more than a quarter, within weeks and with no political preparation whatsoever.
The strangest thing about this arithmetic is its origin. Deficits of this size have historically accompanied wars and recessions; think of 2009 or pandemic-era 2020. Today unemployment is low, the economy is growing, and the hole is widening. The causes are well known. In the summer of 2025, Congress passed a sweeping tax and spending law that extended tax cuts and at the same time raised the debt ceiling by $5 trillion. By the summer of 2026, American budget analysts were noting that the new tax cuts and spending increases had wiped out the entire effect of the tariffs Donald Trump's administration had boasted about a year earlier.
In May 2025, Moody's stripped the United States of its last top credit rating from the Big Three agencies. The market shrugged at the time.
In the fall of 2026, it is no longer shrugging.
The Short-Bill Trap
The most alarming part of Dalio's calculation is hidden in the structure of the debt. Roughly half of new issuance now consists of Treasury bills with maturities of up to one year. The historical benchmark the Treasury aimed for over the years is about 20%.
For the buyer, short-term debt is convenient: the money comes back quickly and interest-rate risk is minimal. Money market funds and banks take bills eagerly, treating them almost like cash. For the government, the logic is reversed. The coupon on a ten-year note issued in 2020 is locked in for a decade. A bill issued today will have to be reissued in three or six months at whatever borrowing cost the market sets at that moment.
Why did the Treasury take this route? The answer is obvious. The long end of the yield curve is expensive in the fall of 2026, and demand for ten- and thirty-year securities is jittery. Short-term issuance makes it possible to save on interest for now and avoid putting pressure on the long-bond market. Dalio names the price of these savings: rollover risk grows, and in a period of rising rates debt service becomes more expensive faster than it would with a longer maturity structure.
The average interest rate on outstanding U.S. debt, by available estimates, is about 3.5%. Market rates are noticeably higher. Every time an old, cheap security matures and is replaced by a new one, the budget pays the difference. The process runs quietly, without headlines, week after week, auction after auction.
In September, the Treasury tripled its buybacks of long-term bonds, raising the maximum size to $6 billion. Officially, the purpose is to support liquidity. In essence, the issuer is scooping up its own long bonds for which the market lacks appetite while simultaneously piling on short-term debt. Washington is shortening the average maturity of its obligations precisely when rates are rising. Judged by today's interest bill, the decision looks thrifty; judged by tomorrow's, it looks risky.
A Trillion Dollars That Fed No One
In fiscal 2026, net interest payments on the federal debt came to roughly $1.05 trillion. That equals about 19.5% of all federal revenue. Nearly one in every five tax dollars goes to bondholders without turning into roads or pensions.
The trend is worse than the snapshot. In the first three quarters of the fiscal year, net interest costs rose by $78 billion, or 10%, compared with the same period a year earlier. Back in fiscal 2024, interest on the debt surpassed defense spending for the first time, and for a country fighting a war in the Middle East, that ratio is especially telling.
Dalio offers another proportion: in 2026 U.S. federal debt equals roughly 579% of the federal government's annual revenue, and by 2028 the figure will rise to 608%. He estimates the annual rollover of principal at about 185% of tax receipts, and total debt payments including principal at about 200% of revenue.
A methodological caveat is needed here, and Dalio himself makes it. Principal repayment is not paid for with taxes: old bonds are retired with new ones. Adding that $10 trillion to current budget outlays would be a mistake. But the figure honestly shows how many times a year Washington has to pass the market's test of confidence.
The most sensitive point in the calculation is the interest rate. By straightforward arithmetic, a 1 percentage point rise in the average cost of marketable debt, on a stock of about $32 trillion, would eventually add about $320 billion to annual interest. Dalio himself, accounting for the maturity schedule, gives a similar figure of about $310 billion. The discrepancy is small, and the conclusion is the same: a single point of interest costs America roughly 6% of all its current federal revenue.
From this follows a mechanism Dalio describes as self-reinforcing. Interest costs rise, the deficit widens, the Treasury issues more securities, supply weighs on bond prices, yields creep up, and interest costs rise again. In credit markets, such a structure is called a death spiral. For now I would call it a cost spiral: death is still far off, but the direction of travel is set.
