The crisis is about cash, not debt. The deeper story is about control. Every dollar Iraq earns from oil lands first in an account at the Federal Reserve Bank of New York (New York Fed), and Washington decides whether it moves. Iraq owns its money in law. It does not control it.

The cash stopped because the oil stopped. Oil pays for ~90% of the Iraqi state, and almost all of it leaves by tanker through Hormuz. Exports fell from ~3.6m barrels a day in February to 0.33m in April. They recovered to ~2.6m in September, but only while tankers are allowed through.

Income fell faster than volume. Buyers now take discounts of up to $37 a barrel below the benchmark, because they carry war-risk: insurance at roughly 3% to 10% of a ship’s value per transit, against 0.1% to 0.25% before the war. No law requires the discount. Buyers can load Saudi crude from the Red Sea or Emirati crude from Fujairah without entering Hormuz. Iraq has no other route and no other income, and the side with fewer alternatives pays.

The bills do not stop, as we know. The state needs ~10tn dinars ($6.7bn) a month, for the salaries and pensions of nearly 8m people. From January to July it spent 65.4tn dinars ($43.6bn) and collected 39.1tn ($26.1bn), a deficit of 26.3tn ($17.5bn). More than 1m state employees had not been paid for July by early August. Health Minister Abdul Hussein Al-Musawi told medical staff: “There is no money.”

The central bank filled the gap. The Finance Ministry sells Treasury bills to the state banks, and the central bank buys them with newly created dinars. It calls this discounting, not printing, which it says its 2004 law forbids. Foreign Minister Fuad Hussein called it printing, putting the sum at 25tn dinars ($16.7bn) in June. If each bill is repaid with a new one, the two are indifferent.

The reserves paid for the rest. The central bank sells dollars to banks so importers can pay for food, medicine and goods, and it sold $31.6bn from January to July while oil dollars dried up. Reserves fell from $102.1bn in February to $80.6bn in July, or ~$58bn excluding gold.

Prime Minister Ali Al-Zaidi told parliament on 8 October that public debt had passed 208tn dinars ($139bn) when he took office. He gave no breakdown, and the debt sits in three piles of different weighting.

The largest is domestic: 109.5tn dinars ($73bn) at the end of July, up 21% this year. Two-thirds is held by the central bank and most of the rest by state banks, so the state largely owes itself. Bonds sold to the public pay 6% to 8.5% a year.

The second is foreign debt that Iraq pays: $10.1bn in April, by the Finance Ministry’s count, of which ~$3bn dates from before 2003. The ministry publishes the total but not a list of lenders. Iraq owes the International Monetary Fund (IMF) nothing in loans; every past programme has been repaid.

The pre-2003 part comes from two settlements. In 2004 the Paris Club, 18 creditor governments including Japan, Russia, France, Germany, the UK and the US, cut $38.9bn of Saddam-era loans by 80%, citing Iraq’s limited capacity to pay rather than whether the debts were legitimate. In 2005 and 2006 private creditors were settled on comparable terms: small claims in cash, large ones swapped into a 5.8% bond at 20 cents per dollar. That bond is Iraq’s only one in the market. About $0.5bn is left, the next ~$183m falls due on 15 January 2027, and it trades at ~99 cents.

The third pile is the one nobody collects: ~$40bn owed mostly to Saudi Arabia, Kuwait and Qatar for loans made in the 1980s. Iraq has paid nothing on it since 2004. The UAE and Algeria wrote off their shares; the others have not. The IMF still lists it as debt in arrears, with relief “requested (but not yet obtained)”.

By the classic test of odious debt, set by Alexander Sack in 1927, this is the strongest modern case. It was borrowed by a dictatorship, spent largely on the war Iraq launched in 1980, and lent by creditors who knew why. No court or tribunal has ever cited the doctrine to cancel a debt, and in 2004 Baghdad and Washington chose not to invoke it. In May 2025 Mazhar Mohammed Saleh, financial adviser to the then Prime Minister, called these “odious debts”.

One debt is closed. Iraq paid $52.4bn of war compensation for the invasion of Kuwait, the last payment on 13 January 2022. The interest once promised on those awards was dropped by the United Nations in 2005 and never paid.

A growing deficit and no immediate law allowing it to borrow, the government turned to the exchange rate. On 7 October the dinar was devalued from ~1,300 to 1,500 per dollar for the Finance Ministry and 1,520 for the public. The street rate moved to ~1,700 on the day.

The reason is the budget. The state is Iraq’s largest seller of dollars, therefore each oil dollar it sells to the central bank now buys 200 more dinars, ~1tn ($0.7bn) more a month. Its domestic debt fell overnight from ~$84bn to $73bn in dollar terms.

The cost unfortunately falls on those paid in dinars. Iraq imports much of its food, medicine and goods, so their prices rise. The devaluation works like a tariff on every import, except that the revenue reaches the state through its dollar sales. State employees, pensioners, savers and holders of fixed dinar bonds pay it, and the poor pay most, because imported food takes the largest share of what they spend.

The choice was political and fiscal. Prime Minister Al-Zaidi named the alternatives he rejected: compulsory savings, salaries every 45 days, or more borrowing. Each shows on a payslip on one day and carries his name. A devaluation leaves the payslip unchanged and spreads the loss over months through prices. Economists call the habit of reading the number rather than what it buys “money illusion”. Prime Minister Al-Zaidi blamed “speculators.” This is the only tool he could use alone. Iraq has no 2026 budget law, borrowing needs an act of parliament, and a devaluation needs only a cabinet decision.

