The Great Depression: Banking Collapse, Deflation, Recovery

Key Takeaways

• The Great Depression is not the 1929 crash. Federal Reserve History dates it from August 1929 to 1941, and the crash itself was over by mid-November 1929, about a year before the first banking panic began.
• The Federal Reserve Bank of St. Louis records real GDP falling 29 percent from 1929 to 1933, unemployment reaching a peak near 25 percent, consumer prices falling 25 percent, and about 7,000 banks, nearly a third of the system, failing.
• The banking collapse arrived in three distinct waves: regional panics from November 1930 through 1931, a national panic beginning in the fall of 1931 after Britain left the gold standard, and a final national collapse in early 1933 that ended in a week-long nationwide bank holiday.
• The Federal Reserve had the legal authority to act as lender of last resort and largely did not. Federal Reserve History calls this failure the System's most serious sin of omission, and the money supply fell by nearly 30 percent from the fall of 1930 through the winter of 1933 as a direct result.
• The 1933 to 1936 recovery was real, and it was interrupted. A second recession from May 1937 to June 1938 cut real GDP by 10 percent and pushed unemployment back to 20 percent, four years into what had looked like a durable recovery.
• Nearly every specific gap this page documents, no lender of last resort, no deposit insurance, a gold standard that overrode domestic banking needs, has since been closed by deliberate policy design, which is the most important fact for a modern reader to carry away.

What Is the Great Depression, and How Is It Different From the 1929 Crash?

Swoopr's 1929 crash case study covers a specific autumn: the Dow Jones Industrial Average peaked at 381.17 on 3 September 1929, fell nearly 13 percent on 28 October and nearly 12 percent on 29 October, and had lost almost half its value by mid-November. That page also covers the 25-year price recovery of the Dow itself, ending in November 1954. This page starts where that one stops being about the stock market and starts covering the decade the crash opened onto: a banking system that failed in three separate waves, a deflation that ran for three and a half years, a currency that was deliberately devalued, a wholesale rewrite of financial regulation, and a second recession that arrived four years into the recovery, in 1937.

Federal Reserve History dates the Great Depression from August 1929, the month the National Bureau of Economic Research marks as the start of the underlying business-cycle contraction, through 1941, when full output and employment returned during the buildup to the Second World War. The essay describes it plainly as "the longest and deepest downturn in the history of the United States and the modern industrial economy." That thirteen-year span is not one continuous slide. It has an identifiable shape: crash, regional bank runs, a national banking and currency crisis, a policy-driven bottom, a genuine recovery, a second recession, and a wartime return to full employment.

Keeping the crash and the Depression separate matters for a specific reason. The equity-market story ends with a number: 89 percent, trough to peak, and a known recovery date. The Depression has no equivalent single number, because it was not one event happening to one market. It was a banking system failing, a currency regime breaking down, and a price level collapsing, each on its own timeline, and each requiring its own explanation.

How Did a Stock Market Crash Turn Into a Banking Crisis?

Federal Reserve History's account of the fall of 1930 makes an important and counterintuitive point: the U.S. economy actually looked like it was on schedule for a normal recovery in the months after the crash. The three prior contractions, in 1920, 1923 and 1926, had lasted an average of fifteen months, and the downturn that began in the summer of 1929 had already run for about fifteen months by the fall of 1930. A typical postwar recession would have been ending. Instead, a series of commercial bank failures beginning in November 1930 converted what had been an ordinary recession into the start of the Great Depression.

The structural weakness that let a regional failure cascade into a systemic one was already in place before 1929. When the crisis began, roughly 8,000 commercial banks belonged to the Federal Reserve System, and nearly 16,000 did not, operating largely as they had before the Fed's creation in 1914: keeping part of their reserves as vault cash and the rest as deposits in correspondent banks in larger cities, a structure known as the reserve pyramid. A country bank needing cash during a panic had to draw on a correspondent, which might itself be under stress and have to draw on a further correspondent in turn. Reserves moved slowly through this chain, and check-clearing float meant a portion of what banks counted as reserves, called fictitious reserves at the time, existed as balances in transit rather than cash in any vault.

That fragility turned a single conglomerate's failure into a multi-state event within weeks, which is the subject of the next section.

What Happened When Caldwell and Company Collapsed in Nashville?