Hormuz Takes Aim at Bonds
If this were only a matter of budget arithmetic, Washington could play for time for years. The war with Iran has compressed those years into months.
Hostilities began in late February 2026. Before the war, the yield on the ten-year Treasury note was dipping below 4%. A memorandum of understanding between the United States and Iran, signed on June 17, briefly brought tankers back into the strait, and on July 2 Brent futures fell to $70. Fighting resumed on July 7, and the market reverted to pricing in a half-closed Hormuz.
By mid-October, according to tanker-tracking services, roughly 10 to 11.6 million barrels a day were passing through the strait, against a prewar norm of about 17 million. Brent closed at $104.28 on October 8 and around $105.92 on October 15, with the year's peaks reaching $109 to $110. Diesel is in the worst shape. Diesel refining margins set records in 2026, and diesel matters more to the economy than gasoline: it powers trucks, farms and mines.
Through fuel, the war entered inflation. In September, the consumer price index rose 0.6% for the month and 3.7% for the year. Gasoline rose 9.1% for the month and 34.2% for the year, accounting for more than half of the monthly price increase. Core inflation, excluding food and energy, accelerated to 2.6%.
Through inflation, the war entered the Federal Reserve. The new Fed chair, Kevin Warsh, who succeeded Jerome Powell, said back in the summer that he had "no tolerance" for elevated inflation. In July the rate nonetheless stayed at 3.50% to 3.75%, with three committee members, including Minneapolis Fed President Neel Kashkari and Dallas Fed President Lorie Logan, voting for a hike. On September 16 in Washington, the Fed raised its rate by 25 basis points to 3.75% to 4.00%, the first increase since 2023. Warsh explained the decision in a single sentence: "Inflation has been too high for too long." Five days later, on September 21, Kashkari added that inflation was excessive across all sectors of the economy, not just in oil.
Yields reacted before the central bank did. On September 15 and 16, the ten-year yield rose intraday to 5.04%, its highest since July 2007, and around September 25 it touched 5.23%. In October, after hot inflation data, it again topped 5%.
The Fed's next meeting opens tomorrow, October 27, and concludes on the 28th. A week later, on November 3, America votes in the midterm elections. Throughout 2025, Donald Trump publicly demanded that the Fed cut rates, and now rates are being raised by a chair he himself nominated. A hike on the eve of the election would hit the ruling party through mortgages and auto loans. Holding off would hit confidence in the Fed through inflation expectations.
The energy shock and the debt shock converge at this point. Expensive oil keeps inflation up, inflation keeps rates up, high rates make refinancing more expensive, and more expensive refinancing widens the deficit. Even without a new jump in yields, old debt is gradually being replaced by more expensive debt, and rising interest costs pour their own fuel onto the deficit.
The Buyer Left, the Seller Stayed
Debt sells only when someone buys it. Washington has a buyer problem.
Foreign investors hold about 32% of Treasury securities, roughly $9.3 trillion. The Fed owns about another 14%. Together they control almost half the market, and both groups are buying less than before. China cut its reported holdings of U.S. government debt from about $1.3 trillion in 2013 to about $618 billion in July 2026, shedding nearly $78 billion in the past year alone. Japan, the largest foreign holder, is also reducing its position.
Beijing is driven by politics. After Russia's reserves were frozen in 2022, every central bank that does not rule out a conflict with Washington recalculated the risk of keeping its savings in dollar assets. Tokyo is guided by economics: the Bank of Japan abandoned yield curve control in 2024, Japanese rates rose, and the country's pension capital no longer needs to travel to New York in search of yield.
The departed buyers are being replaced by more skittish money. In its October Global Financial Stability Report, the IMF devoted a separate chapter to hedge funds and concluded that they improve market functioning but can amplify stress. A leveraged hedge fund holds Treasuries as long as funding is cheap and volatility is low. When conditions change, it sells faster than any central bank.
This is the main fault line. A Chinese regulator sheds bonds over years and on a schedule.
A hedge fund can do the same in a week.
