Why did Bretton Woods collapse?
One Sunday night in 1971, the United States quietly stopped turning dollars into gold. The system that had organized the world’s money for a generation was over — and the collapse built the economics that now teaches it.
The Sunday announcement
“I have directed Secretary Connally to suspend temporarily the convertibility of the dollar into gold or other reserve assets.”
— Richard Nixon, televised address to the nation, August 15, 1971
It was a Sunday night, and the broadcast interrupted Bonanza. The word “temporarily” was doing a great deal of work. The suspension was never lifted. Most people watching had no idea they had just seen the end of the monetary order that had run the postwar world.
To feel why this mattered, you need to know what the dollar had been until that night. Since a 1944 conference at a New Hampshire resort called Bretton Woods, the world’s major currencies had been tied to the US dollar at fixed rates, and the dollar alone was tied to gold: $35 bought one ounce, and foreign governments could march up to the US Treasury and demand the metal. Every other currency floated on top of that promise. The dollar was, in effect, as good as gold — because it was redeemable in gold.
Nixon’s announcement broke the link at the bottom. The rates were “fixed but adjustable”: countries could revalue in a crisis, and the International Monetary Fund stood by to lend a government enough reserves to defend its rate through a rough patch. For twenty-seven years it held. The full story of how that architecture was designed and how it ran — the negotiations, the institutions, the slow strain — is the spine of Economic History Ch.13.
By the late 1960s the promise was getting harder to keep. The US had been printing dollars to pay for Vietnam and the Great Society at home; those dollars piled up in foreign central banks, and a growing number of them wanted gold instead. France led the run. Charles de Gaulle had his central bank convert dollars to bullion and ship it across the Atlantic, and in a 1965 press conference he said the quiet part out loud: the system let the United States pay its debts to the world in paper it printed itself, a privilege no other country enjoyed. In 1968 the major powers gave up trying to hold the open-market gold price at $35 and split it into two tiers — an official price for governments, a free price for everyone else. It was a patch on a leak.
By the summer of 1971 the leak was a flood. Speculators, betting the dollar would have to be devalued, dumped dollars for marks and yen and gold. In the second week of August, Britain reportedly asked the US to guarantee the gold value of its dollar holdings. That request, more than any model, is what brought Nixon and his advisers to the presidential retreat at Camp David for the weekend that produced the Sunday broadcast.
“The dollar is our currency, but it’s your problem.”
— John Connally, US Treasury Secretary, to European finance ministers, Rome, 1971
“The dollar is our currency, but it’s your problem”
Connally’s line is the whole American posture in nine words. Was it arrogance, or just an honest read of who held the leverage when the system broke?
Break it, or fix it?
“We had a problem and we’re sharing it with the world — just like we shared our prosperity. That’s what friends are for.”
— John Connally’s posture at the December 1971 Smithsonian negotiations, as recounted in contemporary accounts
The American voice in the room said the system was already dead and pretending otherwise only prolonged the agony. The gold was draining; the run was on; any orderly defense of $35 an ounce would have meant handing American reserves to whoever lined up at the window first. Better to suspend convertibility on your own schedule, take the import surcharge as a bargaining chip, and force the surplus countries to the table. Nixon, Connally, and Fed chairman Arthur Burns were not solving a theoretical puzzle. They were stopping a bank run, and the bank was the United States Treasury.
“The fact that many states accept dollars as if they were gold gives the United States the exorbitant privilege of settling its deficits with its own currency.”
— the French objection, voiced by de Gaulle and his finance minister Valéry Giscard d’Estaing, mid-1960s
The European voice said the system did not have to die — it had to be reformed, and the Americans had refused. France had been pressing for years to demote the dollar and put a neutral reserve asset at the center. The IMF had actually created one, the Special Drawing Right, in 1969. The 1968 two-tier gold market and a coordinated set of currency revaluations might have bought the system years, even decades. On this reading the collapse was not inevitable physics; it was a political failure to fix a fixable machine, and the country with the most power to fix it chose to walk.
