Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
26 ENVIRONMENTALLY SIGNIFICANT CONSUMPTION 3 Tracking the Flows of Energy and Materials INTRODUCTION Consumption becomes environmentally important because of the manner or extent to which it transforms materials and energy. Therefore, to understand the environmental impacts of consumption, one must un- derstand anthropogenic changes in the flows of materials and energy. This chapter presents four brief reports, taken from presentations at the workshop, that track flows of energy and environmentally important materials or propose methods for tracking them. These reports suggest what can be learned by following materials and energy flows. Their bibliographies point to other related work. Iddo Wernick analyzes aggregate and per-dollar materials flows within the United States, using weight and volume as indicators. Al- though these units are not always good proxies for environmental im- pacts, the analysis provides a first approximation to importance by show- ing which human-environment interactions are the largest; by identifying trends, it highlights the materials that are likely to be increasing or de- creasing as environmental problems. For instance, many materials used in bulk, such as steel and wood, are becoming less important aspects of economic activity, and special-purpose materials used in lesser quantity, such as special alloys, plastics, and coated papers, are becoming more important (see Larson et al., 1986). The new materials have quite different environmental impacts from one another. The use of paper, despite the information revolution, has continued to increase in absolute terms and 26
TRACKING THE FLOWS OF ENERGY AND MATERIALS 27 has held steady on a per-gross-national-product (GNP) basis throughout this century. This sort of analysis, combined with information on the per- unit environmental impacts of the production and consumption of par- ticular materials, can suggest which kinds of consumption are likely to remain, or to become, environmentally important. David Allenâs report focuses on wastes, including air pollutants as well as solid wastes. Allen identifies the sources of these wastes by type of industry. He also illustrates, with an analysis of the inputs and wastes associated with producing a kilogram of polyethylene, how the environ- mental impacts of particular materials or energy transformations can be examined. Data like these can be combined with production data and estimates of the toxicity of each type of emission to yield comparative quantitative assessments of the environmental significance of each prod- uct of the chemical industry or some other segment of the economy. This sort of analysis can clarify the relative environmental importance of dif- ferent kinds of economic activity. David Allenâs approach is similar to one that has been used in energy analysis since the 1970s. Applying the approach to materials is more difficult, however, because materials differ qualitatively to the extent that it is not always meaningful to convert them to a common unit such as joules or kilograms. In addition, unlike energy, which dissipates as waste heat, many materials need to be tracked even after they are âused,â be- cause they continue to be transported through the environment and may reappear in environmentally significant ways. Lee Schipperâs report disaggregates one class of environmentally im- portant consumption. Schipper looks in detail at travel, a significant and growing factor in fossil energy use and associated climatic change and pollution. He disaggregates changes over time in carbon emissions from motor vehicles in wealthy countries into those attributable to levels of activity (passenger-kilometers traveled), energy intensity (fuel per pas- senger-kilometer), and the fuel-use characteristics of the vehicles, and then to finer levels of detail. For instance, he examines activity levels as a function of such variables as numbers of vehicles, load factors, and num- ber and average distance of trips. He finds that while fuel consumption for travel was leveling off in the United States, largely because of de- creases in fuel used per vehicle-kilometer between 1973 and 1991 (but not thereafter), this trend was not observed in other Organization for Eco- nomic Cooperation and Development (OECD) member countries: in all the countries studied, levels of activity have continued to increase since the 1970s with no sign of saturation in any country. The higher level of automobile travel in the United States is attributable to a greater number of trips of about the same average length as in other OECD countries. Schipper also examines such factors as sex and age of drivers. This sort of
28 ENVIRONMENTALLY SIGNIFICANT CONSUMPTION analysis is useful for separating technological influences on environmen- tally significant consumption from behavioral ones; projecting the likely environmental impacts of travel as a function of changes in incomes, age distribution, household composition, labor force participation, and other variables; and estimating the effects of policies to reduce emissions, such as fuel taxes, on different kinds of drivers. Faye Duchin explores the possibility of developing a classification system for households, akin to the Standard Industrial Classification sys- tem, as a way to facilitate disaggregating consumption by household type and by activities within these types, and to make possible systematic study of issues of consumption and âlifestyleâ such as those raised in Schipperâs work. Noting that various market research firms have devel- oped household classifications for short-term marketing purposes, Duchin suggests that a similar form of classification might be useful for detailed analysis of household consumption, including environmentally signifi- cant consumption. She notes that developing classification schemes for different countries may help illuminate the kinds of broad changes in household consumption patterns occurring in developing countries. All four reports illustrate the potential value of tracking and disag- gregating environmentally significant human activities. Such efforts ad- vance understanding by clarifying which actions and which actors are most responsible for particular environmental changesâand which make little difference. The results of such classifications could suggest where the greatest potential lies for reducing the environmental impact of hu- man activity. In addition, by identifying some of the immediate purposes for which people undertake activities that cause environmental damage (e.g., travel to work), this sort of research can identify sources of resis- tance to policy interventions and thus alert policy makers to challenges facing their efforts to reduce the environmental impacts of consumption. REFERENCE Larson, E.D., M.H. Ross, and R.H. Williams 1986 Beyond the era of materials. Scientific American 254 (June):34-41.
