Created 12/12/1998
Go to
Brad De Long's Home Page

Teaching | Writing | Career | Politics | Book Reviews | Information Economy | Economists | Multimedia | Students | Fine Print | Other | My Jobs

Feedback is always very welcome...

Introduction to the Symposium on Business Cycles

J. Bradford DeLong
U.C. Berkeley

Every prolonged economic expansion in the United States generates strong intellectual currents claiming that the boom-bust business cycle is over, that there is a "new economy." The expansion of the 1920s led economists to hope that the new Federal Reserve had learned how to stabilize output--that the decade truly did see a "New Era"--and thus to Irving Fisher's infamous claim on the eve of the 1929 crash that stock prices had reached a "permanent and high plateau" (Galbraith, 195?). The expansion of the 1960s led the Department of Commerce to change the name of its Business Cycle Digest to the Business Conditions Digest, for the New Dimensions of Political Economy (Heller, 1965) opened up by the Keynesian Revolution had led to substantial "Progress Toward Economic Stability" (Burns, 1959). Only the long expansion of the 1980s--under the shadow of the just-past Volcker disinflation and recession--failed to generate a large and vocal group willing to declare that traditional macroeconomic patterns were at an end.

The expansion of the 1990s has generated yet another set of "new economy" advocates. And the expansion of the 1990s has been very impressive along some dimensions--yet less impressive on others. The expansion has been long, and has attained a remarkably low level of unemployment without inflation. Yet increases in real wages have been disappointingly low, as have increases in measured productivity.


Always--or, at least, always up until the time of this writing--the claims of a "new economy" that blossom forth near the end of a prolonged expansion have proven wrong. Expansions do end. They are followed by recessions in which real, nominal, and financial variables follow patterns that bear a very close family resemblance to the prototype of recession outlined in Mitchell and Burns (19??). And either a look abroad at Japan, the NICs of East Asia, and Mexico or a look at the sudden surprise increase in asset risk premia in the U.S. last summer confirms that the business cycle mechanisms that Mitchell and Burns (19??) pointed to are still active.

From one perspective, it is surprising that predictions of a "new economy" as far as the business cycle is concerned have never (or have not yet) come true. Structural changes over the past century have been immense, carrying us from a largely agricultural economy with poor communications, small firms, and relatively unproductive and rigid technologies to today's largely service economy with extraordinary access to information and giant firms. Yet the phenomenon of the business cycle persists, with at least qualitative continuity in its mechanisms and effects. How can this be?

In this symposium, Christina Romer demonstrates that the qualitative continuity in business cycle mechanisms over the past century has been accompanied, in the United States at least, by a quantitative continuity as well: business cycles today are--measured as a proportion of total economic activity--about as large as they were a century ago. Whether they are exactly as large depends on the exact metric (Romer concludes that the standard deviation of year-to-year unemployment rate changes has shrunk by 7 percent, and that the fraction of the time that the economy spends in recession has shrunk by 32 percent). But triumphalist claims that the post-WWII business cycle is but a pale shadow of the pre-Depression phenomenon (like those made in DeLong and Summers, 1986) cannot stand.

Romer, however, does not argue that there have been no changes in the business cycle. The coming of automatic stabilizers and the rise of central banks have allowed monetary policy to offset many of the kinds of shocks that generated pre-Depression business cycles. The absence of significant stabilization springs from the fact that the rise of monetary policy has created a new class of shocks: recessions deliberately induced by monetary authorities to curb rising inflation.

Susanto Basu and Alan Taylor turn their attention not to the absolute size but to the mechanisms of business cycles, arguing that we should be able to learn a lot about which models of business cycles are potentially useful by turning theories "loose on perhaps the greatest macroeconomic laboratory available: the extant record of macroeconomic historical statistics..." for "a robust and useful theory of business cycles should be able to account for the patterns seen in the long-run data for many countries..."

They conclude that business cycle models that do not put monetary economics at the center of analysis are grossly inconsistent with the evidence on the behavior of real exchange rates, and that the comparative pattern of national recoveries from the Great Depression cannot be understood without placing prices that are sticky--either because of Lucas (1974) imperfect-information or other considerations--at the center of the analysis as well. But then they throw up their hands at the prospect that any of our current models of business cycles will ever be made fully satisfactory: the cyclical behavior of the labor market is "an embarrassment for... business-cycle theories" if we maintain the hypothesis that the labor market is a market--in some sort of supply-demand equilibrium.

Victor Zarnowitz presents a different approach: the over-investment approach according to which each boom contains within it the seeds of the subsequent recession, and each recession contains within it the seeds of the subsequent boom. Observers of business cycles have long felt that this approach contains profound truth--yet it has never been well-integrated into old Keynesian, new Keynesian, monetarist, or new classical business cycle theories. Just what is it about the structure of capitalist market economies that generates a complex root in the stochastic differential equation for real economic activity? My assessment at least is that economists will not be able to claim that they fully understand the business cycle until they have successfully integrated Zarnowitz's approach--which is Wesley C. Mitchell's approach as well--with that of other, currently more popular approaches


Feedback is always very welcome...

Professor of Economics J. Bradford DeLong, 601 Evans
University of California at Berkeley; Berkeley, CA 94720-3880
(510) 643-4027 phone (510) 642-6615 fax

This document:

Search This Website