Industrial Energy Management
The industrial sector is made up of a wide variety of manufacturing and other industries that use energy to extract, refine, and process raw materials to produce a variety of goods. The nomenclature used to define industrial categories varies from country to country. In the United States, the Standard Industrial Classification (SIC) system is used to designate industry groups at different levels of aggregation. Table 22.1 shows the major elements of the U.S. industrial sector. The heterogeneity of this sector makes an analysis of energy use and potential savings substantially more difficult than for other sectors of the economy contributing to greenhouse gas emissions.
Table 22.1 also shows the direct consumption of energy by the U.S. industrial sector (classified by SIC code) for 1986 which amounts to 21.7 quads.1 Adding the 6.4 quads lost in the generation and distribution of electricity brings the total sector share to 28.1 quads of total primary energy use. This total consumption amounts to 36 percent of all primary energy used in the United States (Oak Ridge National Laboratory, 1989). Six industry groups account for more than 70 percent of total primary industrial energy use. These major groups, and their corresponding percentage of total industrial energy consumption, are chemicals (21 percent); petroleum refining (19 percent); primary metals (14 percent); pulp and paper (8 percent); stone, clay, and glass (4 percent); and food and kindred products (4 percent). Purchased fuel oil and natural gas each account for roughly one-third of the total direct energy used by industry. Electricity, coal, and other energy sources (primarily wood) account for the remainder.
International studies show that the industrial sector accounts for the largest component of most nations' energy use, averaging nearly 43 percent of total primary energy consumed in the Organization for Economic Cooperation
and Development (OECD) countries in 1985. The percentage is even higher in developing countries, where industrial energy use accounts for nearly 60 percent of all energy consumption. Reported values for the former Soviet Union are slightly under 50 percent (Lashof and Tirpak, 1991).
Several recent studies have documented dramatic decreases in the energy intensity of the U.S. manufacturing sector (Ross, 1989a; U.S. Department of Energy, 1989a; Schipper et al., 1990). Figure 22.1 shows the long-term (27-year) trend in the aggregate energy intensity of U.S. manufacturing (i.e., the ratio of energy used per unit of production as defined by the Bureau of Labor Statistics). In the last 20 years, the average energy intensity of U.S. manufacturing (which does not including the mining, agriculture, and construction industries) has decreased by nearly 40 percent. Most of this decline was due to decreased direct use of fossil fuels, whose intensity fell by 50 percent between 1971 and 1985 (Figure 22.2). Electricity intensity, on the other hand, declined only slightly during that same period. The total levels of manufacturing energy use from 1958 to 1985 are shown in Figure 22.3. 2
Two major factors contributed to the trends shown in Figures 22.1 through 22.3. One was a structural shift in the economy, resulting in lower demands for energy-intensive products such as steel, aluminum, and paper. The other factor was improvements in the efficiency of manufacturing processes, resulting in less energy needed per unit of production (Boyd et al., 1987). The relative importance of structural changes and energy efficiency improvements has been analyzed by the U.S. Department of Energy (1989a), which attributes two-thirds of the overall change since 1970 to energy efficiency improvements. A more recent study by Schipper et al. (1990) draws similar conclusions using other measures of industrial production to examine the effects of structural and efficiency changes.
Effects of Structural Changes
Figure 22.4 shows the impact of structural changes alone on manufacturing energy use (i.e., the energy use that would have occurred if the products and energy intensities of each industry group had remained constant at their 1973 levels while the proportion of manufacturing sector output produced
by each industry group followed its actual historical pattern). By this measure, energy use between 1973 and 1985 would have declined by 18 percent due to structural changes alone. This analysis found that the largest contributor to the structural change was a decline in output from coal-intensive industries, particularly iron and steel (Schipper et al., 1990). The longevity of structural changes in industrial output remains somewhat more speculative. Several studies suggest a decline in the per capita consumption of energy-intensive products as industrial countries attain higher levels of affluence (e.g., Williams et al., 1987; Lashof and Tirpak, 1991). This ''saturation" phenomenon implies a long-term reduction in the demand for energy-intensive materials, which could (if sustained) have important implications for future industrial energy demand.
Other factors believed to have contributed to the changing structure of U.S. manufacturing include higher oil prices and economic policies that affect the competitiveness of U.S. goods. The extent to which structural changes will continue to affect the total demand for energy in the industrial sector depends also on the absolute growth rate of manufactured products. There is some controversy over whether manufacturing represents a constant or declining share of real U.S. gross national product (GNP). To the extent that manufacturing output is coupled to GNP, decreases in energy intensity due to structural shifts may be offset at least partially by a higher total demand for manufactured goods as GNP continues to rise. Recent studies performed for DOE, however, assume a decline in the future manufacturing share of GNP (Ross, 1989b).
Effects of Efficiency Improvements
Recent improvements in energy efficiency for the U.S. manufacturing sector have been analyzed by DOE for the period from 1980 to 1985 (U.S. Department of Energy, 1990) and by Schipper et al. (1990) for the 29-year period from 1958 to 1987. The latter study attempted to develop an indicator of aggregate energy efficiency improvements independent of the structural changes noted earlier. The output of each industry group was assumed to remain constant at its 1973 value, whereas its energy intensity followed the actual historical path. By this measure, the structure-adjusted energy intensity of the manufacturing sector as a whole declined by 2.5 percent per year from 1958 to 1973 and by 2.7 percent per year from 1973 to 1985. The aggregate changes for 1958 to 1985 (Figure 22.5) were 15, 37, and 44 percent for coal, gas, and oil, respectively, and 6 percent each for wood and electricity intensity. For the period from 1985 to 1987, aggregate energy intensity continued to fall by roughly 2 percent per year. Overall, the structure-adjusted reduction in energy intensity between 1973 and 1987 was estimated at approximately 33 percent, due primarily to a reduction in direct fossil fuel use (Schipper et al., 1990).