Where this leaves us
On the morning of August 16, the world woke up to a different monetary order, and most readers of the morning papers did not realize the dollar had stopped being convertible to gold the night before. That December, the major powers met at the Smithsonian in Washington and tried to put the system back together at devalued rates — Nixon called it “the most significant monetary agreement in the history of the world.” It lasted about fifteen months. By March 1973 the major currencies were floating against the dollar, and the thing Bretton Woods had built was gone for good.
But this is the view from the living room. The people inside the Treasury and the Fed had watched this coming for a decade. One economist had even put the collapse in print eleven years before Nixon went on television. If they all saw it, why couldn’t they stop it?
What Treasury and the Fed actually saw
“If the United States corrected its balance-of-payments deficits, the growth of world monetary reserves would slow to a trickle. But if the deficits continued, the foreign dollar balances would soon exceed the American gold stock, and confidence in the dollar would erode.”
— Robert Triffin, Gold and the Dollar Crisis, 1960 (his testimony to Congress, paraphrased in the book)
Triffin published this in 1960 — eleven years before Nixon’s broadcast. He testified to Congress and pressed the warning on the Treasury all through the decade. This was not hindsight. The trap was named while there was still time to do something about it.
Triffin saw a contradiction built into the dollar’s job. The whole world used dollars as reserves, and a growing world economy needed a growing supply of them. The only way the US could feed dollars to the world was to send out more than it took in — to run deficits. But every dollar that went abroad was a potential claim on the same fixed pile of American gold. The more dollars the system needed, the less believable the gold backing became. Supply the world’s money and you erode confidence in it; stop supplying it and you choke world trade. There was no setting of the dial that worked forever.
The numbers were not subtle. American gold reserves and the foreign dollar claims against them crossed like two lines on a chart, and once they crossed, the promise was arithmetic fiction.
| Year | US gold stock | Foreign dollar claims |
|---|---|---|
| 1949 | $25B | $7B |
| 1959 | $19B | $19B |
| 1965 | $14B | $25B |
| 1971 | $10B | $50B+ |
This is what Treasury and the Fed were staring at. The strain was not hidden in a model; it was on the government’s own balance sheet, the same reserves-and-liabilities accounting that governs any issuer of money. The intro-level machinery for reading a government balance sheet — what counts as a reserve, what counts as a claim, what the constraint actually binds — lives in Economics Ch.16.
And the people in the room knew it. The Bretton Woods order ran on a promise that got steadily harder to keep through the 1960s, from the postwar dollar shortage to the moment the gold pool started to crumble — the chronology Economic History Ch.13 follows to its breaking point.
“The gold-exchange standard may, and does, thus help in spreading inflation — or deflation — throughout the world. It is, in this respect, a dangerous mechanism.”
— Robert Triffin, Gold and the Dollar Crisis, 1960
Was Triffin right — or just early?
He predicted the collapse a full decade out, and the collapse arrived. Does that make the diagnosis correct, or did the politics of one decade just happen to deliver what his structural story claimed was inevitable?
Structure or politics?
“The foreign dollar balances would soon exceed the American gold stock several times over. At that point convertibility is a promise the issuer cannot honor, and everyone knows it.”
— the structural reading, after Triffin, 1960
The structural voice points at the table. The gold stock fell from $25 billion to $10 billion while foreign claims climbed past $50 billion. Once redemption demands can outstrip the reserves backing them — and by the late 1960s they could, several times over — the only thing holding the system up is that not everyone asks for gold on the same day. That is the definition of a system living on borrowed confidence. No clever diplomacy reverses an arithmetic this lopsided; it only changes who is standing at the window when the money runs out.
“The system did not collapse of its own weight. It was undermined by specific policy choices — above all the decision to finance the Vietnam War and the Great Society without raising the taxes to pay for them.”