Caldwell and Company was the largest financial holding company in the American South, a Nashville-based conglomerate that sold its clients banking, brokerage and insurance services through a chain of affiliated firms. Its parent company had invested heavily in securities markets and lost substantial sums as prices fell after 1929. To cover those losses, its leadership drained cash out of the operating subsidiaries it controlled.

The Caldwell collapse and its immediate spread, from Federal Reserve History's account of the banking panics of 1930-31.

Federal Reserve History records that within a few weeks, hundreds of banks suspended operations across the region. About one-third of those institutions reopened within a few months; the majority were liquidated. This is the pattern that recurs throughout the next three years: a localized shock exploits the correspondent-clearing structure to spread well beyond the institutions that were actually at fault, and depositors who had no exposure to Caldwell lose access to their own, unrelated banks purely because panic does not respect the boundary between a sick institution and a healthy one standing next to it.

Why Did the Bank of United States Fail in New York?

Panic from the Caldwell collapse was already fading in early December 1930 when a second, unrelated failure reignited it. On December 11, 1930, the Bank of United States, the fourth-largest bank in New York City by deposits, ceased operations. The institution had been negotiating a merger, with the Federal Reserve Bank of New York assisting in the search for a partner. When those negotiations broke down, depositors rushed to withdraw funds, and New York's superintendent of banking closed the bank.

The failure generated national headlines, and Federal Reserve History notes explicitly that it stoked fears of financial panics and currency shortages reminiscent of the Panic of 1907, inducing jittery depositors elsewhere to pull money out of banks that had nothing to do with either Caldwell or the Bank of United States. The name itself did not help: despite sounding like a federal institution, the Bank of United States was a private commercial bank with no government backing, a distinction that was not obvious to a depositor reading a newspaper headline in 1930. A new banking crisis erupted the following June in Chicago, this time driven substantially by real-estate losses among networks of nonmember banks, confirming that the pattern from Nashville and New York was systemic rather than a one-time coincidence of two unrelated failures landing close together.

Why Did the Same Panic Hit Some Federal Reserve Districts Harder Than Others?

One of the more useful facts in the entire episode is that the twelve Federal Reserve districts did not respond identically to the 1930 panic, and the outcomes diverged accordingly. The crisis began in the Sixth District, headquartered in Atlanta. Atlanta Fed leadership treated its lender-of-last-resort responsibility as extending to the broader banking system: it expedited discount lending to member banks, encouraged those member banks to extend credit to their nonmember respondents, and moved funds quickly into towns experiencing runs.

The Eighth District, headquartered in St. Louis, took the opposite approach. Its leadership held a narrower view of the Reserve Bank's responsibilities and refused to rediscount loans for the purpose of assisting nonmember institutions.

The outcomes diverged in the way that view would predict. Federal Reserve History records that after the crisis, the Sixth District's contraction slowed and recovery began, while in the Eighth District hundreds of banks failed, lending declined, and business and employment conditions worsened. This is described in the underlying research literature as close to a natural experiment: two districts facing a similar shock, differing mainly in the response of their regional Reserve Bank, produced measurably different local outcomes. It is direct evidence that the extent of the Depression's damage was not solely a function of the shock itself. It depended on what the institution meant to absorb that shock actually did.

Why this matters for reading later crises. "The Fed didn't stop it" is often treated as a single, uniform failure. The district-level record says otherwise: parts of the same institution, operating under the same charter in the same month, made different choices with different consequences. A crisis response is not one decision made once. It is made by many people, in many places, often disagreeing with each other in real time.

What Happened When Britain Left the Gold Standard in 1931?

The 1930 and early-1931 panics had been regional. That changed on September 21, 1931, when Great Britain left the gold standard, withdrawing its promise to convert bank notes into a fixed quantity of gold on demand. Foreign holders of dollar assets, worried the United States would follow, began converting dollars into gold, producing what Federal Reserve History calls an external drain on the U.S. gold stock. Simultaneously, domestic depositors grew nervous about bank safety and pulled currency out of their own accounts, an internal drain running alongside the external one. Together, the two drains reduced the money supply and deepened the deflation already underway.