History says the street rate is set on another corner. The December 2020 devaluation brought price rises and protests in Baghdad and southern cities. In February 2023, after the New York Fed tightened its rules on Iraqi banks, the street rate reached ~1,750 and the government cut the official rate to 1,300; weeks later the street rate was still ~1,550. Egypt devalued in steps from 2022 and the gap widened each time, until it floated the pound in March 2024 with $35bn of Emirati investment and an IMF programme expanded to $8bn behind it. The gap closes when dollars arrive, and Iraq’s dollars arrive through Hormuz and New York.

Iraq is not short of savings, and most of them are not in the US. The central bank held $80.6bn at the end of July, ~$58bn excluding gold. About a third sits at the New York Fed. By its own data in early 2026, $10.5bn was at the People’s Bank of China, $10.2bn at the Banque de France and $3.3bn at the Bank of England.

About a quarter of the reserves is gold: 175.6 tonnes, by the World Gold Council’s latest count. At the end of 2023, 97% of it was held abroad and 3% in Baghdad. The central bank does not say where the gold abroad sits. A former director general of the central bank lists the New York Fed, the Bank of England and the Bank for International Settlements among the holders of Iraq’s reserves, gold included.

Spreading the savings protects what Iraq has, but not what it earns. Savings abroad can still be frozen: since 2019 the Bank of England has refused to release Venezuela’s ~31 tonnes of gold, first to the Maduro government and now to its successor.

The account in New York is the centre of the story. Oil buyers pay into the central bank’s account at the New York Fed, and every payment out of it is screened there. The central bank buys those dollars from the Finance Ministry for dinars and sells them on to banks for imports. Only banks that pass US-backed checks may buy those dollars, and each payment they make goes through a foreign bank, such as JP Morgan or Citi, after an audit by Ernst & Young. The rest cannot reach the dollar at all, and their customers turn to the street.

The arrangement began in May 2003, under a United Nations resolution and a US executive order that shielded Iraq’s oil money from creditors. The UN shield ended in 2011 and the US shield in 2014, and since then the accounts have been Iraqi in law. They stayed in New York because oil is sold in dollars, and moving them would risk the dollar access Iraq needs to import.

One thing did not end. The US President renews a national emergency over Iraq every May, most recently on 4 May 2026, and Iraq has no part in it. The notice says the obstacles to Iraq’s reconstruction “continue to pose an unusual and extraordinary threat to the national security and foreign policy of the United States”. It allows the US President act on Iraqi assets without declaring a new emergency.

Washington has used the account as leverage. In January 2020, after Iraq’s parliament voted to expel US troops, it warned that Iraq could lose access to the account, and the troops stayed. Since 2022 more than 30 Iraqi banks have been barred from dollar transactions at US request. In January 2026 it warned that Iraq’s oil revenue could be at risk if armed groups it opposes joined the next government, and President Trump said the US would “no longer help Iraq” if Nouri Al-Maliki returned as Prime Minister. Al-Maliki did not return. From late February to late June it halted shipments of dollar cash until Iraq accepted new controls.

Iraq has room to act. It has moved most of its savings out of the US, but it cannot move the flow. Every new dollar it earns still lands in New York, where Washington decides its fate.

That is a question of sovereignty, not finance. Jean Bodin defined sovereignty in 1576 as “absolute and perpetual power”. Among its marks he listed the power to coin money and set its value. For Bodin, sovereignty cannot be conditional: a power that another can override is not sovereign.

By that test Iraq is not sovereign, and Washington is the reason. Iraq can pass laws, sign treaties, issue dinars and set their value. What gives the dinar its value is the dollars behind it, and Iraq cannot receive or spend those dollars without Washington’s consent.

In law Iraq is still a sovereign state. The Montevideo Convention of 1933 asks only for a population, a territory, a government and the capacity to deal with other states. Money is not mentioned. For the Latin American states at the conference, the core demand was Article 8: “No state has the right to intervene in the internal or external affairs of another.”

Albeit, several of them were under US financial control at the time. The US ran the Dominican Republic’s customs until 1941 and Haiti’s finances until 1947. A monetary test would have defined them out of statehood. In 1933, as now, states that did not control their own money were still called sovereign.

The eurozone is the objection. France and Germany gave up their own currencies and are still called sovereign. The difference is consent: they pooled that power by treaty, their central banks sit on the European Central Bank’s governing council, and they can leave the European Union by law. Iraq has no seat in Washington and no exit from the dollar.

However, moving the account would not change that. Every dollar must settle in New York, regardless of where it is held. Real control means being paid in something other than dollars, or depending less on the dollars Iraq is paid. The steps that lead there run in order of speed.

Iraq can bring home its gold, almost all of which was held abroad at last count. Germany did the same when it moved 674 tonnes from Paris and New York to Frankfurt between 2013 and 2017. Iraq can also ask Washington, in the regular talks between the two governments, to end the 2003 emergency.

Iraq can bring its banks up to international standards, and Washington loses its usual reason for cutting them off from dollars. It can keep moving savings out of dollars and into gold, building on the two-thirds of its reserves already held outside the US.