— Barry Eichengreen, Globalizing Capital, on the contingent triggers of 1971
Eichengreen, the leading historian of the international monetary system, argues the timing was made, not fated. The Bretton Woods order had absorbed strain before and could have absorbed more. What broke it in 1971 specifically was a sequence of political choices: deficit-financed war and domestic spending that flooded the world with dollars, a refusal to devalue earlier when it would have been cheaper, and a German and Japanese reluctance to revalue that left the US holding the whole adjustment. Change the politics and you change the date by years. The structural strain set the stage; politics chose the night.
Where this leaves us
Treasury knew. The Fed knew. By 1968 the gold pool was crumbling and the two-tier market was already an admission that the official price was a polite fiction. The question was never really whether the system would break — it was who would break it, and on whose terms. Economists have a name for the trap Triffin described: the Triffin dilemma, the impossibility of being the world’s reserve issuer and a credibly gold-backed currency at the same time. The dilemma was the diagnosis. The politics of August 1971 set the date.
But naming a dilemma is not the same as proving it has no escape. Triffin saw the strain in the gold numbers. The formal proof that the configuration Bretton Woods was trying to hold was logically impossible — not just empirically strained — had been written a decade earlier by a young economist almost nobody outside the profession had heard of. His proof would reshape open-economy economics for the next half-century.
The apparatus the case demanded
“Under fixed exchange rates and perfect capital mobility, monetary policy has no power to alter output; under flexible rates, it regains full power. The choice of exchange-rate regime is a choice about what monetary policy can do.”
— Robert Mundell, “Capital Mobility and Stabilization Policy under Fixed and Flexible Exchange Rates,” 1963 (summarizing the result)
Robert Mundell was a young Canadian economist at the IMF’s research department. His papers, published a decade before Nixon went on air, did something Triffin’s arithmetic could not: they proved, in a small formal model, that the configuration Bretton Woods was trying to maintain was internally inconsistent. Not strained — inconsistent.
Mundell, working alongside Marcus Fleming, asked a simple question: what can a small open economy actually do with its monetary and fiscal policy once money can move freely across its borders? The answer became the workhorse model of international macroeconomics. Its punchline is brutal for a system like Bretton Woods.
In the model, an open economy sits where three relations cross — goods-market balance, money-market balance, and the external balance of payments:
$$Y = C(Y) + I(i) + G + NX(e), \qquad \frac{M}{P} = L(Y, i), \qquad BP: \; i = i^{*}$$With perfect capital mobility the home interest rate $i$ is pinned to the world rate $i^{*}$. Under a fixed exchange rate, defending the peg forces the central bank to passively supply or absorb money, so the money supply $M$ becomes endogenous and monetary policy is powerless. Only by letting the exchange rate $e$ move can the central bank reclaim control of $M$. You cannot fix $e$, keep $i$ free, and hold $i \ne i^{*}$ all at once.
If money can flow freely and you promise to hold your currency at a fixed rate, then any time you try to cut your interest rate to boost your economy, money floods out chasing higher returns abroad, the currency wants to fall, and you have to buy it back — undoing your own rate cut. Promising a fixed rate with open borders quietly hands your monetary policy to the rest of the world. The only way to get it back is to let the exchange rate move.
The result has a name you already half-know: the impossible trinity. A country can have any two of these three, never all three: a fixed exchange rate, free movement of capital, and an independent monetary policy. Bretton Woods wanted all three — fixed parities, increasingly open capital markets, and central banks that set their own course. Triffin saw the mechanical strain in the gold numbers; Mundell proved the strain was not an accident of the 1960s but a logical impossibility baked into the design. The collapse was the apparatus catching up to reality.
If the trinity forbids one combination, it raises a sharper question: when should a group of countries give up their own currencies and share one — accepting the loss of monetary independence on purpose? That is the theory of optimal currency areas, and Mundell himself opened it in 1961, with Ronald McKinnon and Peter Kenen adding criteria soon after: an area can share a currency if labor moves freely across it, if its economies are open and similar enough to face the same shocks, and if it has fiscal transfers to cushion the regions a single interest rate hurts. Born before Bretton Woods fell, this framework became the standard yardstick for the world that came after — and you will see it judging the euro in the next stage.