A DOE industry-by-industry analysis for 1980 through 1985 supports the conclusion of a continuing trend toward greater energy efficiency in the industrial sector (Table 22.2). Although the energy price shocks of the 1970s undoubtedly contributed to improvements in energy efficiency, the more significant driving force for energy improvements in the industrial sector appears to have been the long-term changes in basic process technology, which reduce overall production costs as well as energy costs. Thus, even at the relatively low energy prices prior to 1973 and since 1980, manufacturing processes have become increasingly energy efficient. Although energy prices certainly have affected the fuel mix in the industrial sector (e.g., oil and gas use fell significantly in response to price increases of the 1970s), the sustained improvements in energy efficiency indicate that the industrial sector is not merely substituting one fuel for another (e.g., electricity for oil and gas). Rather, real reductions in energy intensity are being achieved through conservation measures and process technology innovations. The outlook is that this trend will be sustained through the turn of the century.
TABLE 22.2 Energy Efficiency Changes in Manufacturing Industry Groups, 1980 through 1985
Emission Control Methods
Carbon dioxide emissions from the industrial sector are due primarily to the combustion of fossil fuels for heat and power: CO2 is formed from the primary fuels used in boilers and process heaters and from the use of fuel by-products such as coke, petroleum plant gas, and coke oven gas. Additional emissions occur from industrial processes such as calcining and cement manufacture, which give off CO2 when raw minerals (e.g., limestone) are heated.
The use of purchased electricity contributes to CO2 emissions indirectly from the combustion of fossil fuels at power plants. Thus the magnitude of electricity-related CO2 emissions depends on the utility fuel mix in a particular region. Because many energy-intensive industries tend to be located in regions of (historically) cheap electricity (e.g., hydroelectric power), the national average fuel mix may not necessarily be a good indicator of potential CO2 emission reductions from reduced electricity demand.
Because of the extraordinary heterogeneity of the industrial sector, this chapter makes no attempt to discuss specific technological measures for reducing energy use in each of the major industries. Excellent summaries of energy use technologies on an industry-by-industry basis are available elsewhere (e.g., Decision Analysis Corporation, 1990). The intent here is to outline more generally the categories of methods available to industry and to estimateinsofar as possiblethe magnitude of CO2 emission reductions achievable using current technology. Toward this end, 11 general mechanisms for reducing CO2 emissions in the manufacturing sector have been identified by Ross (1990a), who estimated qualitatively their overall potential and limitations (Table 22.3). These measures can be aggregated into four general categories: (1) fuel and energy switching, (2) energy conservation measures, (3) process design changes (including recycling), and (4) macroeconomic structural changes.
Fuel and Energy Switching
Fuel and energy switching measures reduce CO2 emissions by substituting fuels with less carbon per unit of energy for those fuel and energy forms currently in use. For example, switching from coal to natural gas reduces CO2 emissions by approximately 40 percent (for the same energy use), and substituting oil for coal lowers CO2 by roughly 20 percent. The potential for fuel substitution is limited by the technical and economic circumstances of different industries. For example, the largest use of coal occurs in the iron and steel industry for the production of coke, which is used in blast furnaces to produce iron. This use of coal cannot be eliminated by simple fuel substitution. Similarly, when fuel such as coal and oil are used in
TABLE 22.3 The CO2 Reduction Mechanisms in ManufacturingTechnological Opportunities and Constraints
remote locations, such as in the forest products industry, substitution of natural gas (requiring a pipeline) is unlikely to be feasible.
The principal opportunities for fuel substitution lie in industrial boiler applications, particularly in industries that switched more heavily to coal during the 1970s in response to fuel price and regulatory pressures. The price, availability, and reliability of alternate fuel supplies are the key issues in all circumstances.
The technical potential for short-term fuel substitution at existing facilities has been estimated by DOE on the basis of a recent survey of manufacturers (U.S. Department of Energy, 1988). Results are shown in Figure 22.6, which displays the maximum, minimum, and actual 1985 fuel usage for the manufacturing sector. Actual 1985 coal use is seen to be near its maximum technical potential, whereas distillate and residual oil use were near their technical minima. A rough estimate of the potential reduction in CO2 emissions from fuel switching at existing facilities can be obtained by assuming that 0.6 quad of current coal use (the difference between actual and minimum use) is displaced by either oil or natural gas. Based on the average carbon content of fossil fuels, this would yield a CO2 reduction of
24 Mt using gas or 13 Mt using oil.3 The cost-effectiveness of this reduction would depend on oil and gas prices relative to coal prices. Figure 22.7 shows that for recent price premiums of about $1.5 to $2/MBtu for oil and about $1.5/MBtu for gas, the reduction in CO2 would cost about $40 to $80/t CO2.
The longer-term technical potential for fuel switching is, of course, much greater than the short-term estimates based on facilities currently in place. The magnitude of such changes, however, is dependent on future changes in fuel prices, process technology, and the turnover of capital stock. Uncertainty over the long-term availability and price of gas and oil can be expected to inhibit fuel conversions, even where technologically feasible.