Iraq should sell part of the oil in other currencies, meaning settled outside the US system, is the only step that frees the flow itself, and the one Washington will punish. Earning money that is not oil is the final and lasting fix: while ~90% of state revenue arrives as one dollar flow, the account controls the state.

Iraq can take the first steps alone. The ones that matter, the flow and the emergency. The account in New York is the most direct leverage Washington holds over Baghdad, and one of its main instruments in the Middle East. Granting Iraq its sovereignty would mean giving up their ace in the region.

The final say on Iraq’s sovereignty sits in Washington, our proof that Iraq does not have it yet. Sovereignty and Freedom are not bought, but won.

Subscribe now

The Treasury bought at its price, not the market’s. It ran two enlarged long-end buybacks in September, each capped at $6bn. Dealers offered ~$10.5bn both times. The Treasury took $5.2bn and $4.1bn, less than half each time. Even the US Treasury will not overpay for US Treasuries. Total buyback capacity held at $38bn a quarter. The long end got a bigger share. Bessent on August 31: “I don’t think I can change the equilibrium price.” The 10-year rose from ~4.73% to 5.29% on September 29, the highest since 2007. The 30-year cleared 5.6%, the highest since 2002.

The Federal Reserve (Fed) hiked under a chair Trump nominated to cut. On September 16, the Fed raised rates 25 basis points (bps) to a range of 3.75% to 4.00%, 12 to 0, its first hike since July 2023. Governor Barr states more are likely, and futures price another on October 28. We deem that unlikely. Oil fell 4.8% on September 29 on talks to reopen Hormuz. Again. Consumer confidence fell to its lowest since 2014, lower than at any point in 2020. The cap on the Fed’s Foreign and International Monetary Authorities (FIMA) repo facility, which Bessent asked the Fed to raise, remains unchanged at $60bn.

The dollar rose. The dollar index (DXY) went from ~99.6 to ~101.4. The Bank of Japan (BoJ) hiked to 1.25% on September 18, as Washington asked, yet the yen still weakened, to ~157.3 per dollar. The Fed still pays ~2.75 percentage points more than Japan. Investors borrow yen and hold dollars to collect the gap. This is a carry trade. The income is paid today. The fiscal risk tied to the dollar has no date. Therefore, markets collect the income now.

A necessary correction for the record. We stated Japan’s July intervention was funded through FIMA, relying on reports of Bessent’s August 2 post on X. The Fed’s balance sheet shows no FIMA borrowing on any Wednesday from July to September. Tokyo spent ~¥15.4tn between July 30 and August 26 from its own reserves.

What we have not seen yet. A fiscal problem shows when long yields rise and the dollar falls on the same day, as sterling and gilts did in September 2022. September 29 was the opposite: the 30-year at a 2002 high, the dollar index up 0.18%. That pattern needs the Fed to stop hiking. While it hikes, the widening rate gap keeps buyers in the dollar. When the hikes stop, the dollar test begins. Barr wants more hikes. We do not expect them. One of us will be wrong on October 28.

Subscribe now

The story that has not happened yet involves a $10–20bn debt owed to Beijing that legally still exists and that Beijing can no longer collect on. There is no separate bond for this claim to trade: it is a private, bilateral loan between two governments, never issued to the public. The only place this story can show up in price is indirectly, in the same Republic and PDVSA bonds. If China’s claim has effectively dropped out of the queue of creditors expecting to be paid from Venezuela’s oil revenue, there is more left for everyone still in that queue, and the bonds are worth more than they currently trade for.

China’s claim worked as follows. Between 2000 and 2018, China Development Bank (CDB) made 106 loan commitments to Venezuela totalling ~$105.59bn: real cash, handed over up front, spent by Venezuela on whatever it chose. For years afterward, PDVSA sold crude oil to ChinaOil, the trading arm of China’s state oil major, CNPC, in an ordinary commercial sale, at close to the going rate for that grade of crude, to a buyer that needed the oil for China’s own economy regardless of the loans. The one difference from selling to any other buyer: the cash from that sale did not land directly in PDVSA’s account. It first went into a specific account CDB had the contractual right to draw from. The bank took that month’s instalment out and released what remained to PDVSA. One sale, one payment, split in two: not oil traded for money, but an oil sale with the lender given first right to the proceeds. The contract also required that account to hold at least 1.3x the amount currently owed, a buffer rather than the bill itself, so a bad month for oil prices did not automatically tip into a missed payment. Well-engineered protection against price risk. No protection at all against the black swan that flew in.

