Essays on the Great Depression
By Ben S. Bernanke
Princeton, N. J.: Princeton University Press, 2000, 310 pp.
Reviewed for the Cato Journal by Anna J. Schwartz, National Bureau of Economic Research
Economists' fascination with the description and interpretation of the severity and duration of the Great Depression has not flagged in the seven decades since its onset. What began as an investigation centering on the U.S. Depression experience has broadened in recent years to cover the experience of countries around the world.
At least five of the nine essays that Bernanke has collected in this volume (some with coauthors, and only one not previously published) analyze quantitative data for a sample of countries in addition to the United States. He is a careful econometrician, scrupulous in his discussion of weaknesses in the data and problems in the econometrics. His interpretation of his findings is consonant with the current consensus views on the Depression that the economics profession espouses.
Since the 1930s, economists have pursued at least three different directions in research. At first, "overproduction" of some industrial output or a decline in a real sector of the economy (residential housing construction, business fixed investment, or real consumption expenditures) was emphasized as the source of the Depression. Since the 1960s, the emphasis has shifted to monetary shocks as the source of the downturn. Finally, beginning in the 1980s, a select group of proponents of real business cycles have favored technology shocks as the source of downturns in general.
Research in the first direction has been largely superseded by the shift to the monetary shocks approach of the second direction, which Bernanke embraces. He rejects the technology shocks approach of the third direction, discussed in an essay in the final part of the book that examines the evidence for a sample of U.S. manufacturing industries from 1923 to 1939.
The book is organized in three parts. Part one provides an overview of the macroeconomics of the Depression that is found in Bernanke's 1995 Journal of Money, Banking, and Credit lecture. Three essays in the second part deal with money and financial markets. Of the five essays in the third part, the first four deal with labor markets by U.S. industry, while the final essay deals with international data.
There are two key questions about the Depression: What caused the collapse of nominal values--prices, nominal incomes, nominal interest rates, and nominal returns on safe assets? And how was the nominal collapse transmitted to the real economy?
Part two of the book discusses the overwhelming evidence that a contraction in money supplies in all countries at the outset of the Depression was the main factor in producing a sharp decline in demand for goods and services and in their prices. The international data confirm that adherence to the gold standard explains why declines in demand occurred simultaneously in so many countries. From 1931 on, however, the data reveal a sharp divergence between countries that remained on the gold standard and countries that abandoned it, with the latter free to adopt expansionary monetary policies that cut short the Depression and the former experiencing further monetary contraction and deflation.
Bernanke distinguishes at least two channels by which deflation induced depression. A nonmonetary channel that, in his view, linked falling prices and falling output was banking panics and financial crises in choking off normal flows of credit. The decline in financial intermediation that followed from the reduction in banks' ability to lend engendered a fall in the net worth of households and firms holding nominally fixed debt. The ensuing debt crisis increased the number of bankruptcies and became an important propagator of economic contraction. The second channel, to which Bernanke devotes the third part of the book, is the way in which labor markets functioned.
The final essay in that third part, based on data for 22 countries from 1929 to 1936, assesses empirically whether slow adjustment of nominal wages was an important factor in the Depression. The principal finding is that there was a strong inverse relationship between output and real wages during the 1930s across time and across countries. The failure of wages and other costs to fall along with prices contributed to the rise in unemployment and the decline in sales. Employers chose to cut hours of work and the number employed instead of cutting wages. Nevertheless, Bernanke attributes the persistence of unemployment not to "glacially slow wage adjustment" but to the repeated negative shocks to aggregate demand that drove economies far from the path of full employment.
One essay in part three looks beyond the sharp contraction of U.S. manufacturing employment from 1929 to 1933 to the recovery in manufacturing employment in 1933-1937 and 1938-1940, and compares American experience in the 1930s with the European unemployment problem in the 1980s. Bernanke finds self-correcting tendencies of the 1930s economy stronger than is generally acknowledged. It was not a "low-level trap" economy. In his view, the New Deal by ending deflation and rehabilitating the financial system paved the way for a natural recovery rather than serving as the engine of recovery itself. This contrasts with Europe in the 1980s, where unemployment stagnated, yet shocks were absent. Bernanke sees no conceivable parallels between the United States in the 1930s and Europe in the 1980s. Reflation in the 1930s worked for the United States, but inflation would not have solved Europe's unemployment. Higher real wages had positive effects in the 1930s on productivity and aggregate demand, but higher real wages would not have had comparable effects in the European case.
As a final comment on the book, one may ask what consensus views of the economics profession signify. Consensus clearly is not synonymous with truth. For many years the prevailing research strategy was based on the Keynesian model. The monetarist approach that challenged the Keynesian model was fiercely opposed, despite the evidence in its favor, when it was first proposed. Bernanke reports but does not associate himself with a claim that has gained wide acceptance by economists before the evidence for or against it has been examined. That claim (pp. 77-78, 104, n. 23, 153) is that the Federal Reserve could not have mitigated the Depression because the United States would have been forced off the gold standard had the Federal Reserve pursued expansionary monetary policy. When a leading economist exposits such a claim, many adopt it in deference to his insight, with no evidence, in the kind of herd behavior that economists attribute to investors.