It is difficult to overestimate the scale and the impact of the Great Depression. In the United States, where the depression began and where it hit the hardest, real GDP plummeted 27 percent from its peak in 1929 to its trough in 1933. Unemployment rates reached double digits in nearly every industrial country, climbing to roughly 25 percent in the United States. Developing countries suffered as well, mostly because exports of their primary products declined. Globally, prices of goods fell substantially. Deflation reached more than 10 percent a year in the United States.
Speaking at an IMF Institute seminar on October 3, Romer, a distinguished scholar of the Great Depression, drew on her own research, historical narrative, and the works of other noted economists to piece together the causes and effects of the depression. The story, she noted, was one of repeated shocks to aggregate demand. The shocks, particularly in the United States, were largely monetary—precipitous, repeated drops in the money supply—and had a number of causes. Also, in the United States, banking panics and policy failures played a key role. In a number of other countries, international financial strain and the gold standard were important.
What triggered the U.S. depression?
Although the Great Depression ultimately became a global phenomenon, its epicenter was the United States. Why? Romer emphasized that the depression was not a structural adjustment to correct the excesses of the 1920s. The Roaring Twenties were not the sustained boom they are often made out to be. The decade was characterized by moderate growth, punctuated by three short recessions, and remarkably steady prices. Nothing about the 1920s made the Great Depression inevitable.
The Great Depression in the Unites States came in phases: first, as a mild recession, then as a downward spiral as different shocks hit the economy. These shocks were primarily domestic because the United States had become, after World War I, a nearly closed economy: by the end of the 1920s, imports and exports made up only 5 percent of GDP.
Uncertainty. The first shock was the one now famously connected with the Great Depression—the stock market crash of 1929. For years, the crash was viewed as a side issue. But Romer said her research suggested a more central role for the crash in explaining the initial downturn of the U.S. economy. By the summer of 1929, the economy was sliding into recession as a result of monetary tightening. What had been a gradual decline, however, became a dramatic one when, in October, the stock market plunged 40 percent, most of it within a five-day period. Industrial production plummeted soon after, and, by late 1929, consumer spending had declined sharply.
In seeking explanations for the decline in consumer spending, economists typically look at losses in wealth caused by changes in asset prices. But in 1929 (unlike in the 1990s), not many people were in the stock market, and stock market wealth was a small fraction of total wealth. Likewise, negative expectations cannot explain consumer behavior, because neither press reports nor financial forecasts at that phase pointed to a dire economic outlook. What the crash did do—with its dramatic movements in asset prices—was to cause great uncertainty about future economic developments. It was this uncertainty, Romer argued, that drove down consumer spending and, with it, output.
If consumers become more unsure of future income, they postpone spending, particularly on durable goods. But in this situation, they are likely to maintain or even increase spending on nondurable goods. Romer found this thesis borne out by data on different types of consumer spending in 1929-30. In the six months following the crash, automobile registrations and department store sales declined sharply, but sales in “five-and-ten” (cent) stores fell only slightly, and grocery store sales actually rose. Output of durable goods plummeted, while semidurable goods went down less, and perishable goods remained relatively constant.
The relevant lesson for today’s economies, Romer suggested, is that dramatic movements in asset prices can cause high levels of uncertainty, with subsequent deleterious effects on consumer spending. Japan’s experience over the past decade may be the most recent example. Uncertainty created by the bursting of the investment bubble in 1990 may be one of the factors behind the decline in Japanese spending and hence the long and painful Japanese recession of the 1990s.
Monetary shocks and the Federal Reserve. After the crash and until 1930, Romer explained, the actions of the U.S. Federal Reserve (the Fed) were basically correct, with nominal interest rates declining between 1929 and 1930. But between the fall of 1930 and the end of 1931, three successive monetary shocks hit. These created a massive monetary contraction that choked investment, hastened business failures, and accelerated deflation.
Drawing from narrative histories by Milton Friedman and Anna Schwartz, Ben Bernanke, Barry Eichengreen, and others, Romer explained that these monetary shocks were largely independent of the real economy—that is, they were caused by factors other than the fall in output. Federal Reserve mistakes were crucial.
Over the course of the late nineteenth and early twentieth centuries, banking panics were quite commonplace in the United States. Indeed, the Fed was formed principally to prevent them. But after World War I, two aspects of the U.S. banking sector were setting the stage for severe problems. First, the system was dominated by “unit,” or individual, banks. Most states did not allow branch banking—they did not permit successful banks to open branch offices. This meant, in the late 1920s, that there were many small rural banks with little scope for geographical diversification. Second, the country’s agricultural sector, which had boomed during World War I, now saw farm incomes contract sharply. This left many farmers on the margin, unable to repay bank loans. With bad loans and deflated food and land prices, the banking sector was ripe for crisis.
When the first banking panic hit in the fall of 1930, the Fed’s instinct, honed by the prevailing wisdom of the time (and by past experience under the gold standard), was to do nothing and let the money supply plummet. Why the inaction? Romer suggested two possible explanations. First, the Fed was still a relatively young institution and uncertain about its role in handling monetary shocks. Second, its influential head, Benjamin Strong, had died in 1928, creating a power vacuum. Control over monetary policy shifted to the “liquidationists,” who strongly opposed both fiscal and monetary expansion on the grounds that such policies would hinder readjustment and hurt investor confidence.
A second panic hit in the spring of 1931 and a third in the fall of 1931. Because the Fed did nothing to stem these panics either, by the spring of 1932 the U.S. money supply had declined almost 30 percent. This huge monetary contraction had devastating effects through increases in the real interest rate and more pessimistic expectations.