Odious debt is the doctrine attached to this story now, surfacing immediately: by 8 January, a Renmin University economist was already on record suggesting Caracas would invoke it to justify walking away from Beijing’s claim. The doctrine holds that debt taken on by an illegitimate regime, without benefit to the population, does not bind its successor. It has a genuine precedent, though smaller and stranger. In 1923, Great Britain and Costa Rica agreed to submit a dispute to a single arbitrator, former US President William Howard Taft. The dispute concerned a loan Costa Rica’s ousted Tinoco regime had taken from a British bank, funds that went directly to financing Tinoco’s own escape rather than to the country. Taft ruled that Costa Rica did not have to pay, on the basis that the bank knew where the money was going. That is a genuine legal precedent for the doctrine succeeding. But it involved a sum in the low six figures, a single arbitrator both sides had agreed to in advance, and facts specific enough, personal theft, the lender’s knowledge of it, that Taft went out of his way to reaffirm the general rule that successor governments inherit their predecessors’ debts. Ecuador in 2008 is the precedent most commonly cited instead, and it is weaker than it sounds. Correa’s government spent nearly two years threatening default and running an “illegitimacy audit” that pushed the bonds’ market price down, then defaulted outright and used a follow-on tender offer to repurchase most of the debt at those depressed prices. Ecuador did walk away from most of what it owed, but not because any court found the debt illegitimate. The illegitimacy argument was a mechanism for depressing the price ahead of a buyback, not a case Ecuador ever needed to win. Venezuela has no equivalent route available against China. Ecuador’s tactic worked because its debt was a bond, dispersed across many holders on a public market, cheap to repurchase upon price fall. China’s claim is a private loan held by a single lender, never listed, with no secondary market to depress or buy back from. Venezuela’s only options are direct negotiation with Beijing or repudiation, defended without recourse to any market mechanism. Nobody has repeated what Taft did in 1923 at anything approaching the scale Venezuela’s China debt would require, and no equivalent arbitration mechanism, no agreement by both sides to submit to a ruling, exists between Caracas and Beijing today.

The doctrine may prove irrelevant to the actual outcome. China’s claim was never headed for a courtroom: it is a bilateral loan, not a bond under New York law, so there is no International Centre for Settlement of Investment Disputes case available, no US court where Venezuela could seek a ruling, and no Taft-style arbitration to fall back on. What broke the collection mechanism was not a legal argument. It was oil that stopped moving. China took in ~75% of Venezuela’s crude exports through 2025, averaging above 600,000 barrels a day. In December 2025, weeks before the capture, the US imposed a naval blockade on tankers carrying Venezuelan crude, and shipments to China collapsed almost immediately. On 2 September 2026, the US Energy Secretary said in Caracas that China would have no claim to revenue from new Venezuelan production, alongside deals with Chevron, Eni and GE Vernova to double output within five years. None of this required a court. It required an export licence and control of the crude.

The debt has not gone anywhere. It remains owed, arrears accruing every month it goes unpaid, and there are two ways this resolves: Beijing agrees to a formal write-down, or Caracas formally repudiates the claim and defends that position. Until one of those happens, the debt is legally alive and operationally dead, precisely the gap the bond trade above is betting on.

Scale is what turns this from a footnote into a systemic question. China is now the world’s largest bilateral creditor, having disbursed ~$2tn in overseas loans since 2013, much of it under the same resource-backed structure used in Venezuela: bauxite in Ghana, copper and cobalt in Zambia, mineral concessions in the Democratic Republic of Congo. By 2022, distressed borrowers already made up ~60% of that book, which is not charity. The return China takes is not always cash recovered on schedule: it is guaranteed long-term access to the resource and the diplomatic leverage of being a country’s largest creditor, both worth carrying a distressed loan book to preserve. China’s own restructuring record across roughly 22 countries has averaged 15–20% haircuts, delivered slowly and reluctantly, because a Chinese state bank booking a loss requires approval further up the chain than a commercial bank would need. Standard rescheduling stays within CDB’s own credit committee; anything genuinely strategic rises to the State Council and the Central Foreign Affairs Commission, which Xi Jinping chairs personally. Venezuela, along with Cuba, is already documented as one of two cases where Beijing chose not to press a struggling borrower, for political reasons. But that leniency was extended to Chávez and Maduro specifically, not to an abstract “Venezuela.” Beijing does not need to reverse any policy. It can treat Rodríguez’s government as an ordinary distressed borrower owed no special treatment, since the relationship that earned the leniency no longer exists on the other side of the table.

Beijing’s response has been restrained. Foreign Minister Wang Yi stated that China “cannot accept any country acting as the world’s judge,” called for Maduro’s release, and urged dialogue: pointed language, not a countermeasure. China’s financial regulator asked domestic banks to report their Venezuela exposure, a defensive check rather than retaliation. Independent assessments described Beijing as exercising “strategic patience,” judging escalation over this specific matter unlikely.

The one place Beijing has acted, rather than protested, is currency plumbing, and it is worth being precise here. This is a separate question from how much of China’s own dollar reserves have moved between institutions, which we have established in our last piece. What has changed is the currency China lends in when it acts as creditor to others: the dollar share of Chinese banks’ overseas lending has fallen from ~65% in 2017 to ~50% by 2024, and China has converted existing dollar debt into yuan terms with borrowers including Kenya, Angola and Ethiopia, with Mozambique in talks to follow. China’s Cross-Border Interbank Payment System is on pace for ~$30tn this year, also covered previously. All of this defends against one specific kind of American leverage, interdiction: stopping a payment while it is still in transit, before it reaches its intended destination.

It defends against nothing once the leverage stops operating at the payment level. A renminbi-denominated loan, cleared entirely outside the dollar system, is as exposed as a dollar loan the moment the borrower’s own government stops being the one giving instructions. Washington did not intercept a payment to break China’s mechanism in Venezuela. Through a blockade followed by a change of government, it took control of the entity that decides where the crude goes and who the revenue belongs to. Currency was never the variable. Once the party issuing the instructions changes, the unit those instructions were denominated in makes no difference.