AI and the Treasury Fight Over the Same Wallet
Government bonds compete with corporate borrowing for investors' money, and in 2026 the Treasury acquired a competitor with an unprecedented appetite. The largest technology companies building data centers for artificial intelligence issued about $220 billion in bonds in the first half of 2026, compared with $108 billion in all of 2025. Corporate equity issuance in the first half was running at an annualized pace of about $1.8 trillion.
According to the IMF, hyperscalers' share of global investment-grade bond issuance rose from less than 1% in 2024 to about 10% in 2026. Credit spreads on their debt have widened by about 30 basis points relative to the broader market since April. The fund estimates the volume of circular deals among AI companies, in which one invests in another while simultaneously receiving contracts for future revenue from it, at about $1.4 trillion. Such a structure cuts both ways: if it succeeds, it multiplies profits; if a single link fails, it spreads losses along the entire chain.
For a data center developer, expensive credit means a rewritten financial model. Roughly speaking, a project that paid for itself at a 4% rate becomes, at 6%, a bet that AI revenue will arrive before the lender asks for its money back. The higher Treasury yields go, the higher the bar for every corporate borrower, since all private debt is priced as a premium over government debt.
On October 13, at the annual meetings in Bangkok, the IMF presented its full Global Financial Stability Report. The fund named a sharp drop in the prices of AI-related assets as the top risk. Next to it on the list were a spike in government bond yields and a large-scale cyberattack. Term premiums on advanced-economy bonds, by the fund's estimate, are near the upper end of their historical range, pushed up by persistent deficits and expectations of further issuance. The IMF separately modeled a scenario in which the war deepens: oil above $130 through the end of the first quarter of 2027 would shave half a percentage point off global growth next year, and in that scenario the U.S. budget would be hit from two sides at once, through slower tax receipts and through the higher rates at which the Treasury would have to refinance its next trillions.
The fund does not paint an apocalypse, however. The IMF acknowledges the resilience of financial markets and corporate earnings, and its officials stressed at a briefing that price-to-earnings ratios have come down because companies keep beating forecasts. The fund rejected comparisons with the dot-com bubble. The sober conclusion is that a cascade of selling remains a scenario, not yet a diagnosis. What triggers it is a genuine shortage of cash among borrowers and investors who must pay their bills here and now; high valuations merely raise the height of the fall. That shortage is exactly what Dalio is writing about.
The Printing Press That Rescues and Robs
An American crisis does not resemble the bankruptcy of a household or a firm. The United States services its obligations in dollars it issues itself. A technical default is possible only as a political choice, if Congress refuses to raise the debt ceiling. If private buyers fall short, the Fed can buy bonds with newly created money.
History knows such episodes. Starting in 1942, the Fed, by agreement with the Treasury, held long-term bond yields at no more than 2.5% so the war could be financed cheaply. The Treasury-Fed Accord of March 4, 1951, restored the central bank's independence, but by then the federal debt, which had exceeded 100% of GDP after the war, had already begun to melt away thanks to inflation and interest rates below it. Holders of war bonds paid for victory with the erosion of their savings, and few of them understood this when they bought.
A more recent example is closer to hand. In September 2022, the budget promises of Liz Truss's government crashed the British gilt market, leveraged pension funds were forced into fire sales, and on September 28 the Bank of England announced an emergency purchase program of up to £65 billion. The market calmed down. Truss lasted 49 days in office. The British lesson is harsh: a central bank can put out the fire, but the political price is paid by the government that started it.
My generation in Baku remembers the flip side of the printing press better than any textbook. In the early 1990s, savings that Soviet families had accumulated over decades turned into paper within a few months. No one declared a default. The money simply stopped being worth anything.
Dalio calls America's choice an agonizing one, and it is hard to argue with him. Expensive borrowing crowds everything else out of the budget. Money printing relieves pressure on rates, but with inflation at 3.7% and gasoline up by a third over the year, it shifts the costs onto all holders of dollars: American retirees, Asian central banks, Middle Eastern sovereign wealth funds, European insurers. The war with Iran makes both outcomes more expensive for Washington at once, because it pushes inflation up precisely when the budget needs low rates.
