Part 1 - Bretton Woods: a history

Or: how the dollar got its job, and what the job was

The Mount Washington Hotel sits at the foot of the Presidential Range in northern New Hampshire, a white wooden building with red roofs, built in 1902 by a New Jersey industrialist who wanted somewhere fashionable to spend July. In the summer of 1944 it was requisitioned for three weeks by the United States Treasury, which had decided that the post-war monetary order would be designed there, in conference rooms still smelling of furniture polish, by 730 delegates from forty-four countries who would sleep three to a room because the hotel had been mothballed since 1942 and the staff had not yet been recalled.

The choice of venue was deliberate. Henry Morgenthau, the Treasury Secretary, wanted the conference far enough from Washington that delegates could not slip away to consult their embassies, and isolated enough that the American press could be managed by the simple device of controlling the only road in. The conference ran from 1 to 22 July. By the end of it, the assembled delegates had signed off on the International Monetary Fund, the International Bank for Reconstruction and Development (which would become the World Bank), and a system of fixed exchange rates pegged to a US dollar pegged to gold at $35 an ounce. They had also, although none of them put it this way at the time, signed off on the financial architecture of the American century.

The system they produced is universally known as Bretton Woods, after the village. It is less universally understood as what it actually was, which is the monetary leg of a security settlement. The military leg, NATO, would arrive five years later. The two were designed, and have always operated, as a single interlocking apparatus. To understand what is happening now, fifty-five years after the formal monetary system collapsed in 1971, you have to understand what the apparatus was for.

The negotiation that was actually a fight

The standard account of Bretton Woods presents it as a triumph of multilateral cooperation, a moment at which the lessons of the inter-war period were applied with statesmanlike foresight to prevent their recurrence. There is some truth in this. There would have been no conference at all without genuine fear that the failure to construct a stable monetary order after 1918 had been a proximate cause of the Great Depression and, through it, of the war that the Allies were still fighting. The delegates were serious men engaged on a serious problem and they understood the cost of failure.

It was also, beneath the diplomatic surface, a fight between two countries, one of which had already won.

The British delegation was led by John Maynard Keynes, by then sixty-one years old, in failing health, and the most distinguished economist of his generation. The American delegation was led nominally by Morgenthau but operationally by Harry Dexter White, an Assistant Secretary at the Treasury, a competent technical economist, and, as was established with some clarity in the 1990s through the Venona decrypts, a long-standing source of information for Soviet military intelligence. The two men did not like each other. They worked together because they had to.

Keynes arrived at Bretton Woods with a proposal he had been refining for four years. It was called the International Clearing Union, and its operating unit was a synthetic reserve asset Keynes named the bancor. The clearing union would have functioned as a central bank for central banks. Countries with persistent trade surpluses would have been charged interest on their accumulated bancor balances. Countries with persistent deficits would have been able to borrow up to defined limits before adjustment was required. The system was symmetrical. Surplus and deficit countries shared the cost of imbalance.

This was unacceptable to the United States, which in 1944 held approximately two-thirds of the world's monetary gold and was running the largest current-account surplus in history. Under Keynes's system, the United States would have been charged interest on the privilege of having won the economic war. White's counter-proposal, which became the basis of the actual agreement, abolished the bancor, made the dollar the central reserve currency, fixed the dollar to gold at the historical $35 rate, and required deficit countries to adjust toward equilibrium through domestic deflation or devaluation. Surplus countries faced no symmetric obligation.

Keynes lost. He understood that he was losing, and he understood why. In a letter to a colleague during the negotiations he wrote that the Americans had "all the money bags but not all the brains." It is a remark that has been quoted often, usually as an example of Keynesian wit. It is more interesting read as analysis. Keynes was identifying the structural feature of the new order. The country with the gold reserves would write the rules. The rules would favour the country with the gold reserves. Every subsequent strain on the system would emerge from this asymmetry, and every subsequent strain would be resolved, when it was resolved, by the country with the gold reserves changing the rules unilaterally.