Emission reduction measures can also include switching from fossil fuels to electricity, which already is occurring to some degree. This may or may not reduce CO2 emissions, depending on (1) the fuels used for power generation, (2) the fuel for which electricity is being substituted, and (3) the relative efficiencies of the current and substitute processes. Because substituting electricity for fossil fuels entails new capital costs, the much higher price of electricity relative to fossil fuels requires that the electricity-based process be roughly 3 times more efficient than the fossil fuel system to be competitive. Such opportunities, however, do exist (Ross, 1989b).
Switching from fossil fuels to biomass fuels (i.e., waste or by-products from the food processing and forest products industries) is another option that is technically feasible and has already been implemented in certain industries, particularly paper manufacturing. The remaining opportunities appear to be relatively small, barring more aggressive programs to utilize existing forest and field residues, to increase crop production for fuel use, or to improve the current efficiency of biomass combustion (Ross, 1990a).
Energy Conservation Measures
Energy conservation and efficiency measures reduce the CO2 emissions associated with the fuel or energy source that is conserved. Opportunities here range from "housekeeping" improvements, which conserve relatively small amounts of energy at low to negligible cost to more substantial measures requiring much higher capital investments but with potentially larger energy cost savings. A number of recent studies have addressed potential energy conservation measures for the industrial sector (e.g., Oak Ridge National Laboratory, 1989; Ayres, 1990; Decision Analysis Corporation, 1990).
Although many conservation measures tend to be industry- or process-specific, a number of "generic" measures, including more efficient lighting, the use or more efficient variable-speed motors, and the more efficient recovery of waste heat, also have been identified. Table 22.4 shows one example from a European study of how energy efficiency can be improved in the use of electricity for motive power, the largest use of electricity in industry (International Energy Agency, 1989). The opportunity for energy efficiency improvements lies principally in single-phase and three-phase
TABLE 22.4 Breakdown of Electricity Consumption for Motive Power and Possible Efficiency Improvements (example: Germany)
AC motors and associated drive trains. The inability of AC drives to operate at the variable speeds required by many industrial processes (in applications such as fans, pumps, blowers, and conveyors) can result in energy losses of 40 to 80 percent using conventional means of speed control. New electronic control systems, coupled with improved motor efficiencies (from higher quality magnetic materials and other loss-reducing measures), and more attention to correct motor sizing, can yield significant improvements in overall energy efficiency.
In principle, a conservation supply curve (CSC) could be developed on a process-by-process or industry-by-industry basis showing the amount of energy savings obtainable at different energy prices (analogous to supply curves for primary fuels). The functional and sectoral categories of information needed for constructing CSCs needed for the manufacturing sector are indicated in Table 22.5 (Ross, 1990b). However, the present ability to quantify energy conservation potential in this matter remains extremely limited.
Unlike the buildings and transportation sectors, where energy use is concentrated in a relatively small number of processes that have been extensively studied, the enormous diversity and variability of industrial processes remain to be characterized to a similar extent. In particular, there is little publicly available information on the costs of energy conservation measures in the industrial sector. In part this is due to the often proprietary nature of such information. Difficulties also are encountered in ascribing costs to many of the energy savings that accrue from large-scale changes in production methods or plant operation (e.g., converting from ingot casting to continuous casting in steel manufacturing). The "bottom line" is that the types of conservation curves outlined in Table 22.5 are simply not available today to quantify energy and CO2 reduction potential in a manner analogous to the residential, commercial, and transportation sector analyses. Rather, the estimates presented in this chapter derive from a limited number of studies of key U.S. industries.
The most widely studied area in the industrial sector has been the potential for conserving electricity use. For the United States, Ross (1990a,b) has developed a conservation supply curve for electricity use in the U.S. manufacturing sector based on an aggregation of results from several key process industries. Details are described in Appendix D. Figure 22.8 is an electricity conservation supply curve showing that with current technology, a reduction in energy intensity of up to 30 percent could be achieved in the use of electricity by the manufacturing sector, yielding a savings of about 200 billion kilowatt-hours (BkWh) of electricity based on current energy use. The corresponding reduction in CO2 emissions would be about 140
TABLE 22.5 Categories for Constructing Conservation Supply Curves for Manufacturing
Mt/yr based on the current national average utility fuel mix. Comparable maximum technical savings, in the range of 24 to 38 percent, have been estimated by the Electric Power Research Institute (EPRI) for the year 2000 (Barakat and Chamberlin, Inc., 1990; Electric Power Research Institute, 1990). Details of the EPRI analysis are shown in Appendix D. Note that in both the EPRI and the Ross studies the bulk of the savings comes from the use of more efficient motors, electrical drive systems, and lighting, which constitute the primary uses of electricity in industry.
Empirical evidence suggests that for the industrial sector, discount rates of 20 to 30 percent or more, corresponding to payback periods of less than 3 years, are required to stimulate investment in energy efficiency improvements (Ayres, 1990; Ross, 1989a, 1990b). Figure 22.8 shows that for such criteria, electricity prices would have to triple to induce a 30 percent savings in overall electricity use. On the other hand, lower real discount rates typical of public funding projects and large-scale utility investments (i.e., 3 to 10 percent) would substantially lower the annualized cost of conservation investments. As is discussed in Appendix D, at these lower discount rates, electricity conservation investments would yield a net cost savings analogous to those seen earlier for the buildings sector (Chapter 21). For a 6 percent real discount rate, for example, the net cost of conservation (i.e., the investment cost less the electricity savings) for a maximum savings of 200 BkWh is about -$20/t CO2, compared to $100/t for a 30 percent discount
rate. The costs shown in Figure 22.8 are similar to those reported by EPRI for a comparable rate of return (see Appendix D). Policy measures for stimulating investments in energy conservation are discussed later in this chapter and again in Chapter 29.