The Federal Reserve Bank of New York's response was to raise its discount rate and acceptance buying rate in early October 1931, aiming to make U.S. financial assets relatively more attractive to keep gold from leaving. That defended the external position. According to the economic-history literature Federal Reserve History cites, it did nothing for the internal drain, and a more aggressive program of open-market purchases could have offset the domestic damage the rate increase itself was contributing to. The external drain ended by late October 1931, but bank failures continued regardless. Congress created the Reconstruction Finance Corporation in January 1932 partly in direct response to this continuing pressure, and the Fed ran a large-scale open-market purchase program from April to August 1932, buying about $1 billion in government securities, after which the first acute bank-failure wave finally ended.

That relief did not last through the end of 1932, which sets up the final and most severe panic, covered after a short detour through the RFC itself.

What Was the Reconstruction Finance Corporation?

President Hoover signed the Reconstruction Finance Corporation Act on January 22, 1932, creating a new government-sponsored lender with authority to extend credit directly to banks, other financial institutions and railroads. Federal Reserve Board governor Eugene Meyer lobbied for its creation, helped staff it, and chaired its board. The RFC represented a real departure: it was the first large-scale instrument of direct federal lending to troubled financial institutions during the Depression, a role the Federal Reserve itself had been reluctant to fill for member and nonmember banks alike.

The RFC's existence did not by itself solve the problem it was built for, and the reason is instructive. Federal Reserve History records that by early 1933, many struggling banks were reluctant to actually borrow from the RFC, because RFC borrowers' names were published, and a bank whose RFC loan became public could see that disclosure read by depositors as a signal of weakness, triggering exactly the run the loan was meant to prevent. A facility that only works if institutions are willing to be seen using it is a weaker facility than its balance sheet alone would suggest, and this same tension between a support mechanism's design and the market's reaction to observing that mechanism used shows up repeatedly in later financial crises.

How Many Banks Actually Failed, and What Did That Cost the Economy?

The Federal Reserve Bank of St. Louis states the headline figure directly: roughly 7,000 banks, nearly a third of the entire U.S. banking system, failed between 1930 and 1933. That is not a rounding artifact of a bad year. It is the sustained result of the three waves already described: the regional panics of late 1930 and mid-1931, the national panic that began after Britain left gold in September 1931, and the final collapse in early 1933.

The economic cost went well beyond the deposits lost in each individual failure. Federal Reserve History cites research finding that regional banking crises disrupted the process of credit creation itself: firms paid higher prices for working capital, and some could not obtain credit at any price. This mattered most in districts where large numbers of banks failed at once, because a failed bank's loan officers took with them the accumulated, informal knowledge of which local borrowers were good credit risks and which were not. Rebuilding that local credit-assessment capacity is not something a surviving bank can do quickly, which is part of why the damage from a bank-failure wave outlasts the failures themselves.

The three waves of banking failure, in sequence.

What Happened During the National Banking Holiday of March 1933?

By late 1932, the banking system was already fracturing state by state rather than failing as a single national event. Nevada declared the first statewide banking holiday on October 31, 1932, temporarily relieving its banks of the obligation to meet withdrawal demands. Other states followed, and each new statewide holiday pushed more currency demand onto banks in states that had not yet closed, with pressure concentrating heavily on New York City banks that held balances for institutions elsewhere. Michigan's holiday, declared February 14, 1933, came after Detroit banks were unable to secure emergency loans from the RFC, and Federal Reserve History records that the drain on gold reserves accelerated afterward.

The New York Federal Reserve Bank's own gold reserve fell below its legally required 40 percent backing ratio in early March. On March 1 and 2, the Chicago Fed loaned reserves to the New York Fed to help cover the shortfall; on March 3, Chicago refused a further request, citing concern for its own reserve position. The Federal Reserve Board suspended the gold reserve requirement that same day. The System itself, in the words of the definitive Friedman and Schwartz account that Federal Reserve History cites, "shared in the panic that prevailed in New York," and the Federal Reserve Banks closed their doors on March 4, 1933, the day of Roosevelt's inauguration.

The final week of the banking crisis, from Federal Reserve History's accounts of the 1933 banking panics and the bank holiday.

For an entire week, no American could withdraw, deposit or transfer money through a bank. Federal Reserve History's account of the reaction complicates the standard image of 1930s panic: contemporary press reports describe surprisingly little disorder, with stores extending credit for necessities and communities improvising around the closure rather than rioting. About 4,000 banks, roughly a tenth of those that had entered the crisis, never reopened at all. The rest did, on the strength of a federal classification system that separated sound institutions from insolvent ones and let confidence return selectively rather than all at once.