That is the ceiling. The method used to install a government aligned with Washington cannot be repeated, cheaply at least: capturing a head of state is rare and expensive, usable against an isolated government in America’s own hemisphere, not against dozens of countries holding Chinese debt simultaneously. But the mechanism used once that government exists, export licensing and control of state revenue, does not depend on how the government arrived. A capture achieves it. So does an election, an IMF programme, or diplomatic pressure, all cheaper than a raid, and none requiring a single foot on ground. China’s currency strategy is a hedge against leverage that reaches it through the payment system. It is not a hedge against a government changing hands by any means at all. Beijing is owed its $10–20bn, for as long as anyone cares to check the books. Whether a single dollar is ever collected is now a decision that belongs to Washington, not Beijing.

Subscribe now

The money has not left. Part of the headline decline is real, but it reflects how the money is counted rather than who owns it. TIC data attributes a Treasury bond to whichever country its custodian is based in, not to whoever actually owns it. When China shifts custody of a bond it still owns to an account based in Belgium, Luxembourg, the UK or France instead of one that reports directly as China, the TIC figure for China falls even though nothing has been sold.

A separate pool of dollar exposure sits outside this reporting category entirely. China’s state-owned commercial banks hold dollars in their own right, through their own lending and trade financing, ~$1.1tn. Its policy banks, state lenders focused on infrastructure and trade financing rather than ordinary commercial banking, hold a further ~$1tn, mostly in dollars, with no official breakdown published. China’s sovereign wealth fund, the China Investment Corporation, known as CIC, holds a further $450bn in foreign assets, also mostly dollar-denominated. None of this counts as part of China’s official reserves, because reserves are defined narrowly: assets held directly by China’s central bank, the People’s Bank of China, known as the PBoC, and its foreign exchange manager, the State Administration of Foreign Exchange, known as SAFE. Money sitting on a separate institution’s balance sheet does not qualify, however state-owned that institution is.

Add it all together, custody reattribution and separately-held pools both, and China’s real exposure to the dollar looks close to unchanged from before 2022. It has not shrunk, but scattered across more places than it used to sit in, which makes it harder to add up from outside and easy to misdiagnose for a Chinese retreat.

Simultaneously, the opposite is happening with gold. The PBoC has bought gold for 22 consecutive months, taking its declared reserves to ~2,386t (tonnes) by the end of August, ~8% of its own total reserves. But the published numbers do not match the amount of gold entering the country. In May alone, China imported 163t of gold, declared only 9.95t as an addition to reserves, and saw withdrawals from the Shanghai Gold Exchange, known as the SGE, the domestic wholesale market that normally accounts for where imported gold ends up, fall to 63.5t, the lowest reading since February 2020. Widen the lens to the full first half of the year (H1) and the same gap holds at scale: ~865t imported, ~598t moving through the SGE, and only 40t declared as a reserve addition. That leaves ~227t unaccounted for by either official channel, across H1.

Analysis of UK customs data, tracking physical gold shipments leaving London for China rather than relying on anything China itself discloses, put the total brought in through that route at ~1,050t since February 2022, as of last year. That estimate has not been refreshed since, and the real number today is certainly higher. Albeit, a genuine counter-argument exists: some of the PBoC’s reported increase in reserve value likely reflects the rising gold price itself, not new tonnes bought. It does not touch the import versus domestic demand gap above. That gap concerns physical gold entering the country, not how it is valued afterward.

These two facts do not confirm each other. Gold and dollars are not being compared here in a direct sense, and showing that one figure is overstated while the other is understated does not, on its own prove the why. The natural logic may be that China is swapping one for the other: reducing real dollar reliance and using gold as the backing behind that shift. However, the evidence does not support that reading. If gold were ‘replacing’ dollar exposure, the true scale of that exposure, declared reserves plus the hidden pools sitting in state banks, policy banks and CIC, should be falling as gold rises to take its place. It is not, it has stayed close to flat. That is what rules out a currency swap. What is happening is a jurisdiction swap. China is not shrinking its dollar exposure and buying gold in its place. It is holding dollar exposure roughly steady while separately building a gold position that sits inside China’s own vaults rather than a foreign custodian’s. The two moves manage the same underlying risk from opposite directions. Scatter the dollars across enough institutions and countries that no single freeze order can reach all of it. Bring the gold home so it never has to sit inside someone else’s jurisdiction.

The same underlying risk is live again this week, for Russia, whose frozen reserves in 2022 were the reason China drew this lesson in the first place. The EU is discussing moving roughly €210bn of Russia’s frozen assets out of Euroclear, the Belgium-based institution that holds and settles securities on behalf of investors worldwide, and into a new EU-run structure. Belgium does not want to keep carrying the legal exposure alone, and Russia has filed a $230bn lawsuit against Euroclear in a Moscow court. Nothing has been seized, but the assets remain frozen, not confiscated. The detail that matters is that the custodian wants out. Euroclear was built to be a neutral piece of plumbing, holding and settling securities without taking any side. Being the mechanism used to freeze a sovereign’s reserves has pulled it into a geopolitical dispute it was never designed to be part of. It is not only sovereign reserves that turned out to be exposed. The ‘apolitical’ infrastructure turned out to be exposed too.

Infrastructure tells us what politics will not. CIPS, short for the Cross-Border Interbank Payment System, lets banks settle yuan transactions without routing through the Western financial system, meaning the SWIFT messaging network alongside the dollar clearing systems, CHIPS and Fedwire, that actually move dollars between banks. CIPS processed ~$25.5tn in transactions in 2025 and is on pace for ~$30tn this year.