What the system actually did

The Bretton Woods agreement came into legal force on 27 December 1945. The system it established did not begin operating in any recognisable form until 1958, when European currencies became convertible on current account. In the intervening thirteen years, the international economy was held together by Marshall Aid, by the European Payments Union, and by a series of bilateral arrangements between the United States and its strategic partners that bore only an indirect relationship to the architecture agreed in New Hampshire. This is worth noting because the period during which Bretton Woods functioned as designed was therefore not twenty-seven years, which is the conventional figure, but at most thirteen, from 1958 to 1971. The first fourteen years were a workaround. The last few years were a managed collapse.

During the period it functioned, the system did three things. The first was to provide a stable framework for the expansion of international trade, which grew at approximately 7 per cent a year in real terms through the 1960s, a rate not seen before and not seen since. The second was to underwrite the reconstruction of Western Europe and Japan, both of which were integrated into a dollar-denominated trading bloc whose terms were favourable to American exporters and whose security was guaranteed by American military power. The third, less often acknowledged, was to extend dollar liquidity into the international system through persistent American balance-of-payments deficits, deficits which were politically tolerable only so long as foreign central banks did not exercise their formal right to convert accumulated dollar balances into gold at the Treasury window.

This third function was the one that broke the system, and it broke it on a timetable that had been visible in the data for at least a decade before the breakage occurred.

The mechanism is known as the Triffin dilemma, after Robert Triffin, a Belgian-American economist who described it in a Congressional testimony in 1960. Triffin's observation was simple. The Bretton Woods system required the United States to supply the world with dollar liquidity, which it could do only by running persistent current-account deficits. Persistent deficits, however, would eventually produce a stock of foreign-held dollars exceeding the US gold reserves backing them. At that point the convertibility commitment at $35 an ounce would become unsustainable. Either the United States would have to deflate its domestic economy severely enough to reverse the deficit, which was politically impossible, or it would have to suspend convertibility, which was diplomatically impossible. Triffin's conclusion was that the system contained a structural contradiction that would, given time, destroy it.

He was correct on the analysis and approximately correct on the timing. By 1960, foreign-held dollar balances first exceeded US gold reserves. By 1966, the gap was significant. By 1971, it was unmanageable. The London Gold Pool, an arrangement under which the major central banks coordinated to defend the $35 price by selling official gold into the market when private demand exceeded supply, collapsed in March 1968 and was replaced by a two-tier system in which official transactions continued at $35 while the private market priced gold at whatever level willing buyers and sellers could find. The two-tier system was a polite fiction. Everyone in the relevant central banks understood that it was a holding action.

The Eurodollar market and the parallel system

There is a subplot to this story which most accounts of Bretton Woods omit, and which matters considerably for what comes later. From approximately 1957 onwards, a market developed in London for dollar-denominated deposits held outside the United States. The market had multiple parents. The Soviet Union and the People's Republic of China both wanted to hold dollar reserves but did not want to hold them in American banks where they could be seized in a diplomatic incident, and so deposited them in European banks instead. American commercial banks, finding their domestic operations constrained by Federal Reserve interest-rate ceilings under Regulation Q, set up London branches that were not so constrained. British merchant banks, denied the use of sterling for international trade finance after the 1957 sterling crisis, found that dollars worked just as well. The Bank of England, anxious to preserve London's role as an international financial centre after the eclipse of sterling, declined to regulate the new market on the grounds that the transactions were not denominated in its currency.

The Eurodollar market grew from approximately $1 billion in 1959 to approximately $50 billion by 1970 to approximately $1 trillion by 1984. It functioned as a parallel monetary system, denominated in dollars but operating outside the regulatory perimeter of the Federal Reserve and outside the formal Bretton Woods architecture entirely. By the late 1960s, the volume of speculative capital available in the Eurodollar market for short-term currency operations exceeded the foreign-exchange reserves of any single central bank, including the Federal Reserve itself. When the speculative tide turned against the dollar, as it did with increasing frequency from 1967 onwards, the formal architecture of Bretton Woods proved inadequate to the defence.

