The energy crisis of 1973–1974 coincided with a dramatic decline in U.S. stock market capitalization. Real energy prices jumped by 80 percent from 1973 to 1974. At the same time, the market value of nonfarm, nonfinancial corporations plunged by 40 percent. Because the two events coincided, the energy price hike is often considered a potential explanation for the stock market decline. One of the leading advocates of a causal link, Martin Neal Baily (1981), holds that the jump in energy prices made a substantial fraction of the capital stock obsolete. Energy inefficient machines were shut down, and expected profits of machines in operation declined. The value of existing capital decreased because it was not technologically suited to new economic conditions. This link between rising energy prices and capital obsolescence may explain the low level of stock market prices during that period. Despite the link, there has been no modern general-equilibrium evaluation of the extent to which the energy price shock was responsible for the dramatic drop in the market value of firms in 1974. I construct a dynamic generalequilibrium model with production and capital accumulation to examine the magnitude of the energy price effect. Contrary to the conventional wisdom, I find that an 80-percent increase in the real energy price causes the stock market value to decline by only 2 percent. Labor compensation, not claims to the capital stock, bears the brunt of the energy cost increase. The key element of the model is a putty-clay production technology. The neoclassical production function allows for smooth substitutability between factors after installation and conversion of capital to consumption goods at little cost. By contrast, the putty-clay production technology features ex ante substitutability of production factors, while there is no substitutability across them after machines are installed. Melvyn A. Fuss (1977) provides empirical evidence supporting the notion that capital and energy are complementary in the short run and substitutable in the long run. Because the technology is embodied in the capital stock in a putty-clay framework, changes in factor prices cause capital obsolescence and a decline in the value of capital. As a result, the putty-clay model is particularly suitable for studying the hypothesis put forward by Baily (1981). This paper adapts the putty-clay model developed by Simon Gilchrist and John C. Williams (2000) to include energy as a factor of production. I take the production technology ex ante to be Cobb-Douglas with constant returns to scale, but for capital goods already installed, production possibilities take the Leontief form: there is no substitutability of capital, energy, and labor ex post. An energy price shock affects the market value of firms through three channels. The first channel is the endogenous depreciation of the old vintage machines from both decreases in capacity utilization and declines in expected profits; the second channel is the effect of the energy price shock on investment; and the third channel is the effect on the interest rate. The impact of the fundamental shocks on the securities market depends on the resulting movement of price variables, such as the wage and the interest rate. Only a full general-equilibrium model can sort out all the interactions. * Department of Economics, CB#3305 University of North Carolina at Chapel Hill, Chapel Hill, NC 27599 (e-mail: cdwei@unc.edu). This paper is based on my Ph.D. dissertation (2001) at Stanford University. I owe a substantial debt to Robert Hall and Thomas Sargent for their invaluable advice and encouragement. Thanks also to Victor Chernozhukov, Tim Cogley, Simon Gilchrist, Kenneth Judd, Hanno Lustig, Sergei Morozov, Beatrix Paal, Michael Salemi, Stijn Van Nieuwerburgh, the Sargent group members, two anonymous referees, and numerous seminar participants. The financial support of the Dissertation Fellowship from the John M. Olin Foundation is gratefully acknowledged. Any errors are my responsibility.
Stored only when both papers are in the corpus — 237 citations exist upstream.
Imported from OpenAlex (CC0) — nothing here is generated.