The German Hyperinflation of 1923: How a Currency Dies

A Currency That Stopped Being a Unit of Account

Money does three jobs: it settles debts, it stores value between the moment you earn and the moment you spend, and it lets you compare a bicycle to a haircut without haggling in bicycles. Hyperinflation destroys these functions in a specific order. The store-of-value job goes first — nobody holds cash overnight if cash loses a visible fraction of its purchasing power by morning. The unit-of-account job goes second, and its collapse is the more disorienting of the two, because when prices are re-quoted daily and then hourly, the ordinary mental furniture of economic life becomes unusable. You cannot tell whether a wage is good. You cannot tell whether a business is profitable. Contracts written last month describe a world that no longer exists.

Germany reached that state in the autumn of 1923. By November, one U.S. dollar traded for something on the order of 4.2 trillion paper marks — a number that is less a price than a statement that the price system had stopped conveying information. The Hanke–Krus hyperinflation table puts Germany’s peak monthly inflation, in October 1923, at roughly 29,500 percent, which is another way of saying that the mark’s purchasing power halved every few days.

Jargon note: the working definition of hyperinflation used in economics comes from Phillip Cagan, in a 1956 study published in Milton Friedman’s Studies in the Quantity Theory of Money. Cagan defined a hyperinflation as beginning in the month prices rise more than 50 percent and ending in the month before the monthly rise drops below that threshold and stays below it for a year. It is a convention, not a law of nature, but it has the virtue of drawing a line that almost nothing crosses.

This essay is not primarily a chronology. It is an attempt to explain why the 1923 German inflation became the single most argued-over episode in monetary history, and what the argument was actually about — because the fight over the German mark is the ancestor of every modern quarrel over whether inflation is “monetary” or “structural.”

The Fiscal Hole Before the Monetary One

The German inflation did not begin in 1923. It began during the war. Imperial Germany financed the First World War overwhelmingly by borrowing rather than taxation, on the assumption — never stated so crudely, but never far from official thinking — that victory and indemnities from the defeated would settle the bill. Defeat inverted the assumption. The debts remained; the indemnities went the other way.

The Treaty of Versailles (1919) established the principle of reparations without fixing the sum. The London Schedule of Payments of May 1921 fixed it: 132 billion gold marks, structured into tranches of bonds. Contemporary observers understood that the final tranche was, in practical terms, a political instrument rather than a debt anyone expected to collect in full — a way of satisfying Allied electorates while leaving room for later revision. But the headline number did real work. It hung over German fiscal politics, made every negotiation about domestic taxation also a negotiation about foreign creditors, and gave every German government a reason to argue that the state’s finances were impossible.

John Maynard Keynes had made the case against the reparations settlement in The Economic Consequences of the Peace (1919), arguing that the sums demanded exceeded Germany’s capacity to transfer and that the attempt would destabilize Europe. The book made him famous and is often read backward as prophecy. It is worth being careful here: Keynes’s argument was about the transfer problem — the difficulty of converting a domestic budget surplus into foreign currency without collapsing your own exchange rate — not a prediction of hyperinflation as such. That distinction turns out to be the hinge of the whole subsequent debate.

Jargon note: the transfer problem is the gap between raising money domestically and paying it abroad. A government can tax its citizens in marks; it cannot pay French creditors in marks they do not want. To transfer, it must earn foreign exchange through exports or sell marks on the exchange market, which pushes the exchange rate down. Keynes and Bertil Ohlin argued this point at length in the 1920s, and it remains alive in every modern discussion of external debt and development finance.

Meanwhile the German state ran persistent deficits it could not close politically. The Weimar coalition governments faced a public that had already absorbed defeat and would not absorb austerity; the propertied classes resisted a capital levy; the parties of the left resisted taxes on wages and consumption. What could not be agreed was printed. The Reichsbank discounted government bills more or less on demand, and the note issue expanded.