The panics also took a huge toll on the U.S. banking sector. By 1933, nearly half of the banks in existence in 1929 were no longer operating. Bernanke’s research argued that the scale of this loss destroyed knowledge crucial to the process of credit intermediation. Many banks no longer had long-term relations with their clients or knew which of their small borrowers were creditworthy. In this climate, the cost of credit intermediation rose significantly, aggravated by continuing deflation, which reduced the value of collateral. One sign of the rising cost of credit intermediation was the greatly widened spread between safe yields on government securities and risky interest rates on business loans in the early 1930s. Small borrowers also faced important credit rationing. Bernanke found that these credit channel effects compounded the direct monetary effects of the panics and further depressed real output.
In September 1931, in the midst of this scenario, Great Britain was forced off the gold standard. In response, the Fed raised interest rates substantially to avert fears that the United States might also be forced to go off the gold standard. At the depth of the depression, with the economy reeling from a credit crunch, deflation, and falling output, such a policy choice was tantamount to pumping water into a sinking ship. With the rise in nominal rates, real interest rates also rose. This, Romer emphasized, was not the characteristic response of money to falling output. It was a deliberate policy action that produced a large monetary contraction—and another shock to aggregate demand.
Were the Fed’s actions required by U.S. adherence to the gold standard? Romer discussed her current research with Chang-Tai Hsieh. They found that the United States had significant scope for monetary expansion even during the worst years of the depression. In the spring of 1932, the one period when the Fed did expand the money supply substantially, there is no evidence that monetary expansion triggered large gold flows or expectations of devaluation. This suggests that the monetary contraction that was central to the Great Depression in the United States was the result of policy mistakes, not institutional constraints.
The international scene
What caused the depression to become worldwide? One factor was what Barry Eichengreen refers to as the “golden fetters”—the constraints imposed by the gold standard. The other was a sharp decline in international lending. In both cases, events in the United States played a large part.
The U.S. recession had an immediate and direct effect on primary commodity—producing countries, particularly in Latin America, which were among the first to show economic decline. Although the United States was a small buyer of manufactured goods in international markets, it was a significant buyer of primary commodities. The decline in U.S. production had a direct impact on the exports of primary goods producers. It prompted a drop in prices for primary goods and resulted in a steep decline in the purchasing power of exports in those countries, making the depression especially painful for them. In response, primary commodity—producing countries (and others facing a foreign exchange crisis) had to either increase their interest rates substantially to defend the gold standard or relinquish the standard altogether. One by one, they chose the latter, starting with Argentina in 1930.
Even when the direct effect of the U.S. downturn was small (as in the nonprimary commodity—producing economies), countries were affected by the U.S. decline through induced policy changes. The tightening of U.S. monetary policy resulted in falling output and prices in the United States. The high real interest rates and low prices attracted significant gold inflows into the United States. While many countries were close to the statutory minimum in their gold reserves, the United States and France (which had a deliberately undervalued currency) accumulated gold, draining reserves from other countries. By 1927, roughly half of the world’s gold was in the United States and France. Countries losing gold were forced to deflate by running very tight monetary policies. The result was a massive global monetary contraction that set in motion the worldwide economic downturn.
Lessons from the recovery
The recovery from the Great Depression, Romer argued, holds many lessons for today’s policymakers. In the United States, devaluation and monetary expansion were the key sources of the recovery. Real interest rates plummeted in 1933, and the first types of spending to recover were those typically thought to be sensitive to interest rates, such as automobiles and investment goods. The New Deal’s fiscal policy elements, she said, had only a small direct effect on spending and output. That monetary expansion worked effectively in the 1930s—a time when deflation was rampant and nominal interest rates were near zero—may suggest a note of hope for modern economies facing prolonged recession and deflation. Monetary expansion, when coupled with concrete changes in the policy regime, appears to be able to generate expected inflation and lower real interest rates even in severely depressed economies.
The experience of the 1930s showed that the gold standard both spread the downturn more widely and prolonged it. In fact, countries that relinquished the gold standard early (Argentina, for example) experienced less deflation and recovered earlier than countries that remained on the gold standard until the bitter end (United States, 1933; France, 1935). Once free of the gold constraint, countries were able to devalue, which allowed them to generate more exports and run more expansionary monetary policies. This lesson has resonance for today’s economies as well, Romer said. A system of rigidly fixed exchange rates can be destructive, particularly in the presence of large external shocks.
But simply choosing a more flexible exchange rate regime may not resolve or prevent a crisis. Good policies are critical, too. In the United States in the late 1920s and early 1930s, policymakers did not understand how the economy worked and essentially relied on the wrong model. The choices they made, Romer argued, created a huge and avoidable monetary contraction. It was truly, she concluded, a colossal policy mistake whose effects were felt around the world.
Julio R. Prego
The IMF Survey (ISSN 0047-083X) is published in English, French, and Spanish by the IMF 23 times a year, plus an annual Supplement on the IMF and an annual index. Opinions and materials in the IMF Survey do not necessarily reflect official views of the IMF. Any maps used are for the convenience of readers, based on National Geographic’s Atlas of the World, Sixth Edition; the denominations used and the boundaries shown do not imply any judgment by the IMF on the legal status of any territory or any endorsement or acceptance of such boundaries. Text from the IMF Survey may be reprinted, with due credit given, but photographs and illustrations cannot be reproduced in any form. Address editorial correspondence to Current Publications Division, Room IS7-1100, IMF, Washington, DC 20431 U.S.A. Tel.: (202) 6238585; or e-mail any comments to