The significance of the Eurodollar market for this series is that it established the operational template for everything that follows. A dollar-denominated financial system, operating outside US regulatory jurisdiction, beyond the reach of the Federal Reserve, in jurisdictions chosen for their regulatory laxity rather than their economic importance, in volumes that eventually overwhelm the formal monetary architecture. The Eurodollar market in 1971 was the prototype. The offshore stablecoin sector in 2026 is the production version. Both rest on the same regulatory arbitrage. Both extract yield from the dollar's reserve status while operating outside the institutions that produce that status. The technical details have changed. The structural logic has not.

The fifteen minutes

By the summer of 1971, the position was untenable. Foreign-held dollar balances exceeded US gold reserves by a factor of approximately four to one. Speculative pressure against the dollar was running at levels the Federal Reserve could not match through open-market operations. The French government, under de Gaulle and then Pompidou, had been systematically converting dollar balances into gold for several years and showed no inclination to stop. The British, on 11 August, formally requested conversion of $3 billion in dollar reserves to gold. This was the proximate trigger.

Richard Nixon convened a meeting at Camp David over the weekend of 13 to 15 August 1971. The attendees were Treasury Secretary John Connally, Office of Management and Budget Director George Shultz, Federal Reserve Chairman Arthur Burns, Under Secretary for Monetary Affairs Paul Volcker, and a handful of senior aides. The meeting was not authorised in any constitutional sense. The Bretton Woods agreement was a treaty obligation. The President had no domestic legal authority to repudiate it. He repudiated it anyway.

The announcement was made on Sunday evening, 15 August 1971, at 9pm Eastern, in a fifteen-minute televised address that pre-empted Bonanza. Nixon called the package the New Economic Policy. It contained a ninety-day wage and price freeze, a 10 per cent import surcharge, a $4.7 billion cut in federal spending, and, almost in passing, the suspension of the convertibility of the dollar into gold at the Treasury window. The suspension was described as temporary. It has now been temporary for fifty-five years.

The fifteen-minute address is worth pausing on because it is the moment at which the dollar's relationship to gold ceased to be an architectural fact and became a discretionary policy of the United States government. Before Nixon's announcement, the dollar was a claim on metal held in vaults at West Point and Fort Knox. After Nixon's announcement, the dollar was a claim on the credibility of American institutions. The metal had not gone anywhere. The promise to convert paper into metal at a fixed rate had. What replaced it was something the textbooks would later call fiat currency but which is more accurately described as trust currency. The dollar held its reserve status not because it could be exchanged for anything in particular but because the institutions of the American state, including the Federal Reserve, the Treasury, the Department of Justice, the Securities and Exchange Commission, and ultimately the United States military, were trusted to defend the value of the instrument.

This is the substitution that matters for everything that follows in this series. The post-1971 dollar is an institutional product. Its value derives from the integrity of the institutions that issue and police it. Anything that erodes those institutions, whether from outside or from within, erodes the value of the instrument. The current administration of the United States contains figures whose private financial positions benefit from the erosion of those institutions. This is not a rhetorical claim. It is a cap-table observation. The series will demonstrate it line by line.

What survived 1971

The collapse of the gold convertibility commitment did not collapse the dollar's reserve status. This is the central historical puzzle of the post-1971 monetary order, and it is the puzzle the rest of the series is in some sense about.

By the conventional logic of monetary history, the dollar should have lost reserve status fairly quickly after 1971. Reserve currencies are reserve currencies because they are reliable stores of value, and a currency whose convertibility commitment has been repudiated by unilateral executive action is, by definition, not a reliable store of value. Foreign central banks held approximately $50 billion in dollar reserves at the moment of Nixon's announcement, balances they had accumulated on the strength of a promise that had just been broken. The rational response was to diversify away from the dollar over the subsequent decade. They did not diversify away. They diversified into.