How Deep Was the Economic Contraction, in Real Numbers?

The Federal Reserve Bank of St. Louis summarizes the four-year contraction in four figures, each independently verifiable and each measuring a different part of the economy.

The Great Depression's economic impact, 1929 to 1933, from the Federal Reserve Bank of St. Louis.

Federal Reserve History adds a related, separately sourced figure for the same period: the money supply fell by nearly 30 percent from the fall of 1930 through the winter of 1933, which the essay states reduced average prices by an equivalent amount. The unemployment figure deserves a methodological note that a page built on precision should not skip. No national monthly survey of unemployment existed until 1940, when the Work Projects Administration launched what eventually became the Current Population Survey; the Census Bureau took over the survey in 1942, and the Bureau of Labor Statistics did not become its co-sponsor until 1959. So the 25 percent peak for 1933 is a historical reconstruction rather than a contemporaneously measured monthly rate; a RAND Corporation analysis of the period independently states that unemployment hovered near 25 percent for a few months in 1933, corroborating the St. Louis Fed's figure from a separate institutional source, but the reader should understand it as a well-supported estimate rather than a survey reading in the modern sense.

Read together, these numbers describe an economy that was smaller, had fewer employed workers, and where every dollar of remaining output was chasing goods whose prices had fallen by roughly a quarter. That combination, shrinking output and falling prices simultaneously, is the specific and unusual signature of a debt-deflation spiral, which is the subject of the next section.

Why Did Falling Prices Make the Depression Worse Rather Than Better?

Falling prices sound, in isolation, like good news for anyone holding cash. Federal Reserve History is explicit that the deflation of 1930 to 1933 did the opposite, and traces the mechanism through several distinct channels rather than treating "deflation" as one undifferentiated harm.

The deflation had an identifiable source: bank panics convinced bankers to hold more reserves and the public to hoard cash rather than keep it on deposit, reducing the money actually circulating through checking accounts, the era's principal means of payment. As the effective money supply fell, prices necessarily followed.

From there, Federal Reserve History lists the specific harms in sequence: deflation forced banks, firms and individual debtors into bankruptcy; it distorted ordinary economic decision-making, because a rational actor delays purchases when prices are expected to keep falling; it reduced consumption directly; and it increased unemployment as firms facing falling revenue cut costs. The debt mechanism is the one most worth internalizing, because it explains why deflation is uniquely dangerous in a way that a proportional decline in both prices and debts would not be. A loan is fixed in nominal dollars. When the price level falls 25 percent, a debtor's income and the value of their collateral fall roughly in step with prices, but the dollar amount they owe does not fall at all. The real burden of every fixed-rate debt in the economy rises automatically, without the debtor doing anything differently, purely because the dollars that debts are measured in became scarcer and more valuable.

The gold standard then generalized the problem beyond the United States. Because gold-standard countries had linked interest rates and monetary policy together, U.S. deflation transmitted to trading partners, contributing to financial crises abroad that then reflected back onto the U.S. system, in what Federal Reserve History calls a deflationary feedback loop running both directions across the Atlantic at once.

How Did Roosevelt's Gold Policy Actually Work?

Since 1900, the dollar had been legally convertible into gold at $20.67 per ounce, and the Federal Reserve was required to hold gold equal to 40 percent of the currency it issued. The March 1933 crisis broke that system in practice: the New York Fed could no longer honor the conversion promise, which is the immediate trigger for the national bank holiday covered above. The Roosevelt administration's response unfolded in three distinct phases over less than a year, according to Federal Reserve History's own account.

Phase one: suspending gold, spring and summer 1933. The Emergency Banking Act of March 1933 gave the president authority over gold movements and let the Treasury compel surrender of gold coin and certificates. On April 20, 1933, Roosevelt formally suspended the gold standard by proclamation, halting gold outflows. In May, the Thomas amendment to the Agricultural Relief Act gave the president power to reduce the dollar's gold content by up to 50 percent. On June 5, a congressional resolution voided gold clauses in every contract, public and private, that had promised repayment in gold or its equivalent value; the Supreme Court upheld the constitutionality of that action in a series of 1935 cases.