Another energy-conserving measure applicable to the industrial sector is the use of co-generation to produce heat and power simultaneously. Applications of co-generation to industrial processes have been practiced for many years and are discussed extensively in the literature. Economical applications of this technology require fairly steady heat loads, which are not available in many industrial processes that operate in a more cyclical fashion. Thus co-generation tends to be found primarily in the chemicals, paper, and petroleum refining industries. Nevertheless, opportunities do exist to utilize co-generation more extensively, for example, by using newer gas turbine technology that better matches typical industrial energy demands and provides larger overall fuel savings (Williams et al., 1987).
New financial incentives for co-generation also now exist as a result of the 1978 Public Utility Regulatory Policies Act (PURPA), which requires utilities to purchase independently generated electricity that is available below the utility's own avoided cost. However, studies of the chemical and paper industries, where co-generation is most prevalent, suggest that the absolute price of purchased electricity, rather than the potential revenue from the sales of the electricity generated through co-generation, is the primary driving force for this technology (Ross, 1989b).
Largely as a result of PURPA, co-generation capacity has expanded rapidly in the past decade, principally in California and Texas. It is expected that this trend will continue. The California Energy Commission (CEC) and the Northwest Power Planning Council are among the leading regulatory agencies who have analyzed the potential for increased co-generation on a regional basis (Bonneville Power Administration, 1989; California Energy Commission, 1990; Northwest Power Planning Council, 1990). The CEC, for example, estimates that co-generation capacity in California could more than double over the next 10 years (California Energy Commission, 1990). A national estimate of future industrial co-generation potential has been developed by RCG/Hagler, Bailly, Inc. (1991), as shown in Table 22.6. Here, too, a doubling of current industrial capacity from 22 GW to 47 GW is projected in the next decade. This result is roughly consistent with the amount of new co-generation capacity estimated by DOE in 1984 based on the replacement of aging industrial boilers (Department of Energy, 1984).
The CO2 implication of increased industrial co-generation principally is a reduction in emissions from the combustion of natural gas, which is the
TABLE 22.6 Projected Industrial Co-generation Market Development
fuel of choice for such installations (which range in size from a few megawatts to over 100 MW). Co-generation units typically employ either a boiler/steam turbine, combined cycle, or gas turbine to achieve higher overall energy efficiency in comparison with to the separate generation of steam and electricity (Larsen and Williams, 1985). Where co-generation displaces electricity purchased from a utility, the CO2 implications depend on the fuel mix that obtains. In many cases, industrial demand is met by peak load electricity also generated by natural gas. For purposes of this study, a rough estimate of the CO2 reduction potential of co-generation assumes that natural gas used by industrial and utilities to generate steam and electricity is instead utilized in co-generation facilities to supply energy demands more efficiently. Based on recent studies of future co-generation potential, an additional 25 GW of industrial capacity is assumed. These savings will be available only as long as the substitution of natural gas for other energy sources is possible. Appendix D summarizes the performance and economic assumptions employed.
The result is an estimated 45 Mt/yr of CO2 that could be reduced using current technology at a net cost of roughly -$20 to -$5/t CO2. This 45 Mt CO2 reduction amounts to 4 percent of current industrial sector emissions (Edmonds and Ashton, 1989). This figure is similar to an independent estimate by Ross (1990a) of a 5 percent reduction in fuel CO2 from wider
use of co-generation in the next two decades. The net negative cost reflects the same societal perspective used throughout this report, i.e., a 6 percent real discount rate to annualize the investment cost of energy savings (in this case, an estimated $1000/kW for a co-generation installation saving natural gas). In practice, of course, the economics of co-generation depends strongly on the selling price of electricity, the local tariffs for natural gas, and the actual discount rates of industrial investors. Such factors have been considered in the capacity expansion estimates of California Energy Commission (1990) and others cited earlier in this report.
Other Conservation Measures
A wide variety of other energy-saving measures have been identified in industry-specific studies that focus on the details of industrial processing. such measures range from improved waste heat recovery at chemical plants to soaking pit enhancements in steel mills to boiler efficiency improvements in the food processing industry, and so on. As noted earlier, however, a major limitation of most studies is the absence of information on the costs of energy savings. Current analyses by DOE, for example, rely heavily on regression models of energy use versus fuel price, plus independent judgments of ''autonomous" energy efficiency improvements in forecasting future industrial energy demands and their environmental implications (e.g., AES Corporation, 1990; Energy Information Administration, 1990). At the present time, the limited amount of engineering costs data compared with that available for the residential and transportation sectors precludes a rigorous analysis of the cost-effectiveness of CO2 mitigation measures for the industrial sector.
For purposes of the present study, rough estimates of the magnitude and cost of additional fuel conservation measures have been derived from a few published studies of specific industrial plants. For example, Figure 22.9 shows cost versus energy reduction curves for a petroleum refinery and an integrated steel mill (Ross, 1987; Larsen, 1990). Overall energy savings of up to nearly 30 percent are represented. To the extent these results are generally applicable to the industrial sector, CO2 reductions of up to about 350 Mt/yr may be achievable at little or no net cost using the societal discount rates assumed in this report (see Appendix D for details of this estimate).