The reasons are several and they are worth enumerating because they describe the structure of dollar dominance as it has operated for the half-century since.

The first is that there was no credible alternative. Sterling had been a reserve currency and had collapsed into managed decline through the 1950s and 1960s. The Deutschmark was a candidate but the Bundesbank did not want it to be a reserve currency, on the perfectly reasonable grounds that reserve status produces persistent overvaluation and damages export competitiveness. The yen had the same problem and the Bank of Japan held the same view. Gold itself was a candidate, and in fact the gold price rose tenfold between 1971 and 1980, but gold's role as a reserve asset is limited by its physical properties. You cannot settle a Eurodollar interbank transaction in bullion.

The second is that the United States retained the institutional infrastructure of a reserve issuer. Treasury markets were the deepest and most liquid in the world. American capital markets were the most sophisticated. The dollar was already the unit of account for most international trade, particularly oil trade, and changing the unit of account is a coordination problem of formidable difficulty.

The third is that the United States actively cultivated the conditions of continued reserve status, most consequentially through the petrodollar arrangement reached with Saudi Arabia in 1974. The arrangement was simple. Saudi Arabia priced its oil exports in dollars and recycled the proceeds through US Treasury markets and US arms purchases. In exchange, the United States guaranteed Saudi territorial integrity. The arrangement extended through OPEC more broadly over the subsequent decade and converted the dollar's reserve status from a monetary fact into a security fact. Anyone who wanted to buy oil needed dollars. Anyone who held dollars in size needed somewhere to put them, and US Treasuries were where they put them. The dollar's reserve status was no longer a function of convertibility into gold. It was a function of convertibility into oil, and oil's convertibility into dollars was guaranteed by American carriers in the Persian Gulf.

The fourth is the one that gets least attention and which matters most for the present argument. The dollar's reserve status, after 1971, became a public-goods system underwritten by American institutional and military capacity. The yield on that status, which is to say the seigniorage available to the issuer of a reserve currency, accrued to the United States Treasury and through it to the American public balance sheet. Foreign holders of dollar reserves were, in effect, paying a small tax to the United States for the privilege of using the dollar system. The tax was tolerable because the system was useful and the alternative was worse. The yield was socialised through the Treasury and recycled into American public expenditure: defence, infrastructure, social programmes, debt service.

This is the system that is now being privatised. The yield that, before stablecoins existed, accrued to the US Treasury, is now accruing to a small set of equity holders whose political representatives sit in the Cabinet. The mechanism by which the privatisation is occurring is the subject of Part 3. The vocabulary through which it is being narrated is the subject of Part 2. The figure who supplied that vocabulary, fifty years ago, is the figure to whom the series now turns.

Coda: what Bretton Woods was for

It is worth ending this first instalment with a sentence that the rest of the series will assume.

Bretton Woods was a public settlement. The architects in New Hampshire understood that international monetary arrangements are public infrastructure, in the same sense that roads and harbours and telegraph systems are public infrastructure. They are produced through public effort, sustained through public institutions, and they generate public benefits, which are then taxed to fund the public effort that sustains them. The system was imperfect. The negotiations were rigged in favour of the country with the gold. The fixed exchange rates produced strains that eventually destroyed the system. The Eurodollar market grew up alongside it and undermined it. None of these flaws alters the basic structural fact, which is that Bretton Woods was a public system producing public goods.

What is replacing it is not a public system producing public goods. What is replacing it is the asset-stripping of a public system by a private equity-holder class whose political representatives are simultaneously dismantling the institutions that produce the value of the asset being stripped. This is not a normal transition between monetary regimes. It is something for which the monetary literature has not yet developed adequate vocabulary, because the transition has not previously occurred.

The transition is occurring now. The series is about how.

Coming next: Part 2. The man who has been calling for a "New Bretton Woods" since 1975, and the employees who carried his vocabulary into the State Department, the Treasury, and the broadcast booth from which the current US Treasury Secretary announced his candidacy.