Phase two: deliberate devaluation, October 1933. The government authorized the Reconstruction Finance Corporation to buy gold at rising prices, which mechanically lowered the dollar's value in terms of gold and foreign currencies still pegged at the old rate. The explicit goal, in the Federal Reserve's own telling, was to raise the domestic prices of commodities like wheat and cotton back toward their 1926 level, reversing the deflation and relieving debtors, banks and businesses simultaneously.

Phase three: a new fixed rate, January 1934. The Gold Reserve Act, signed January 30, 1934, transferred ownership of all monetary gold in the United States to the Treasury and reset the official conversion price to $35 per ounce, a rate that valued the dollar at 59 percent of its pre-1933 gold content. Individuals and firms holding gold, from banks to a dental-supply manufacturer that had previously applied to the Federal Reserve Bank of Cleveland for gold bars used in making false teeth, were compensated in currency at the new rate rather than the old one. The act also created a $2 billion Exchange Stabilization Fund, letting the Treasury manage the dollar's value independently of the Federal Reserve, a tool later used, among other things, to backstop money-market mutual funds during the 2008 financial crisis.

Private ownership of gold bullion and coin, restricted since 1933, was not restored to U.S. citizens until 1974. Federal Reserve History records that a chart of the U.S. price level, indexed to its 1926 average, shows prices bottoming out in March 1933 and then rising, but not returning to their 1926 level until 1944. Most retellings of the New Deal's gold program treat 1933-34 as the moment deflation ended. It is closer to the moment deflation's direction reversed.

What Did Glass-Steagall and the FDIC Change?

The Emergency Banking Act of March 1933 stopped the immediate bleeding. The Banking Act of 1933, commonly called Glass-Steagall after its sponsors, Senator Carter Glass and Representative Henry Steagall, was the structural reform that followed, signed into law on June 16, 1933.

Its best-known provision separated commercial banking from investment banking: institutions that took deposits and made loans could no longer underwrite or deal in securities, and firms that underwrote securities could no longer share directors or ownership with a deposit-taking bank. Commercial banks were capped at earning 10 percent of total income from securities activity, with an exception carved out for underwriting government bonds. Federal Reserve History notes this separation was not especially controversial at the time; the broad view in 1933 was that mixing the two business lines had contributed to the instability just experienced. It became more contested decades later and was substantially repealed by the Gramm-Leach-Bliley Act of 1999.

The act's more contested provision at the time, and its most consequential one for depositors going forward, created the Federal Deposit Insurance Corporation. Representative Steagall insisted on it specifically to protect small rural banks and their depositors; large banks that expected to end up subsidizing weaker small banks opposed it, and President Roosevelt himself reportedly threatened to veto the bill over the provision. It passed anyway, helped by public support in the immediate aftermath of the bank holiday, when many Americans blamed large banks and Wall Street for the preceding four years.

Federal deposit insurance limits since creation, from Federal Reserve History's account of the Glass-Steagall Act. This is a rule that Congress has changed multiple times and can change again.

Deposit insurance directly addresses the mechanism behind every bank run described earlier on this page: a depositor's individually rational response to any rumor of trouble was to withdraw immediately, since being first in line was the only way to be certain of getting paid in full. An insured deposit removes that incentive, and is the single reform here that most directly targets the failure mode that turned a Nashville conglomerate's losses into runs on unrelated banks in Kentucky and Tennessee.

Why Did the Depression Return in 1937, Four Years Into the Recovery?

This is the part of the Great Depression most casual retellings skip entirely, and it is directly relevant to any reader trying to learn something transferable about recoveries. The recovery that began in 1933 was genuine. By 1936 and 1937, output and employment had improved substantially from the 1933 trough. Then, according to the National Bureau of Economic Research's dating, a new recession ran from May 1937 to June 1938, which Federal Reserve History describes as America's third-worst downturn of the twentieth century, behind only the 1920 and 1929 contractions.

The recession within the recovery, from Federal Reserve History's account of the recession of 1937-38.

Two policy decisions are most commonly identified as the proximate causes, and Federal Reserve History presents both without declaring a clean winner, a useful discipline against oversimplifying. First, the Fed doubled reserve requirements in 1936 to absorb what it judged an "injurious" buildup of excess bank reserves, which had swelled from about $859 million in December 1933 to more than $3.3 billion by December 1935, largely because banks scarred by 1930-1933 preferred holding cash buffers over lending even at low rates. Doubling the requirement converted reserves banks had held voluntarily into reserves they were now required to hold, sharply tightening credit. Second, in June 1936 the Treasury began sterilizing gold inflows, preventing gold entering from Europe from expanding the money supply the way it normally would, cutting off a channel quietly fueling the recovery. Fiscal policy added a third drag: the first Social Security payroll tax took effect in 1937, layered on a 1935 income-tax increase, withdrawing purchasing power at the same moment monetary policy was tightening.

