Abstract:
The primary metals industry is defined by the Bureau of the Census as the Standard Industrial Classification (SIC) 33. In terms of both absolute amount of energy use and energy intensiveness, SIC-33 is large relative to other industries.' This paper estimates essential energy demand relationships for this important sector. Although there are basically four energy sources used in the primary metals industry (fuel oil, natural gas, coal, and electricity), the response to changes in own price and cross price is dampened as a consequence of technology, plant vintage, and historical growth in the industry. Forecasts of future energy demand hinge on these price elasticities and on assumptions about future energy prices.

Abstract:
We propose a novel approach to analyze interfuel substitution that explicitly incorporates heterogenous fuel-using capital stocks in the estimation of the optimal fuel choice. Our econometric framework structurally estimates the frictionless level of fuel-using capital stocks and employs non-parametric analysis to reveal information on the form of adjustment costs facing firms. To illustrate this approach we use a large panel of Irish manufacturing firms over the period 2004-2009. The econometric estimates show a large variation in the optimal response of capital to changing fuel prices across different fuel-using technologies and imply substantial costs to capital adjustment. These results underscore the significance of the frequently ignored link between capital adjustment and the choice of fuels used by manufacturing firms.