One more piece closes the genealogy. The case for letting exchange rates float was not invented in 1971; Milton Friedman had argued it in 1953, eighteen years early, in an essay called “The Case for Flexible Exchange Rates.” And the formal account of how a fixed peg actually dies — why speculators attack a currency whose reserves are draining, and why the attack can become self-fulfilling — arrived as back-fill: Paul Krugman’s 1979 model of speculative attacks, then Maurice Obstfeld’s 1986 and 1996 models of crises that happen because everyone expects them to. Those crisis models are named here; you will watch them do their work on 1992 and 1997 in Stage 4. The intellectual lineage that carries Friedman’s flexible-rates case forward sits in the counter-revolution era of the History of Economic Thought timeline.
“The case for flexible exchange rates is, strange as it may sound, very nearly identical with the case for free prices in general.”
— Milton Friedman, “The Case for Flexible Exchange Rates,” 1953
Did Friedman win the argument before it started?
He made the floating-rate case in 1953, when fixed rates were gospel and almost everyone thought he was wrong. By 1973 the world had switched to his side. Was he proved right — or just rescued by events he didn’t cause?
Abandon it, or rebuild it better?
“Flexible exchange rates are a means of combining interdependence among countries through trade with a maximum of internal monetary independence.”
— Milton Friedman, “The Case for Flexible Exchange Rates,” 1953
The Friedman–Mundell voice says the collapse was a liberation, not a catastrophe. Bretton Woods had been forcing countries to choose between defending an arbitrary peg and running their own monetary policy, and the impossible trinity meant they could not honestly have both once capital started moving. Floating gave the choice back. A central bank could finally set interest rates for its own economy instead of for the defense of a number. This is the same monetarist program in which Friedman argued that a central bank should control the money supply — an argument that only made sense once the exchange rate stopped tying the central bank’s hands. The collapse didn’t break the system; it freed it.
“The objective should be to keep exchange rates within target zones around equilibrium values — managed flexibility, not a free float that lets rates wander far from the fundamentals trade depends on.”
— the managed-rates tradition, after John Williamson’s “fundamental equilibrium exchange rate” framework
The reform voice, argued at its strongest by economists like John Williamson and C. Fred Bergsten, says the choice was never “rigid peg or free float.” That is a false binary. A reformed Bretton Woods — wider bands, crawling pegs adjusted on a schedule, target zones anchored to estimated equilibrium rates, a real reserve asset to replace the overburdened dollar — could have kept the discipline of fixed rates without the brittleness. And it is not obvious, looking only at what was known by 1971, that the costs of floating would be lower: the 1970s brought exchange-rate chaos, the 1980s brought competitive devaluations, and Europe spent the next two decades trying to rebuild fixed rates it had just lost. The verdict that floating was obviously right is a judgment made comfortable only by later hindsight; in 1971 the reformers had a serious case, and the world chose the rougher road partly by default.
Where this leaves us
Mundell proved in the early 1960s what Friedman had argued in 1953: the Bretton Woods configuration was internally inconsistent, and floating was the regime the logic implied. The textbook chapter on open-economy macroeconomics — the impossible trinity, optimal currency areas, the policy mix under capital mobility — is largely a formalization of what the system’s failure made impossible to ignore. The collapse demanded the apparatus, and the apparatus is now orthodox. For the companion story — what the central bank can and cannot do once the economy is open, and the tools it reaches for when the exchange rate ties its hands — see the walkthrough on whether central banks can control the economy.
If the apparatus the case demanded is the apparatus we have, the real test is whether it still explains the world the collapse opened. Floating was supposed to hand national central banks back their independence. The next fifty years — the European experiment, the euro, the Asian crisis of 1997 — would test whether the trilemma binds as hard as Mundell said, or whether the modern world of global capital has found a way around it.
What the apparatus explains
“Even a passing acquaintance with monetary history suggests that this center won’t hold. We need to recognize the short-, medium- and long-term challenges this system creates … and consider a synthetic hegemonic currency.”