Compare that with mBridge, a separate project China has built alongside several other central banks. mBridge has a more ambitious aim than CIPS. Rather than building a parallel messaging system, it lets central banks settle payments directly with each other on a shared digital ledger, bypassing the correspondent banking system completely. Total volume moved through it since 2021 sits at only ~$55bn, a fraction of CIPS’s scale, and the Bank for International Settlements, known as the BIS, the institution often described as the central bank for central banks, withdrew from the project entirely in October 2024, citing concerns that it could be used to circumvent sanctions. Not everything China builds succeeds at the same pace. CIPS is scaling fast. mBridge, the more ambitious bet, has stayed small. Despite CIPS’s growth, the yuan’s share of global payments messaged through SWIFT has stayed flat at ~3% for years, close to fifth among world currencies, against an economy that accounts for ~18% of global GDP and ~13.5% of global trade. That gap between economic weight and currency weight is worth an eye.

The same direction shows up outside of payments infrastructure. In 2025, China converted dollar-denominated debt owed by Kenya, Angola and Ethiopia into yuan terms, part of a broader lending relationship in which China is now a major creditor to dozens of lower-income countries. China’s onshore bond market is now the world’s second largest, with more than $25tn outstanding. Foreign investors hold ~¥4.2tn of it, mostly through Bond Connect, a route that settles offshore in Hong Kong and avoids dealing with mainland capital controls directly. Bilateral trade settled directly in yuan has ~tripled as a share of China’s total trade since 2017, from a tenth to ~a third, though Beijing has attached no public target to how far it intends to take this. Russia settles more than 90% of a $245bn trading relationship with China in yuan and roubles. Brazil has run a real-yuan settlement arrangement since 2023, backed by a swap line worth ~¥190bn, covering ¥171bn of bilateral trade last year.

China does not hide the goal. It hides the pace. The Five-Year Plan calls for the yuan’s internationalisation to proceed “steadily” and “prudently.” Officials describe it as a process measured in decades, not a campaign with a deadline. A 2026 industry survey of more than 2,500 domestic and over 1,000 foreign firms found that more than 60% of the foreign companies polled plan to hold their yuan usage steady rather than increase it, which cuts against reading this as a sudden or accelerating shift. The approach is selective, favouring trade settlement over any move to open China’s capital account.

That caution holds real weight, because it reflects a constraint rather than a free choice. Fully opening China’s financial system would reopen the door to the capital flight the country fought hard to shut down a decade ago. Between mid-2014 and early 2017, China’s reserves fell by ~$1tn as the central bank sold dollars to defend the yuan, after a surprise devaluation in August 2015 triggered a rush of money leaving the country. Much of it left through Hong Kong, converted into dollars, some disguised as ordinary trade through manipulated invoices. A visible share went straight into property in Vancouver, Sydney, London and several U.S. cities, all of which saw a noticeable wave of Chinese buying during this window. Beijing has not forgotten that episode. Its slow, gradual approach to internationalising the yuan is that memory, still shaping policy today.

There is a historical precedent for this gap between what the official numbers show and what actually happened. The conventional account of the dollar’s rise says sterling still held somewhere between 80 and 90% of global currency reserves as late as 1945 to 1947, and that the dollar only took over afterward, formalised by the Bretton Woods agreement in 1944.

Look past the official reserve statistics and the picture changes. Research using bond market data, by the economic historians Barry Eichengreen and Marc Flandreau, found that the dollar had already overtaken sterling as the currency international borrowers preferred to issue debt in, by the mid-1920s, ~two decades before the reserve figures caught up. That is not the same as saying investors chose US government debt over British government debt. It means that governments and companies anywhere in the world, borrowing internationally with no particular connection to the United States, increasingly chose to denominate their bonds in dollars rather than sterling. That shift was effectively complete by the late 1920s. It simply did not show up anywhere official for another two decades.

When sterling’s decline finally did show up in the official reserve data, the catching-up process itself was not smooth. Sterling’s share of global reserves fell to ~58% by 1950 and to ~11% by 1973. The sharpest jolt came in 1956, during the Suez Crisis. Britain, France and Israel invaded Egypt without American backing, and Washington saw the whole coalition as a liability. Britain and France looked like colonial powers reasserting control, and Israeli presence made the episode look like a Western-aligned attack on an Arab nationalist leader, exactly the optics the US was trying to avoid across the wider region during the Cold War, for fear of pushing Arab nations toward Soviet support. The United States made its support conditional. It declined to back sterling and blocked Britain’s access to International Monetary Fund, known as IMF, assistance until British troops withdrew, even though Britain was one of America’s closest allies. Britain withdrew, and sterling suffered a serious run in the process. It took until 1976, and a full IMF bailout of the United Kingdom, for sterling’s fall from reserve-currency status to reach its final marker.

The lesson in that episode is not about Britain. It is that financial dominance gets used as leverage even against allies, whenever it serves the dominant power’s own strategic interest, and that the official numbers describing a currency’s status are reliably a lagging indicator, catching up years or decades after the underlying shift has already happened.