Other recent reports also document the potential for continued energy savings in the industrial sector. For example, Nelson (1989) reports the results of an energy conservation contest held at one division of the Dow Chemical Company since 1982. To date, over 500 different projects have been identified and undertaken. As shown in Table 22.7, the average return on investment of the winning projects was nearly 200 percent, corresponding
TABLE 22.7 Summary of Energy Conservation Projects of Less Than $2 Million Each in the Annual Dow Chemical Company Louisiana Division Contest
to payback periods of much less than a year. (The minimum return on investment established by Dow was 30 percent.) While the total fraction of energy saved from these projects is proprietary information, it is nonetheless reported to be considerable (K. Nelson, Louisiana Division, Dow Chemical USA, Plaquemine, Louisiana, personal communication to E. Rubin, Carnegie-Mellon University, Pittsburgh, Pennsylvania, January 1991). More importantly, data for very recent years show that energy savings continue to be available at attractive rates of return. Anecdotal information, however, suggests that the type of contest run by Dow at this division is not widespread within the industry. More aggressive attention to energy-saving measures in the industrial sector undoubtedly could yield national energy and CO2 reductions at costs competitive with other investment opportunities.
Changes in Process Design
To the extent that new industrial process technology continues to reduce the energy intensity of manufacturing, CO2 emissions will decline. As seen earlier, this method of energy reduction appears to be one of the most significant long-term mechanisms for reducing energy use and consequent emissions. Two types of measures are discussed here: recycling and fundamental process changes.
The principal opportunities for reducing energy use and CO2 emissions through increased recycling lie in the primary metals, pulp and paper, organic chemicals, and petroleum refining industries (Ross, 1990a). Recycled materials substitute for raw materials whose processing and refining typically are the most energy-intensive phases of manufacturing. Thus significant energy savings can be achieved when the demand for raw materials is reduced. Because of impurities typically associated with recycled material, some process modifications or preprocessing steps often are necessary to utilize recycled materials. In some cases, impurities preclude a material's being reused for its original purpose (e.g., printers ink in recycled newspaper). Thus other applications of recycled material may be sought (e.g., wastepaper for insulation materials rather than newsprint). Major limitations to increased recycling at the present time include the creation of markets for postconsumer recycled material in the manufacture of higher-quality products than heretofore and the reliable and clean collection of a high fraction of selected postconsumer materials (Ross, 1989b).
More fundamental changes in process technology hold theoretically large potential for energy intensity reductions (Ross, 1987) but need to be quantified through more detailed industry-specific studies, including costs. Although the energy savings actually achievable through process innovations cannot be quantified readily, a number of potentially attractive opportunities
TABLE 22.8 Estimates of Energy Efficiency Potential by Industry, 1990 through 2020
have been identified in industry-specific studies (e.g., Oak Ridge National Laboratory, 1989; Decision Analysis Corporation, 1990). Historical trends, however, indicate that the time scales for introducing new technology are relatively long (i.e., on the order of a decade or more). The role of research and development in accelerating the pace of process innovation is discussed in the section "Research and Development Needs" below. In the long run, it is clear that technology innovation holds the key for sustainable and significant reductions in greenhouse gas emissions from the industrial sector. Table 22.8 summarizes one set of judgments on energy-efficiency potential from experts in several major industries (National Research Council, 1990). A sustained improvement of 1.5 percent per year in energy efficiency through new process technology would yield a 25 percent decrease in overall industrial energy demand shortly after the turn of the century. The corresponding CO2 reduction would be on the order of 300 Mt/yr, based on current energy use.
Macroeconomic Structural Changes
The measures discussed above reduce greenhouse gas emissions by lowering the energy requirements of a particular industrial process. In turn, the mix of processes and industries that constitutes the overall industrial sector determines the total magnitude of greenhouse gas emissions at the national level. Three mechanisms contribute to structural changes. One is materials
substitution, which alters the demand for various industrial products (e.g., the substitution of plastics for metals in automobiles or the use of ceramics in engines). Changing demands for industrial products contribute in part to the historical sector shifts shown earlier for the United States. The impact of materials substitution on greenhouse gas emissions is complex and remains largely unstudied. Policy measures such as increased emphasis on recycling could induce further changes in materials substitution patterns.
Another structural factor has been a shift in activity away from primary manufacturing of materials such as steel and aluminum toward downstream processes such as fabrication and assembly (Ross, 1989b). In terms of greenhouse gas emissions, this trend implies a reduction in CO2 emissions as a result of a shift toward less-energy-intensive processes. To some extent, however, such emission reductions may be offset by increased manufacturing and processing of raw materials in other parts of the world.
Barriers to Implementation
The most significant barriers to achieving more rapid reductions in energy intensity and CO2 emissions in the industrial sector are (1) the relatively short payback period demanded by industrial decision makers for investments in energy conservation technology and (2) the relatively long time required to replace existing capital equipment with newer processes that are more energy efficient. Because the most energy-intensive manufacturing processes are also extremely capital-intensive (e.g., chemicals, primary metals, and paper), the rate of capital turnover is typically measured in decades. Policy measures that can hasten the rate of capital turnover and the introduction of more-energy-efficient process technology can therefore speed the long-term trend in energy efficiency improvements indicated earlier. Related to this, regulatory requirements for process operating permits (i.e., as required by air and water pollution control agencies) may pose an additional barrier to the introduction of new technology. Changes in permitting procedures to minimize the "hassle" and delays in bringing new or modified processes on-line may significantly affect industrial willingness to implement process improvements.
