The Great Depression, 1929–1933

The Great Depression is the foundational episode of modern macroeconomics — the event that produced Keynes’s General Theory, the empirical core of Friedman-Schwartz monetarism, the Rothbard Austrian reading of the 1920s as a Fed-induced credit boom, and the Bernanke financial-balance-sheet program. No other episode receives as much sustained explanatory effort across as many traditions, and no other episode’s competing readings are as load-bearing for the traditions producing them. The depression is, for macroeconomic theory, what the Industrial Revolution is for economic history: the canonical event that every framework must explain, and explaining it is part of what each framework means.

What happened

US real GDP peaked in the third quarter of 1929 and fell continuously through the first quarter of 1933 — a peak-to-trough decline of approximately 30 percent. The unemployment rate rose from approximately 3 percent in 1929 to 25 percent at the trough in 1933. The price level (CPI) fell by approximately 25 percent over the same period. The banking system suffered four major waves of failures (October 1930, March 1931, October 1931, January–March 1933), with the final wave producing the bank holiday FDR declared on his first day in office. By March 1933, more than 9,000 US commercial banks — approximately one-third of the prewar total — had failed.

The depression spread internationally through the gold-exchange standard, with Germany suffering a comparably severe contraction, the UK leaving gold in September 1931, and the international trading system fragmenting under successive waves of tariffs (Smoot-Hawley 1930) and competitive devaluations. Recovery began at different times in different countries, generally tracking departure from gold; US recovery began in March 1933 with the bank holiday and the suspension of gold redemption.

The Old Keynesian reading

Keynes’s General Theory (1936) is the foundational Old Keynesian reading. The depression is a coordination failure: aggregate demand collapses (in 1929, on Keynes’s account, principally because investment expectations collapse with the stock-market crash and asset-price losses); the resulting unemployment is involuntary; there is no automatic mechanism returning the economy to full employment because falling wages and prices fail to restore demand and may exacerbate the contraction through debt-deflation dynamics; the policy response should be expansionary fiscal and monetary policy to restore demand directly. The framework’s principal claim is that the depression was a demand failure that competent demand-management policy could have prevented or substantially shortened.

The framework’s empirical claim — that fiscal contraction in 1937 (the “Roosevelt recession”) deepened a recovery that had been underway since 1933 — is widely accepted; the framework’s explanation of the initial contraction is more contested.

The Monetarist reading

Friedman and Schwartz’s A Monetary History of the United States (1963) recast the depression as fundamentally a monetary contraction. The Fed permitted M2 to fall by approximately one-third between 1929 and 1933 — the result of bank failures destroying deposit money the Fed failed to offset through expansionary open-market operations. Absent this monetary contraction, the framework holds, the 1929 stock-market crash would have produced a sharp ordinary recession rather than the depression that actually occurred. The depression is on this reading not a demand failure that requires fiscal correction but a monetary failure that the Fed could and should have prevented through aggressive monetary expansion.

The framework’s central evidence is the timing: each of the four bank-failure waves coincides with deepening of the contraction; recovery in 1933 coincides with the suspension of gold convertibility (which permitted monetary expansion). The reading was, by the 1980s, broadly accepted across mainstream macroeconomics as identifying a load-bearing causal mechanism the Old Keynesian framework had underweighted.

The Austrian reading

Rothbard’s America’s Great Depression (1963) reads the 1920s as a Fed-induced credit boom and the 1930s contraction as the inevitable liquidation. On this reading, the Fed’s expansionary monetary policy through the 1920s produced unsustainable malinvestment — capital flowed into long-horizon, capital-intensive projects that consumer time-preferences did not warrant. The 1929 crash and the subsequent contraction were the necessary readjustment; policy intervention to prevent the liquidation (which both Hoover and Roosevelt attempted) prolonged the depression rather than shortening it.

The Austrian framework’s principal disagreement with both the Old Keynesian and Monetarist readings is structural: the depression is not an aberration to be prevented but the inevitable correction of a prior credit boom. The framework’s empirical claims about 1920s credit dynamics are contested; the broad direction of its argument (that prolonged easy money creates unsustainable conditions) has substantial support across other traditions even where the specific Austrian capital-theoretic apparatus is rejected.

The Bernanke financial-balance-sheet reading

Ben Bernanke’s early academic work (1983 AER paper “Nonmonetary Effects of the Financial Crisis in the Propagation of the Great Depression”) extended the Monetarist account by adding financial-system mechanics the Friedman-Schwartz reading had treated as proximate consequences of the monetary contraction. On Bernanke’s reading, the bank failures destroyed not just money but the relationship-based credit-intermediation infrastructure on which small and mid-sized firms depended; this destruction had real effects independent of the monetary contraction, persisting long after price levels stabilised.

The reading is now substantially absorbed into the New Keynesian financial-accelerator program (Bernanke-Gertler 1989, 1995, 1999) and supplied the intellectual framework for the Federal Reserve’s 2008 response under Bernanke’s chairmanship.

The Post-Keynesian / Minsky reading

Hyman Minsky’s financial-instability hypothesis reads the 1920s as an extended period of stable prosperity that endogenously produced increasingly fragile financial structures — the hedge-to-speculative-to-Ponzi progression — culminating in the 1929 crash and subsequent debt-deflation dynamics that Irving Fisher had identified contemporaneously. The depression is on this reading not a monetary failure (Friedman-Schwartz), not a demand failure (Keynes), and not a credit-boom liquidation (Rothbard), but the inevitable unwinding of unsustainable financial structures the prosperous 1920s had produced.

The Minsky framework received minimal mainstream recognition during his lifetime; the post-2008 rediscovery (when “Minsky moment” entered the financial-press vocabulary) substantially elevated this reading of 1929–1933.

Status today

The mainstream synthesis on the Great Depression is principally Friedman-Schwartz monetary contraction plus Bernanke financial-balance-sheet propagation, with Keynesian demand-management mechanics in the recovery phase post-1933. The Austrian and Post-Keynesian/Minsky readings sit outside this synthesis but are taken more seriously than they were a generation ago — partly because the 2008 crisis revealed financial-system mechanics the synthesis had underweighted, and partly because the post-2008 macro-prudential turn drew explicitly on Minsky-tradition framings.

The single most-discussed unresolved question is the policy-counterfactual: would aggressive monetary expansion in 1930–1932 (against the gold-standard constraint) have prevented the depression’s deepening? The Friedman-Schwartz reading says yes; the Bernanke reading says yes but with caveats about the credit-system damage that had already occurred; the Old Keynesian reading is sceptical that monetary policy alone would have sufficed against the demand collapse; the Austrian reading is hostile to the question itself. The Federal Reserve’s 2008 response was structured around the assumption that the Friedman-Schwartz reading was correct; whether the assumption is fully borne out by the post-2008 experience is itself a live debate.