The recession ended once policymakers reversed course: the Fed rolled back the reserve requirement increase, the Treasury stopped sterilizing gold and released what it had already sterilized, and the administration returned to expansionary fiscal policy. The recovery that followed, from 1938 to 1942, was described by Federal Reserve History as spectacular, with output growing 49 percent, driven by renewed gold inflows from Europe and the onset of wartime defense spending.

Why 1937 belongs on this page. Chicago Fed president Charles Evans, discussing the aftermath of the 2008 crisis in 2012, pointed to 1937 directly: "There is a natural tendency for policymakers to pull back on accommodation too early before the real rate of interest has fallen to low enough levels. Such errors happened in 1937 when the Fed prematurely withdrew accommodation." Economist Christina Romer separately called it a cautionary tale against withdrawing support too early. A recovery in progress is not proof that the underlying support can be removed; 1937 is the specific historical case that makes policymakers hesitate before doing so.

When Did the Economy Actually Recover, and By Which Measure?

There is no single honest answer to "when did the Great Depression end," because different parts of the economy recovered on different clocks, and a page that picks one number and calls the matter closed is hiding the more useful information.

Recovery on separate clocks. Figures are as verified in Federal Reserve History's own essays and charts, except the pre-holiday bank count, sourced separately below; the Dow's price recovery is covered in depth on Swoopr's 1929 crash page rather than repeated here.

Notice what this table implies about sequencing. The banking system was functionally stabilized within weeks of March 1933. The price level took over a decade longer to normalize. Full employment took even longer than that, and only arrived through wartime industrial mobilization rather than through peacetime policy alone. A recovery that looks complete by one measure, deflation has stopped, banks have reopened, can still leave the economy years away from complete recovery by another measure, output and jobs. Any statement of the form "the economy recovered from the Depression by year X" is choosing one of these clocks and presenting it as though it were the only one running.

Which Warning Signs Were Visible at the Time, and Which Only in Hindsight?

The 1929 crash itself had visible warning signs, margin buying, an eight-year advance, a central bank openly tightening to slow speculation, covered in depth on that page. The Depression that followed had a separate, largely distinct set of signals, and they were legible to different people at different times.

Signals classified by whether they were actionable in real time.

The recurring theme: structural fragility was often visible to people paying close attention, while the specific triggering events and their timing were not. That same distinction, structure as knowable versus timing as unknowable, is what Swoopr's 1929 crash page draws about margin lending ahead of the crash itself. It recurs here because it is a property of how financial crises work, not a coincidence of this episode.

Common Misconceptions About the Great Depression

"The stock market crash caused the Great Depression." The crash was the opening event, not the cause of the decade that followed. Federal Reserve History is explicit that the Depression's severity traces to monetary and banking failures: a nearly 30 percent contraction in the money supply and a Federal Reserve that did not act as lender of last resort during the 1930 to 1933 banking panics. A crash with a functioning deposit-insurance system and an active lender of last resort in place would very likely have produced a materially shorter and shallower downturn.

"It was one continuous decline from 1929 to the end of the 1930s." It was not. There was a genuine, multi-year recovery from the March 1933 trough through 1936 and into 1937, followed by a distinct second recession from May 1937 to June 1938 that cut real GDP by 10 percent again, followed by a second recovery. Treating the entire decade as a single uninterrupted slump erases the 1937-38 recession, which is arguably the more instructive episode for a modern investor because it happened after policymakers believed the crisis had already been handled.

"Everyone lost their savings when the banks failed." Losses were real but concentrated: depositors in the roughly 7,000 banks that actually failed bore direct losses, while the larger population banking at surviving institutions did not, though they still lived through the unemployment and deflation that hit the wider economy regardless of which bank they used.