— Mark Carney, Governor of the Bank of England, Jackson Hole, August 2019
Fifty years after the collapse, the most senior central banker of his decade stood up and proposed re-architecting the international monetary system — a “new Bretton Woods” in carefully institutional language. The question that broke the old system never really went away.
Start with the era the collapse opened. Floating did roughly what Mundell’s model said it would: exchange rates got more volatile, but national monetary independence got real. The clearest proof is the Volcker disinflation — the brutal early-1980s interest-rate campaign that broke American inflation could not have happened under a fixed rate, because defending a peg would have overridden the policy. The Plaza and Louvre accords of the mid-1980s, where the major economies tried to manage the dollar by agreement, showed the limit from the other side: cooperative management works only at the margins, and only when everyone agrees.
Then watch every attempt to rebuild fixed rates hit the same wall. Europe spent twenty years trying to recreate Bretton Woods on a smaller scale. The European Monetary System held through periodic re-pegs until 1992, when speculators — running exactly the Krugman-1979 and Obstfeld-1996 playbook from the last stage — broke the British pound and the Italian lira out of the system in a single September. Britain had tried to hold a fixed rate, keep its capital markets open, and run its own monetary policy. The trinity does not forgive. The euro then took the experiment to its logical end: not a peg but a shared currency, the impossible trinity resolved by deleting monetary independence entirely.
And here optimal-currency-area theory, born in 1961, finally got graded. For the European core — economies that trade heavily with each other and move together — the euro roughly clears the OCA bar. For the periphery it did not. Greece, Portugal, and Ireland after 2008 were the textbook failure: a single interest rate set for Germany, no fiscal transfers to cushion the shock, and labor that could not move fast enough to absorb it. The euro-periphery crisis is a currency union failing exactly the test Mundell wrote down half a century earlier. The full European story — the deepening integration, the capital flows, the design choices — runs through Economic History Ch.18.
The 1997 Asian financial crisis is the same proof in emerging-market form. Thailand, Indonesia, and South Korea each ran a currency informally pegged to the dollar while keeping their capital accounts open and trying to run their own monetary policy — all three of the trinity at once. When confidence turned, the pegs broke in a cascade, a self-fulfilling run exactly as the second-generation crisis models predicted. The crisis literature named in Stage 3 cashes out here: the formal account of why fixed pegs get attacked, applied to a region that learned the trinity the hard way. Economic History Ch.17 treats the Asian crisis as the development model’s stress test.
Now the live frontier, where the apparatus is under real pressure. Hélène Rey argued at Jackson Hole in 2013 that the trilemma overstates the freedom floating buys: a global financial cycle, driven by US monetary policy and global risk appetite, sweeps across borders and compresses even a floating economy’s independence. On her reading the choice is not three-into-two but a starker “dilemma” — you get monetary independence only if you manage the capital account. The IMF, which spent decades preaching open capital accounts, has since 2012 rehabilitated capital controls as a legitimate tool, and a Mundell-Tobin-style tax on financial flows has resurfaced in the post-2008 macroprudential literature. This challenger frame sits in the modern-pluralism strand of the History of Economic Thought, the post-2008 cluster that keeps the question open: History of Economic Thought Ch.17 (modern pluralism).
“The deficiencies of the international monetary and financial system have become increasingly potent. Even a passing acquaintance with monetary history suggests that this center won’t hold.”
— Mark Carney, Jackson Hole, August 2019
Do we need a new Bretton Woods?
Carney, the BRICS bloc, and a chorus of gold and crypto advocates all say the dollar-centered system is broken and needs replacing. The apparatus from Stage 3 has a sharp question for every one of them.
Did floating work?
“The genius of the post-Bretton-Woods system is its flexibility. It has absorbed shocks — oil, debt crises, the end of communism, a global financial crisis — that would have shattered any fixed-rate regime.”
