Petroleum and natural gas are the two fuels most commonly consumed by the industrial sector. In 2002, they accounted for about 44 percent and 40 percent, respectively, of the sector’s primary energy use. While petroleum use in the industrial sector increased by some 24 percent from 1985 to 2002, coal consumption dropped by 27 percent. The use of renewable energy has exhibited a fluctuating pattern over the years, totaling about 1.4 quads in 1978, rising to about 1.9 quads in 1985, and then retreating to about 1.68 quads in 2002. Energy use in the manufacturing industries continues to be significantly higher than in the non-manufacturing industries, which include agriculture, forestry, fisheries, mining, and construction.
Between 1985 and 2004, real GDP in the U.S. industrial sector increased by nearly 45 percent, while total energy use was virtually unchanged; this led to a decrease in energy intensity by nearly a third. However, this apparent improvement in energy intensity was due primarily to a change in the mix of products manufactured in the United States rather than to improvements in energy efficiency. The share of industrial GDP accounted for by such energy-intensive industries as petroleum refining and paper manufacturing had declined and was replaced by relatively non-energy-intensive sectors such as computers and electronics. In general, industries in most industrialized countries are more energy-efficient than their counterparts in the United States.23 The differences in energy use among countries stem from multiple sources, including natural resource endowments, energy-pricing policies, and the average age of industrial plants.
This section briefly reviews two studies that attempted to assess the potential for cost-effective energy efficiency improvements across the U.S. industrial sector. Many other studies of energy efficiency potential either examine individual industrial subsectors, such as aluminum, chemicals, or paper manufacturing, or they focus on the potential impacts of specific technologies (such as membranes or combined heat and power) or a family of technologies (e.g., sensors and controls or fabrication and materials). Such cross-sectional studies are also reviewed here