"Roosevelt's gold policy was just a symbolic New Deal gesture." The economic-history literature Federal Reserve History cites, including Milton Friedman and Anna Schwartz and later research by Barry Eichengreen and Jeffrey Sachs, finds that recovery in the United States and other major economies began at the point each country suspended the gold standard and reflated. The devaluation from $20.67 to $35 per ounce, which cut the dollar's gold value to 59 percent of its prior level, was a deliberate, targeted policy action against deflation, not a symbolic move, even though contemporaries genuinely viewed it as reneging on the government's own gold-backed promises.

"It could never happen again, because we have deposit insurance now." Deposit insurance addresses one specific failure mode documented on this page, the incentive for a depositor to run first, but it does not address every mechanism a modern crisis could exploit, and a modern reader should not treat the presence of the FDIC as proof that a systemic banking crisis is now impossible. What can more defensibly be said is narrower: several of the specific gaps that turned the 1930-1933 panics into a near-total collapse, no deposit insurance, no consistently exercised lender of last resort, a gold standard forcing a choice between defending a currency peg and rescuing banks, have each been deliberately closed since.

What a Reader Can Actually Carry Forward

The Great Depression is the case study in this library least useful as a direct analogy to a modern decline, and most useful as a lesson in how institutional response, not just the size of the initial shock, determines how bad an economic crisis actually gets.

What generalizes

• A lender of last resort that does not act is not a neutral absence. The Federal Reserve had the legal authority to stem the 1930-1933 panics and largely chose not to use it consistently. The gap between what an institution can do and what it actually does under pressure is where a modern reader should look first when assessing any crisis response, current or historical.
• Deflation is not the mirror image of inflation. Because debts are fixed in nominal terms, falling prices make every existing debt heavier in real terms automatically, without any change in the debtor's behavior. This is the specific mechanism, documented in Federal Reserve History's own account, by which a price decline converts survivable debts into unpayable ones across an entire economy at once.
• A recovery in progress is not proof the support behind it can be safely withdrawn. The 1937-38 recession happened four years into a genuine recovery, caused substantially by policymakers who believed, in 1936, that the recovery was durable enough to tighten into. It was not.
• The same institution can respond differently in different places at the same time. Atlanta and St. Louis operated under the same Federal Reserve charter in 1930 and produced measurably different local outcomes because their leadership made different lending decisions. A crisis response is a set of individual choices, not a single monolithic policy.

What does not generalize

• The specific 29 percent GDP decline and 25 percent unemployment peak. Those figures required the combination of an absent lender of last resort, no deposit insurance, and a gold standard forcing contractionary policy at the worst possible moment. Each of those conditions has since been deliberately changed.
• A gold-price devaluation as a policy tool. The dollar has not been convertible into gold at a fixed rate since 1971, so the specific mechanism Roosevelt used to reflate the economy in 1933-34 is not available to a modern policymaker in the same form.
• The 1937-38 double-dip as an exact template. Its specific triggers, doubled reserve requirements and gold sterilization, were era-specific monetary tools. The general lesson, that premature tightening can reverse a fragile recovery, generalizes; the specific instruments used to tighten in 1936 do not.

The one question worth asking now

The most transferable question this decade poses is not "could the 1930s happen again," which requires assuming away deposit insurance, the lender-of-last-resort doctrine and the end of the gold standard all at once. It is narrower: for any institution or market a reader is exposed to today, is there a run dynamic or funding mismatch that would force a rational, well-informed actor to act pre-emptively in a way that makes the outcome worse for everyone, the way an uninsured depositor in 1930 was individually correct to withdraw first? Risk management is where that question gets applied to an actual portfolio rather than left as history.

References

Every figure on this page was verified against the following sources, each retrieved on 26 August 2026:

• Federal Reserve History: The Great Depression: the 1929-1941 dating, the crash/panics/1937-recession shape of the decade, the nearly 30 percent money-supply contraction, the Fed's lender-of-last-resort failure, and the list of reform legislation.
• Federal Reserve History: Banking Panics of 1930-31: the reserve-pyramid structure, the Caldwell and Company collapse dates (November 7, 12 and 17, 1930), the Bank of United States failure (December 11, 1930), the Atlanta-versus-St. Louis district comparison, and the June 1931 Chicago crisis.
• Federal Reserve History: Banking Panics of 1931-33: Britain's September 21, 1931 gold-standard exit, the RFC's January 1932 creation and the Fed's 1932 open-market purchases, the Nevada and Michigan statewide holidays, and the March 1-4, 1933 collapse into the national bank holiday.
• Federal Reserve History: Bank Holiday of 1933: Proclamation 2039 (1 a.m., March 6, 1933), the Class A/B/C bank classification, the Emergency Banking Act's March 9 passage, reopenings from March 13, the March 15 figure of 90 percent of banking resources resumed, and the roughly 4,000 banks that never reopened.
• Federal Reserve History: Banking Act of 1933 (Glass-Steagall): the June 16, 1933 signing, the commercial/investment banking separation, the FDIC's creation and its $2,500 to $250,000 insurance-limit history, and the 1999 Gramm-Leach-Bliley repeal.
• Federal Reserve History: Roosevelt's Gold Program: the pre-1933 $20.67-per-ounce standard, the three-phase 1933-34 suspension and devaluation, and the price-level chart showing 1926-level prices not returning until 1944.
• Federal Reserve History: Gold Reserve Act of 1934: the January 30, 1934 signing, the $35-per-ounce reset (59 percent of $20.67), the $2 billion Exchange Stabilization Fund, and the 1974 restoration of private gold ownership.
• Federal Reserve History: Recession of 1937-38: the May 1937-June 1938 dating, the 10 percent GDP decline, 20 percent unemployment peak and 32 percent industrial-production decline, the reserve-requirement and gold-sterilization causes, and the 1938-42 recovery.
• Federal Reserve Bank of St. Louis: How Bad Was the Great Depression?: the 29 percent real GDP decline, 25 percent unemployment peak, 25/32 percent consumer/wholesale price declines, and roughly 7,000 bank failures between 1930 and 1933.
• Federal Reserve History: Reconstruction Finance Corporation Act: the January 22, 1932 signing date and Eugene Meyer's role.
• RAND Corporation: Is the Unemployment Rate Now Higher Than It Was in the Great Depression?: independent corroboration of the near-25-percent 1933 unemployment rate.
• U.S. Bureau of Labor Statistics: Current Population Survey History: the 1940 launch of the monthly unemployment survey by the Work Projects Administration, the Census Bureau's 1942 takeover, and BLS's 1959 arrival as co-sponsor, supporting this page's note that the 25 percent unemployment peak is a reconstruction rather than a contemporaneous survey reading.
• National Bureau of Economic Research: Lessons from the Federal Bank Holiday of 1933: the figure of roughly 16,790 commercial banks operating immediately before the March 1933 holiday, of which 11,793 had fully reopened by the end of that month, summarizing Jaremski, Richardson and Vossmeyer, "Signals and Stigmas from Banking Interventions" (NBER Working Paper 31088).

Figures deliberately not stated. This page does not give a precise national bank-failure count beyond the "about 7,000" figure the St. Louis Fed itself uses, a state-by-state failure breakdown, an exact monthly unemployment series for 1929-1940 (no such series was published contemporaneously; the CPS began in 1940), or a single "the Depression ended on this date" claim, because the record does not support one date across every measure. Where the shape of an event is well documented but a more precise figure was not found in a source verified this session, the shape is described and the number is left out.

A rule that can change. The $250,000 FDIC deposit insurance limit cited in this article is current federal law as of publication, set by the 2010 Dodd-Frank Act. Congress has raised it numerous times since 1934 and can do so again; readers should verify the current limit at fdic.gov rather than relying on this page for a real-time figure.

This is educational content about a historical episode. It is not investment advice, it is not a forecast, and nothing here should be read as a claim about how any future economic downturn will unfold.

Related Reading

• Market History Case Studies · the full library, with every drawdown and recovery in this series compared side by side.
• The 1929 Crash and the Depression That Followed · the autumn 1929 equity crash and the Dow's own 25-year price recovery, the specific chapter this page does not retell.
• The 2008 Financial Crisis · a banking crisis where the lender-of-last-resort function this page shows the Fed withholding in 1930-33 was actually exercised.
• The Savings and Loan Crisis · a later, smaller test of the same deposit-insurance system Glass-Steagall created in 1933, including the moral-hazard problem that came with it.
• The Great Inflation, 1965-1982 · the opposite monetary failure mode: sustained inflation rather than the debt-deflation spiral this page documents.
• Risk Management · turning the run-dynamics and forced-selling mechanisms in this case study into questions to ask about a real portfolio today.

Frequently Asked Questions