— the floating-worked reading, after Barry Eichengreen, Globalizing Capital
The floating-defended voice says look at the survival record. The post-1971 world has weathered shocks no fixed-rate system could have taken — the oil crises, the Latin American debt collapse, the fall of communism, 2008, a pandemic — and the exchange-rate system bent each time instead of breaking. Every currency crisis since has been the trinity binding a country that tried to have all three, not a refutation of floating: Britain in 1992, Thailand in 1997, Argentina later all broke because they reached for the forbidden combination. The volatility is real and it is a genuine cost. But a system that has flexed through fifty years of shocks without a single Sunday-night collapse has earned its keep. Floating did the job Bretton Woods could not.
“The global financial cycle transforms the trilemma into a dilemma. Independent monetary policy is possible if and only if the capital account is managed.”
— Hélène Rey, Jackson Hole, 2013
The challenger voice, argued at its strongest by Rey and the post-2008 capital-flow literature, says floating bought less independence than advertised. A global financial cycle pulses out from US monetary policy and global risk appetite, and it floods into and out of every open economy regardless of its exchange-rate regime — a floating currency does not insulate you from it. So the real menu is harsher than Mundell’s: a country gets genuine monetary independence only if it is willing to manage capital flows. The IMF’s quiet rehabilitation of capital controls since 2012 is the institutional admission. And from the other flank, voices like Lyn Alden and the de-dollarization camp argue the dollar’s reserve role is a burden the world keeps trying to escape. None of this restores fixed rates — but it says the floating era’s costs were larger, and its freedoms smaller, than the 1973 verdict assumed.
Where this leaves us
Bretton Woods collapsed because the impossible trinity is real, and the floating era that followed has been turbulent in specific, nameable ways — exchange-rate volatility is a genuine cost, reserve accumulation as quasi-mercantilist insurance is a genuine cost, currency crises clustering on capital-account-open emerging markets is a genuine cost. But it has been more sustainable than continued Bretton Woods would have been. Every attempt to rebuild fixed rates at scale re-creates the same binding constraint: the European Monetary System held until 1992 and then it didn’t; the euro core clears the OCA bar and the periphery does not; the Asian pegs broke for the same reason the dollar-gold peg broke. The live disagreement among serious economists is not about the frame — the impossible trinity itself is not contested. It is a disagreement about how completely the trinity binds once global capital moves freely: a parameter-magnitude argument inside Mundell’s frame, not a challenge to the frame. Rey contests how much independence floating really buys; the de-dollarization camp contests the dollar’s institutional role; neither displaces the trilemma. The case for some managed capital-account architecture for emerging markets is genuinely open. The case for re-pegging the major currencies is not.
Where this leaves us
We started on a Sunday night in 1971, watching a president interrupt Bonanza to tell the country, in the word “temporarily,” that the dollar would no longer turn into gold. That was the surface event. Underneath it, Treasury and the Fed had watched the trap close for a decade — Robert Triffin had put the contradiction in print in 1960, and the gold numbers had been crossing the danger line since the start of the 1960s. Then the apparatus arrived to explain why the trap was a trap: Robert Mundell’s model proved that fixed rates, open capital, and independent monetary policy form an impossible trinity, that Friedman had named the failure mode in 1953, and that the textbook chapter on open-economy macroeconomics is largely a formalization of what the collapse forced into the open.
And then we watched that apparatus grade the half-century since. The European Monetary System, the euro, the 1997 Asian crisis — each one is the impossible trinity binding a system that reached for all three. The honest verdict lives in the conditional. Bretton Woods collapsed because the binding constraint was binding; the floating era has been turbulent in specific ways but more durable than the alternative; and the apparatus the case demanded is mainstream-orthodox today. The frontier argument — Rey’s dilemma, the rehabilitation of capital controls, the recurring dream of a new Bretton Woods — is a fight about how hard the trinity binds in a world of global capital, not about whether it binds at all. The next time someone tells you the dollar system is broken and we need to anchor the world again, you have the tool that every such proposal has to survive: pick two of three. The thing that broke Bretton Woods has not gone anywhere.