Similarly, measures that directly or indirectly affect the payback period or discount rate used to evaluate investments in energy conservation can also influence the pace of CO2 emission reductions. Relative to other industrial nations, investment in energy conservation measures by the U.S. industrial sector is still low. Although this difference is due at least in part to lower real energy prices, the emphasis on short-term profitability that characterizes much of the U.S. industrial sector also plays a fundamental role.
The total level of production also significantly influences the industrial
sector's contribution to greenhouse gas emissions. To some extent, sectoral shifts and the total demand for products can be influenced by government policies either directly or through indirect effects on international competitiveness and the overall costs of doing business. While it is beyond the scope of this study to assess the complex interactions between market forces and public policy initiatives, their existence certainly must be recognized in considering policy measures that affect U.S. industry.
A number of policy measures that can accelerate the reduction in CO2 emissions from the industrial sector are summarized in Table 22.9. These measures include various forms of regulation, fiscal incentives, and information programs (U.S. Department of Energy, 1989b). In considering these options, it must be recognized that manufacturing processes tend to vary significantly both within and across industries and are typically proprietary in some or all of their designs. Thus direct government regulation of industrial energy use would be much less effective than price as a means of inducing energy efficiency and other measures that reduce CO2 emissions. Policy instruments likely to be most useful in managing CO2 emissions in industry thus include fuel taxes, regulatory changes to encourage co-generation and other conservation measures, and tax credits or other incentives to induce investments in research, recycling, and new process technology that uses energy more efficiently.
Fiscal measures such as a fossil fuel use tax or a carbon tax can provide a flexible and general incentive for reducing fossil fuel consumption and encourage wider use of lower-emission fuels. However, fuel taxes do not directly address the high discount rates implicit in industrial decision-making and could also cause serious dislocations in international trade unless similar taxes are adopted by other countries. Tradeable emission permits, analogous to current Clean Air Act allowances for SO2 emissions, could offer similar CO2 reduction benefits, although implementation and allocation procedures could prove much more cumbersome than for SO2. Fuel use fees or a carbon tax are preferred over direct regulation of energy use because industries differ greatly in their energy needs (U.S. Department of Energy, 1989b).
A related concept that has been proposed is to incorporate externality costs into the prices of fuel and/or raw materials. Such externalities might include the environmental damages from mining through end use. For example, any increase in virgin materials price due to externality costs would
increase the attractiveness of recycled materials. The effect on net energy consumption would depend on the magnitude of the price changes. However, actual implementation of this policy measure could be very difficult. Not only would externality costs be very difficult to determine, but the adoption of different policies in different countries could lead to widespread disruption in international trade absent any new and enforceable international accords imposing similar externality costs.
As noted earlier, the Public Utility Regulatory Policies Act (PURPA) encourages the development of electricity co-generation, i.e., the combined production of electricity and process heat. For many manufacturing processes, co-generation saves fuel and produces excess electricity that can be sold to a utility. Although some industries (e.g., petroleum refining and petrochemicals) have taken advantage of these savings, others (e.g., cement) have not. Industries that do not use co-generation could be encouraged to do so through regulatory incentives, although more information is needed to first determine why existing incentives are insufficient (U.S. Department of Energy, 1989b).
Significant potential also exists for electric utility companies to provide energy services to industry, much like the residential and commercial sector opportunities that are now receiving attention in some parts of the country (see Chapter 21). Such programs effectively lower the applicable discount rate for investments in new equipment. As discussed in Chapter 21, pioneer programs in California and Massachusetts are now providing financial incentives to electric utilities to invest in end-use technologies that are more cost-effective than building new plant capacity. Extension of such programs to other states, and to the industrial sector, could be accomplished through regulatory changes at the state level. Model programs and legislation could help guide the introduction of such changes. Indeed, cooperative efforts between utilities and industrial firms may constitute one of the most promising policies for achieving more rapid energy and CO2 reductions in the industrial sector. In principle, such arrangements could be extended to other utilities, such as regulated gas companies, where fuel savings opportunities could be targeted.
Investment Tax Credits
Another method of reducing energy consumption is to change the corporate tax code to encourage quicker capital turnover and investment in more fuel-efficient production processes. This policy would encourage the many industries with old capital equipment and low rates of capital turnover (e.g.,
TABLE 22.9 Policies for the Manufacturing Sector
(continued on page 275)
(Table 22.9 continued from page 274)
(continued on page 276)
(Table 22.9 continued from page 275)
(continued on page 277)
(Table 22.9 continued from page 276)
(continued on page 278)
(Table 22.9 continued from page 277)
steel and paper) to invest in more efficient processes. Past experience shows that encouraging investment through the tax code can have a large and rapid effect on industry investment, although there is some disagreement over the benefits of the 1981 corporate tax revisions (Hall and Jorgenson, 1967; Bosworth, 1985; Summers, 1985). While it appears that some past efforts to encourage energy conservation through federal tax credits for specific qualifying equipment did not succeed in significantly influencing investment behavior (Alliance to Save Energy, 1983), larger tax credits and more broadly defined qualifying criteria likely would get a strong response. An investment tax credit also provides certainty of a payback when firms are risk averse due to uncertainty about future energy prices, as well as a justification for investment when financial market imperfections restrict the availability of funds (U.S. Department of Energy, 1989b). The windfall implications of investment tax credits, however, also must be considered in evaluating this option.