1923: The Ruhr and the Acceleration

In January 1923, French and Belgian troops occupied the Ruhr, Germany’s industrial heartland, on the grounds that Germany had defaulted on deliveries in kind. The German government responded with a policy of passive resistance: workers and officials in the occupied zone were instructed not to cooperate, and the state undertook to support them.

This was the decisive fiscal turn. The government simultaneously lost the tax revenue and coal output of its most productive region and took on the obligation to pay the wages of an idled workforce. It met that obligation with the printing press. What had been a rapid inflation became an accelerating one, and once expectations adjusted, the acceleration became self-feeding through a mechanism worth naming.

Jargon note: the Olivera–Tanzi effect describes how inflation erodes real tax revenue. Taxes are assessed on past income and collected with a lag; if prices double between assessment and collection, the state receives half the real revenue it expected. High inflation therefore widens the deficit it was meant to finance, which requires more money creation, which raises inflation further. The fiscal and monetary sides are not separate stories but one loop.

There is a large folklore around this period — wheelbarrows of banknotes, wallpaper made of currency, children building with bricks of money. Some of it is documented, some of it is retrospective embellishment, and it is not where the analytical interest lies. The economically important behaviors were quieter: wages renegotiated weekly and then daily; workers paid at midday so they could spend before evening; firms invoicing in dollars, in gold marks, in tons of coal; a spontaneous drift toward foreign currency and barter that economists now call currency substitution. People did not need a theory of money demand to act on one. When holding cash becomes a losing bet, everyone reduces their cash balances at once — and reduced money demand raises prices for any given money supply, which is why the final phase of a hyperinflation runs faster than the printing presses themselves.

The Argument: Money or the Balance of Payments?

Here is where 1923 earns its place in the history of economic thought. German economists, officials, and journalists at the time were not in agreement about what was happening, and their disagreement maps almost perfectly onto arguments still running.

The balance-of-payments school, whose best-known advocate was Karl Helfferich, held that causation ran from the outside in. Reparations and the trade deficit forced Germany to sell marks for foreign currency; the exchange rate fell; import prices rose; domestic prices followed; and the Reichsbank issued notes only to accommodate a price level that had already risen for reasons beyond its control. On this reading, the note issue was a symptom. Restricting it would not have stopped the inflation; it would only have starved commerce of the means of payment.

The quantity theory school held the reverse. The exchange rate fell because the mark was being issued without limit; the causation ran from money to prices to the exchange rate. On this reading, the balance-of-payments story confused a consequence for a cause, and the remedy was straightforward in principle even if impossible in politics: stop financing the deficit with new money.

The most thorough postwar verdict came from Costantino Bresciani-Turroni, whose The Economics of Inflation (Italian edition 1931; English translation 1937, with an introduction by Lionel Robbins) assembled the statistical record and came down decisively on the monetary side. His account became the standard one, and it is the ancestor of the modern textbook treatment. For a wider tour of how these mechanisms are formalized, see our primer on the quantity theory as an organizing principle and the entity page for the quantity theory of money.

The honest summary is that the monetary account is correct about the proximate mechanism and incomplete about the politics. No quantity of monetary theory explains why the German state could not close its budget, and the reason it could not is inseparable from reparations, occupation, and a distributional conflict that no coalition could resolve. The balance-of-payments school was wrong about the transmission and right that the constraint was external. Readers who want the earlier version of exactly this argument — with different names and the same structure — should read our piece on the bullion controversy, where British pamphleteers ran the identical fight over the Napoleonic-era pound.

The Stabilization, and Why It Was So Fast

The end came quickly enough to embarrass any theory that treats inflation as a slow-moving physical process.

In November 1923 the German government introduced a new currency, the Rentenmark, which entered circulation on 15 November alongside the existing paper mark. Its notional backing was a mortgage charge on German land and industry — an arrangement whose economic substance was thin and whose symbolic substance was the entire point. Hjalmar Schacht, appointed Currency Commissioner on 13 November, fixed the conversion at one trillion paper marks to one Rentenmark, which restored the pre-war dollar parity of 4.2 marks. Alongside the new unit came the part that mattered: the Reichsbank stopped discounting government paper on demand, and the government moved to close its deficit.