Which is where this has now landed us, a pattern. Central banks are not abandoning the dollar at the pace the headlines suggest, and China’s own numbers, read properly, show a country repositioning rather than retreating. But the pressure on the dollar’s reserve-currency status is real, and it is not coming from a rival currency winning the argument on its own merits. It is coming from a world moving away from a single, unquestioned centre of financial power toward one with several competing centres, at exactly the moment several of those centres have watched financial infrastructure become weaponised, and a live conflict over the Strait of Hormuz puts a fifth of the world’s oil supply, priced overwhelmingly in dollars, at risk. Nobody knows which currency, or which system, ends up ahead. That uncertainty is the condition under which gold thrives, and it is thriving again now.

The dollars leaving are the least interesting part of this story. Where trust is going is the part worth our time.

Subscribe now

The Treasury is capping the long-end because the Federal Reserve (Fed) will not cut for it. Capping a yield does not remove the pressure that produced it. It re-routes it. The only valve left open is the currency. The dollar is hiding behind the bond’s shadow, which is why the evidence surfaces everywhere bar the bond market.

The Treasury calls the buyback program liquidity support, and has since Yellen revived a dormant tool in May 2024 to help primary dealers offload bonds nobody wanted. What changed on August 19, 2026, was not the tool. It was the size and timing. The 30-year had just hit its highest yield in almost two decades. Bessent answered by doubling the long-end operation, $2bn to at least $4bn per operation, naming it a “Treasury Twist,” lifted from the Fed’s 1960s playbook. Outside the normal quarterly schedule. He called it liquidity support while calling the yields themselves wrong, “not reflecting the underlying fundamentals.” Participants read it the other way: a sign of desperation.

The mechanics explain themselves. The Treasury cannot create money, only the Fed can. So a buyback is funded first, mostly by selling short-term bills, sometimes by drawing down the Treasury’s cash account at the Fed, kept unusually full for this purpose. The bond is not resold or stored. It is retired the moment the trade settles. None of this shrinks the debt by a single dollar, it changes the shape. A liability owed in twenty years becomes one owed in months. Cheaper borrowing now, paid for by refinancing more often, at whatever rate prevails when the bill comes due. And that rate is not the one the Treasury manages. It is the Fed’s. By fighting the long-end, the Treasury has increased its exposure to the one lever it does not control.

Which matters, because the Fed is not cooperating. The Fed’s portfolio still carries duration built for a different decade, when the goal was holding long yields down. New Fed Chair, Kevin Warsh, resigned a governorship in 2011 over exactly this bloat and now runs a task force to unwind their ancient portfolio toward short bills. Follow it through and it does the opposite of what Bessent wants. Duration back into private hands, long yields up, at the exact moment the Treasury buys that same duration down. Two arms of one government, near-identical tools, opposite directions, same curve. At Jackson Hole on August 28, Warsh settled which side he is on: inflation trends have “not meaningfully improved.” September hike odds went from a third to two-thirds within hours.

The Treasury needs the long-end lower. The Fed pushes the short-end higher. Neither touches a deficit that must be funded either way. Nothing is resolved. It is redirected, and redirected pressure will surface.

Deutsche Bank’s George Saravelos names the reality: soft-form financial repression. Rules and quiet official pressure, rather than ‘free’ market pricing, holding a government’s borrowing costs down. Picture the bond market as a kettle, the deficit as the flame. Left alone, pressure escapes as expected. Bond price falls, yield rises, a new buyer is tempted in. The kettle whistles rather than bursts.

Buybacks weld that valve part-shut. Pressure that cannot escape the valve everyone watches finds the one nobody does. If the price of US Treasuries is not allowed to fall, their value to a foreign holder falls, through a weaker dollar. A foreign central bank is not holding a yield. It is holding a yield converted back into its own currency. Suppress the first and the second adjusts. The holder takes the loss either way.

Japan is the demonstration. The yen sits near a 40-year low. Washington and Tokyo jointly intervened in late July, the first since 1998. Not by selling Japan’s Treasuries, which would push US yields up, but by letting Japan borrow dollars against them through the Fed’s FIMA facility, the line that lets foreign central banks pledge Treasuries rather than sell them. Bessent wants its limit raised. He has also pushed the Bank of Japan (BoJ) to hike. Same man, asking the Fed to soften here and the BoJ to tighten there, depending which currency needs defending. In simple terms, the United States is arranging for its largest foreign creditor not to sell. That is not a liquidity operation. That is a country managing an exit.

Now the objection worth taking seriously. The bond market is not flashing. The term premium is the extra yield investors demand for lending long rather than rolling short, and it is measured in percentage points. It sits near 0.8% today. Across 65 years the median has been 1.41. Investors are charging roughly half the historical premium to lend the US government money for 10 years, and that has barely moved in a year. Nobody sets the number. It is pulled from the shape of the curve. If fiscal fear were building, there is where it should show.

Except the curve is the thing being administered. A gauge stops being a gauge when what it measures is held down on purpose. The Treasury buys the long-end so the yield does not rise. Reading a suppressed term premium as calm is reading the sedative as the diagnosis.

Market positioning says the same and complicates a cleaner story. The obvious read was a crowded steepener, caught out by a Fed hiking into a suppressed long-end. September 1 CFTC data does not support it. Hedge funds are net short across the entire curve, 2-year through 30-year. Asset managers are net long across the entire curve, near-mirror image. That is not a curve view. It is the cash-futures basis trade, now the largest structural exposure in the Treasury market. 10-year open interest fell by over 900,000 contracts in the week to September 1. Positioning got lighter into Jackson Hole, not more crowded. Nobody wishes to express a fiscal view in bonds because bonds are where the intervention is.