Research and Development Needs
Increased research and development holds the potential to accelerate the application of new process technologies that reduce industrial energy requirements, with a corresponding reduction in CO2 emissions. A number of recent studies (e.g., Oak Ridge National Laboratory, 1989; Decision Analysis Corporation, 1990) have identified research and development opportunities for specific industries (e.g., chemicals, petroleum refining, steelmaking, cement, and paper), as well as general technological developments that would provide benefits across the industrial sector (e.g., improved waste heat recovery. Table 22.10 summarizes the results of one comprehensive study that evaluated some promising research and development options for reducing energy consumption (Oak Ridge National Laboratory, 1989) and also attempted to estimate some of the ancillary benefits of research and development in addition to energy-saving potential (e.g., economic competitiveness, secondary environmental impacts, energy security, social feasibility, and technology transfer to developing countries). In general, these ancillary benefits are positive.
The major conclusions that emerge with regard to CO2 mitigation measures for the industrial sector are the following:
• The industrial sector typically imposes the greatest demand for primary energy, making it (in most cases) the largest contributor to greenhouse gas emissions associated with the use of energy. For developing countries, the industrial sector accounts for up to 60 percent of the primary energy demand, compared to about 40 percent for developed countries.
TABLE 22.10 Evaluation of Promising Research and Development Options for Energy End Use
• In the United States the energy intensity of the industrial sector (i.e., the amount of energy per unit of production) has been declining steadily over the past three decades. Analyses of the manufacturing sector suggest that improvements in energy efficiency account for most (on the order of two-thirds) of the changes observed to date. The remainder are due to structural shifts that have resulted in less demand for energy-intensive products such as steel, aluminum, and paper. This trend in energy intensity reduction is expected to continue, exerting downward pressure on CO2 emission growth.
• A major factor in the long-term improvement in energy efficiency has been innovations in process technology that appear to be independent of changes in energy prices. Although the cost of energy is certainly a factor in decisions regarding fuel choice and energy consumption levels, other factors related to industrial productivity generally dominate energy considerations.
• While the potential for further energy savings in the industrial sector has been widely studied, there is relatively little information on the costs of energy reduction measures. Further study of costs is needed to analyze more rigorously the relationships between industrial energy use and greenhouse gas mitigation measures.
• Estimates of the energy savings from energy conservation investments that reduce the use of electricity for manufacturing (via more efficient motors, drives, process technology, and so on) indicate that savings up to about 25 to 30 percent are achievable with current technology. The reduction in CO2 emissions for these savings would be roughly 140 Mt/yr. However, electricity prices would have to increase by a factor of 2 to 3 at the implicit rates of return now prevalent in the industrial sector (i.e., payback periods of 3 years or less). On the other hand, for lower rates of return, typical of public sector and utility investments, significant energy savings appear achievable at a net negative cost based on current electricity prices.
• Expanded use of co-generation and other existing measures to improve fuel use efficiency at industrial plants was estimated to achieve an overall energy savings of roughly 30 to 35 percent, producing a reduction in CO2 emissions of nearly 400 Mt/yr. As with the estimated electricity savings, the implicit rates of return required to achieve these savings would have to be substantially lower than those now prevalent in the industrial sector. However, for a social discount rate of 6 percent, substantial CO2 reductions appear achievable at a net negative cost.
• Given the long-term trend in energy efficiency improvements through process technology innovation, policies that stimulate research and development, and encourage more rapid capital turnover, may offer some of the best long-term strategies for mitigating CO2 emissions from the industrial sector.
• Government policy measures or incentives that effectively lower the rate of return (increase the project payback period) could accelerate investments in energy conservation. State-level incentives by public utility commissions to encourage utilities to invest in cost-effective measures at industrial facilities, similar to programs now emerging for the residential and commercial customers, may constitute one of the most promising means of addressing the industrial sector.
1. 1 quad = 1 quadrillion (1015) British thermal units (Btu).
2. Values are in exajoules (EJ); 1 EJ = 1018 J = 1/1.054 quad = 85 bkWh of electricity.
3. Throughout this report, tons (t) are metric; 1 Mt = 1 megaton = 1 million tons; and 1 Gt = 1 gigaton = 1 billion tons.
AES Corporation. 1990. An Overview of the Fossil2 Model. Prepared for the U.S. Department of Energy. Arlington, Va.: AES Corporation, July 1990.
Alliance to Save Energy. 1983. Industrial Investment in Energy Efficiency: Opportunities, Management Practices, and Tax Incentives. Washington, D.C.: Alliance to Save Energy.
Ayres, R. U. 1990. Energy conservation in the industrial sector. In Energy and the Environment in the 21st Century. Cambridge, Mass.: MIT Press.
Barakat and Chamberlin, Inc. 1990. Efficient Electricity Use: Estimates of Maximum Energy Savings. Report No. EPRI CU-6746. Palo Alto, Calif.: Electric Power Research Institute.
Bonneville Power Administration. 1989. Assessment of Commercial and Industrial Cogeneration Potential in the Pacific Northwest, prepared by Tech Plan Associates, Inc., for Bonneville Power Administration, Portland, Oreg., March 1989.