The inflation stopped almost immediately. Prices that had been doubling in days stabilized within weeks. The following year, the Dawes Plan (1924) restructured reparations, arranged a foreign loan, and reorganized the Reichsbank, supplying external credibility the domestic arrangements alone could not.

The speed of this reversal is the analytical payoff of the episode, and it is the subject of one of the most cited papers in modern macroeconomics: Thomas Sargent’s “The Ends of Four Big Inflations” (1982). Sargent’s argument is that the four great interwar inflations — Austria, Hungary, Poland, and Germany — ended not through a gradual grinding-down of expectations but through an abrupt and credible change of regime: a set of institutional commitments (an independent central bank forbidden to finance the deficit, a fiscal plan, sometimes external support) that changed what people expected the government to do in future. If expectations are forward-looking, and if the regime change is believed, the disinflation need not cost years of unemployment.

Jargon note: a policy regime is the rule people believe a government is following, not the action it takes on a given day. The distinction is the core of the Lucas critique: relationships estimated under one regime do not survive a change in the rule, because people’s behavior was a response to the rule. Sargent’s hyperinflation paper is the empirical showcase for that idea, and it is the reason rational-expectations macroeconomics kept returning to 1923.

The Sargent reading has its critics. Skeptics note that the German stabilization involved real costs that arrived shortly afterward — a sharp rise in unemployment through 1924, business failures among firms that had prospered on cheap credit — and that “costless disinflation” oversells the case. Others point out that credibility in 1923 was purchased partly with foreign money and foreign supervision, which is not a policy option most governments possess. What survives the criticism is the central point: expectations and fiscal arithmetic, not the printing press alone, determine whether a stabilization holds.

What 1923 Does and Does Not Teach

The German hyperinflation is invoked constantly and often badly. Three cautions are worth stating plainly.

It is not a template for ordinary inflation. Hyperinflations are fiscal events. They occur when a state cannot tax, cannot borrow, and will not stop spending — typically after war, occupation, revolution, or state collapse. An advanced economy with a functioning tax administration and a central bank that can refuse to buy government paper is not one bad quarter away from Weimar, and analogies that skip the fiscal precondition are doing rhetoric rather than analysis. Our survey of inflation’s causes and measures works through the ordinary cases.

The political lesson is more complicated than the popular version. It is common to hear that hyperinflation brought Hitler to power. The chronology does not support the simple version: the Munich putsch of November 1923 failed, and the Nazi electoral breakthrough came in 1930–32, during the deflation and mass unemployment of the Great Depression, under a government pursuing austerity. The more defensible claim is slower and darker — that the inflation destroyed the savings and the social position of a broad middle class, delegitimized the republic among people who had done everything the old order asked of them, and left a reservoir of grievance that the Depression later mobilized. Money illusion does not vote. Ruined creditors, eventually, do.

The distributional story is the one usually skipped. A hyperinflation is a colossal transfer. Debtors are released; creditors, bondholders, pensioners, and holders of savings accounts are expropriated. Firms with physical assets and foreign-currency receipts do well; wage earners do badly during the acceleration and worse during the stabilization. Keynes made the general point in A Tract on Monetary Reform (1923), written as the German episode ran its course: inflation is a form of taxation, and the question of who pays it is a political question wearing a technical costume. Anyone reading a modern inflation debate as a purely macroeconomic dispute is missing the part the participants care most about.

Connection to the Broader Reckonomics Graph

The 1923 episode sits at the junction of several threads on this site. On the monetary side, it is the extreme case that disciplines the quantity theory and connects forward to Friedman on long and variable lags and the later story of monetarism becoming inflation targeting. On the expectations side, it is Sargent’s evidence, which links to the Lucas critique and the design of monetary rules and central-bank reputation. On the international side, the transfer problem runs directly into the design questions settled two decades later at Bretton Woods. And Keynes is present at both ends: as the critic of the reparations settlement in 1919 and as the analyst of inflation-as-taxation in 1923.