Not all of it is American. German, French, British, and Japanese yields sit at their own multi-decade highs this month, each for local reasons. Berlin broke its debt limit for defence. Paris is stuck in a budget standoff. London is funding a deficit it has not addressed. Tokyo is defending a currency. Isolate the American charge, and there is a measure for it. Robin Brooks tracks the extra yield the US pays over a currency-hedged basket of its peers, stripping the global move out. That number flipped in 2026 from a small US advantage to a small penalty. The US used to be paid for being the safest borrower in the world. It now pays. Not dramatic, but it is the cleanest read available, denominated in the terms this thesis cares about, and moving as per expectation.

One thing is about to make it worse. Warsh says inflation has not improved. Oil is rising on Hormuz, a supply shock rather than a demand one. Buybacks ease financial conditions into a Fed trying to tighten. The debt rolls shorter, into the rate that rises if he hikes. In a normal cycle hiking brings inflation down. Against a supply shock it raises the cost of capital without producing a barrel. Warsh is not choosing between inflation and growth. He is choosing which to sacrifice, and the Treasury undercuts whichever he picks. More flame under the same welded kettle.

A word on gold, since the last piece was about it. Gold is not the best performer. Silver and platinum ran roughly twice as far trough to peak. They ran further because nobody official is buying them, which is why silver surrendered more than half its move by mid-year and gold did not. Central banks bought a record 289 tonnes in the second quarter, into falling prices. That is a price-insensitive bid, and a currency decision rather than a commodity one. Gold is the low-beta expression of this trade. Worst upside capture, best floor, an official buyer underneath who does not care what he pays because he is not buying an asset. He is leaving one.

Real question: where does this express. Not duration. Both ends of the curve are administered by officials who want opposite outcomes, which makes the 10-year a bet on who wins a bureaucratic argument rather than a view on the economy. The bond market stopped being an economic instrument this quarter and became a political one.

The currency is where the mechanism vents, and it has not started. Hike odds near two-thirds are dollar-supportive, and the dollar index (DXY) already bounced off its August low near 98.8 to around 99.6. Repression is a medium-term dollar negative fighting a near-term dollar positive, and the near-term one is winning. That is the cost of being early, not evidence of being wrong. The cleaner expression is long yen. A currency at a 40-year low, a central bank pushed to tighten, an intervention floor beneath it, against an issuer suppressing its own yields. Although, a cost to carry while the Fed is hawkish. That is the price of a front row seat.

Three things to watch. The November 4 refunding, and whether the Treasury holds this pace or quietly backs off it. The September FOMC, a real hike decision incoming rather than the formality it was a month ago. And the FIMA cap. If the Fed raises it for Japan, that is the tell for how far this pressure bends an independent central bank.

Bessent keeps calling this liquidity, and the bond market keeps pricing this as a confession. The dollars leaving are the receipt, where they are going is the next piece.

Subscribe now

The headline Q2 number, 288.9t bought, +62% YoY, flatters itself. It is a selling story wearing a buyer’s coat. In Q1, Türkiye sold ~75t through currency-defence swaps, gold pledged and due to return, and Azerbaijan trimmed 22t. Russia sold ~22t, as it did again in Q2. By Q2, Türkiye’s swaps had slowed to 4t and Azerbaijan did not repeat. Russia’s steady 22t became the quarter’s largest single sale. Not because it grew, because everyone around it went quiet. The core buyers, China, Poland, Uzbekistan, and Kazakhstan remain.

Russia’s position deserves care in the telling. Exports are running at record volume, yet revenue is down, squeezed by a widening Urals-Brent discount (back to $27/barrel by July), a third of refining capacity knocked offline by Ukrainian drones, and a stronger ruble. The National Wealth Fund’s fiscal rule draws gold and FX automatically when oil revenue falls short. That is the mechanism working as intended, not a country in retreat. Reserves eased from 2,327t in January to 2,277t by August. 2% of the total, ~6-9t/month, 10 straight months. The reserve is paying out exactly as it was built to.

China and Poland remain the trade’s spine. China’s declared 2,346t undersells the truth; analyst reconstruction of London bullion flows points north of 5,000t, because the PBoC buys in London specifically to step away from the dollar, and the metal moves long before Beijing ever confirms it. Poland is buying toward 700t on record. NBP’s Sobon put it plainly: “the price is not a primary consideration for us.” That is insurance against a Russian border, not a trade idea in the Azerbaijani sense.

The clearest tell sits in the split. Gold ETFs sold 45t out of North America the same quarter central banks bought 289t net. Same metal, same quarter, opposite instincts. Insurance capital and trading capital, pulling in different directions.

H1’s total, 345t, is still the weakest since 2022. Buying has not sped up. Türkiye’s crisis simply faded while Russia’s steady drawdown carried on. China and Poland never needed a reason to stop.

Three things to watch from here. The Urals-Brent discount, which sets Russia’s pace. Poland’s climb toward 700t, the clearest confirmation the buy side holds. And the gap between UK gold exports to China and what the PBoC actually discloses, still the sharpest read on how fast the world is really leaving the dollar behind. The next WGC print lands late October. That is when we find out which way this is really running, until then we play our favourite sport; speculation.

Subscribe now