Bosworth, B. 1985. Taxes and the investment recovery. Pp. 1–45 in Brookings Papers in Economic Activity. Washington, D.C.: The Brookings Institution.
Boyd, G., J. F. McDonald, M. Ross, and D. A. Hanson. 1987. Separating the changing composition of U.S. manufacturing production from energy efficiency improvements: A divisia index approach. The Energy Journal 8(2):77–96.
California Energy Commission. 1990. Staff Testimony on qualifying Facilities/Self-Generation Forecast. Docket No. 88-ER-8. Sacramento, Calif.: California Energy Commission.
Decision Analysis Corporation. 1990. Energy Consumption Patterns in the Manufacturing Sector. Report on Subtask 7B prepared for U.S. Department of Energy, Washington, D.C. Vienna, Va.: Decision Analysis Corporation.
Edmonds, J., and W. Ashton. 1989. A Preliminary Analysis of U.S. CO2 Emissions Reduction Potential from Energy Conservation and the Substitution of Natural
Gas for Coal in the Period to 2010. Report DOE/NBB-0085. Washington, D.C.: Office of Energy Research, U.S. Department of Energy.
Electric Power Research Institute. 1990. New push for energy efficiency. EPRI Journal 15(3):4–17.
Energy Information Administration. 1990. PC-AEO Forecasting Model for the Annual Energy Outlook. Model Documentation. Report DOE/EIA-M036(90). Washington, D.C.: Energy Information Administration, U.S. Department of Energy.
Hall, R. E., and D. W. Jorgenson. 1967. Tax policy and investment behavior. American Economic Review 57:391–414.
International Energy Agency. 1989. Improving the Efficiency of Electricity End-Use. Paris: International Energy Agency, Organization for Economic Cooperation and Development.
Lashof, D. A., and D. A. Tirpak, eds. 1991. Policy Options for Stabilizing Global Climate. Washington, D.C.: U.S. Environmental Protection Agency.
Larsen, E. D., and R. H. Williams. 1985. A Primer on the Thermodynamics and Economics of Steam-Injected Gas Turbine Cogeneration. Report PU/CEES 192. Princeton, N.J.: Princeton University.
Larsen, W. G. 1990. Energy Conservation in Petroleum Refining. Ph.D. dissertation. University of Michigan, Ann Arbor.
National Research Council. 1990. Confronting Climate Change: Strategies for Energy Research and Development. Washington, D.C.: National Academy Press.
Nelson, K. C. 1989. Are there any energy savings left? Chemical Processing (January).
Northwest Power Planning Council. 1990. Draft 1991 Northwest Conservation and Electric Power Plan, Volume II. Portland, Oreg.: Northwest Power Planning Council.
Oak Ridge National Laboratory. 1989. Energy Technology R&D: What Could Make a Difference? Report ORNL-6541/V2/P1. Oak Ridge, Tenn.: Oak Ridge National Laboratory.
RCG/Hagler, Bailly, Inc. 1991. Industrial Cogeneration Markets. RCG/Hagler, Bailly, Inc., Washington, D.C. January 29. Memorandum to E. Rubin, Carnegie-Mellon University.
Ross, M. 1987. Industrial energy conservation and the steel industry of the United States. Energy 12(10/11):1135–1152.
Ross, M. 1989a. Improving the efficiency of electricity use in manufacturing. Science 244:311–317.
Ross, M. 1989b. Energy and transportation in the United States. Annual Review of Energy 14:131–171.
Ross, M. 1990a. Modeling the energy intensity and carbon dioxide emissions in U.S. manufacturing. In Energy and the Environment in the 21st Century: Proceedings of a Conference at Massachusetts Institute of Technology. Cambridge, Mass.: MIT Press.
Ross, M. 1990b. Conservation supply curves for manufacturing. In Proceedings of the 25th Intersociety Energy Conversion Engineering Conference. New York: American Institute of Chemical Engineers.
Schipper, L., R. B. Howard, and H. Geller. 1990. United States energy use from
1973 to 1975: The impacts of improved efficiency. Annual Review of Energy 15:455–504.
Summers, L. H. 1985. Comments and discussion. Brookings Papers on Economic Activity 1:42–44.
U.S. Department of Energy. 1984. Industrial Cogeneration Potential (1980–2000) for Application of Four Commercially Available Prime Movers at the Plant Site. Report DOE/CS/40403-1. Washington, D.C.: U.S. Department of Energy.
U.S. Department of Energy. 1988. Manufacturing Energy Consumption Survey: Fuel Switching 1985. Report DOE/EIA-0515(85). Washington, D.C.: U.S. Department of Energy.
U.S. Department of Energy. 1989a. Energy Conservation Trends, Understanding the Factors That Affect Conservation Gains in the U.S. Economy. Report DOE/PPPE-0092. Washington, D.C.: U.S. Department of Energy.
U.S. Department of Energy. 1989b. A Compendium of Options for Government Policy to Encourage Private Sector Responses to Potential Climate Change. Report DOE/EH-0103. Washington, D.C.: U.S. Department of Energy.
U.S. Department of Energy. 1990. Manufacturing Energy Consumption Survey: Changes in Energy Efficiency 1980–1985. Report DOE/EIA/0516(85). Washington, D.C.: U.S. Department of Energy.
Williams, R. H., E. D. Larsen, and M. H. Ross. 1987. Materials, affluence, and industrial energy use. Annual Review of Energy 12:99–144.