Warsh, at least for now, has chosen the first path: rates up, the printing press in reserve. The market is testing how long that choice will hold.
Who Pays for Someone Else's War
There are more losers than winners in this setup. The American middle class pays twice: at the pump and on mortgages, whose rates are tied to the ten-year yield. Developing countries with dollar debts pay through higher rates. Oil importers in Europe and Asia pay the war premium in every barrel: according to major investment banks, the risk premium in the Brent price exceeds $20 a barrel, and only a durable agreement between the United States and Iran, which did not exist as of mid-October, can remove it.
The winners are those standing outside the line of fire. Gold, which Dalio has long called the most time-tested form of money, is drawing demand from central banks moving away from dollar concentration. Oil exporters whose flows do not pass through Hormuz are winning too. Azerbaijani oil travels through the Baku–Tbilisi–Ceyhan pipeline to the Mediterranean, and not a single barrel of it depends on the fate of the Iranian strait. With Brent above $100, that means additional foreign currency revenue for Baku and, for Europe, a vivid reminder that the reliability of a route is worth no less than its volume. The same logic adds weight to the Middle Corridor across the Caspian: every month of a half-closed Hormuz works in favor of routes that bypass the Persian Gulf.
Washington itself is in an ambiguous position. The United States remains the world's largest oil producer, and American oil companies are profiting from expensive barrels. Yet by my estimate, the federal budget gains noticeably less from their windfall profits than it loses on the interest costs driven up by that same expensive oil.
Three Years, Give or Take Two
The baseline scenario looks like a slow rise in costs without a catastrophe. Hormuz stays half-open, Brent holds around $100, inflation does not fall below 3%, the Fed keeps its rate around 4%, and the ten-year yield hovers around 5%. Judging by the trend in the first three quarters of fiscal 2026, the budget's interest bill keeps growing by roughly $80 billion to $100 billion a year. Nothing breaks, but every month Washington's room to maneuver narrows.
The optimistic scenario begins with a durable deal on Iran. Some ratings analysts allow that once supplies are restored, Brent could return to $70. Inflation would then retreat, the Fed would get a pause, and yields would fall. The relief would be temporary, because the structural deficit of 6% of GDP is not going anywhere, and the share of bills in the debt has already been built up.
The pessimistic scenario can be summed up in one word: auction. A failed sale of long-term securities triggers selling by leveraged hedge funds. Next comes a forced Fed intervention on the British model of 2022. Rates stabilize for a time, the dollar weakens, gold rises, and the inflation tax on dollar holders becomes explicit.
The fourth option is voluntary fiscal consolidation. After the midterms, both parties will have two years until the 2028 presidential race and not a single incentive to cut benefits or raise taxes. I put the odds of such a turn at minimal.
I state my expectations so that they can be checked. Through December 31, 2026, the U.S. ten-year yield will not once close below 4.5%. At its December 8–9 meeting, the Fed will not lower its rate below the current 3.75% to 4.00%, whatever happens on October 28.
One question remains that neither the Treasury nor the Fed is answering today. Who will buy American debt if Beijing and the hedge funds start selling in the same week?
A Privilege at a Floating Rate
The dollar's strength remains enormous. There is no alternative reserve currency: the yuan remains nonconvertible, and the eurozone has no single sovereign debt of sufficient size. That is precisely why Washington allowed itself to turn its budget into a system of weekly refinancing: a monopolist is forgiven bad habits for a long time.
America's weak spot is the maturity of its debt; the size is secondary. A country that finances half its new borrowing with bills has effectively put its fiscal policy on a floating rate and pegged it to the price of a barrel in the Strait of Hormuz. America's privilege has not disappeared. It has become more expensive, and it keeps getting more expensive every week.
On Monday, November 2, the day before the midterm elections, the Treasury, as it does every week, will auction 13- and 26-week bills. The voters who go to the polls the next day to vote on taxes and war will not notice this auction. The market will. The rate it sets that day will be the most honest verdict on American fiscal policy in 2026: the price of three more months of living on borrowed money for the richest country in the world.