TRACKING THE FLOWS OF ENERGY AND MATERIALS 29 CONSUMING MATERIALS: THE AMERICAN WAY1 Iddo K. Wernick I focus in this paper on characterizing consumption by providing an account of all the physical materials consumed in the United States and a framework for assessing the relative scales and environmental relevance of that consumption. Assessing the materials consumption of a nation requires viewing (a) the total volume of materials consumed, (b) the com- position of that total, (c) how these change with time, (d) forces driving those changes, (e) foreign trade in raw materials, and (f) the prospects for large-scale materials recovery. Together, these allow us to view materials consumption comprehensively and place particular instances and anec- dotes in proper perspective. Since the oil price shocks of the 1970s many have studied energy consumption at the national level, examining consumption trends over time, the mix of fuels used, and alternatives for the future (United Na- tions, 1978; World Energy Conference, 1974). Such studies provide the analytic tools that have documented the slowing growth of primary en- ergy consumption and its decoupling from U.S. economic development (NakiÄenoviÄ, 1996). Although the analogy is imperfect, materials would similarly benefit from this approach but have not yet enjoyed the same scrutiny for several reasons. The same fear of imminent shortages that focused attention on energy never gathered momentum with respect to materials, as the proven reserve base for most material resources has actually grown in the last decades (U.S. Congress, 1952; Goeller and Weinberg, 1976; World Resources Institute, 1994). Although exhausting materials resources may, in fact, not be a priority concernâwith the no- table exception of high-grade energy fuelsâthe environmental degrada- tion resulting from extracting, processing, consuming, and disposing materials is. The heterogeneity of materials consumed in modern society presents a further barrier to comprehensive analysis. Materials possess numerous and diverse properties that make them attractive to consumers and deter- mine their environmental impacts, thus weakening generalizations. While the energy from firewood, coal, or gas is readily reduced to common units such as joules or British thermal units, the utility of the gravel, ore, and fertilizer materials we consume cannot be. Although less than an ideal measuring stick, mass will serve here as 1A longer version of this paper appeared in Technological Forecasting and Social Change, 1996, 53:111-122. Printed with permission of the journal.
30 ENVIRONMENTALLY SIGNIFICANT CONSUMPTION the common currency for describing materials. Using mass alone ob- scures environmentally important features of materials use, such as the growing volume of plastics in U.S. waste streams, the high toxicity of relatively trivial masses of industrial effluent, and the acreage disturbed in extracting both renewable and nonrenewable resources. Nonetheless, kilograms and tons provide objective measures for grasping the sheer quantities of bulk materials mobilized to serve society and the relative sizes of different materials classes. Moreover, most of the available data on materials are either given directly in mass or can be converted to it. CURRENT NATIONAL MATERIALS CONSUMPTION AND TEMPORAL DYNAMICS In 1990, the average American consumed over 50 kg of material per day, excluding water (Wernick and Ausubel, 1995). Consumer goods compose a small fraction of this total; the materials required for their production and distribution, as well as the auxiliary materials used in their manufacture, contribute a far greater amount. To gain some per- spective on the ratio of direct to indirect consumption, the mass of mu- nicipal waste that Americans directly dispose of each day accounts for less than 5 percent of the daily quantity (Franklin Associates, Ltd., 1992). Figure 3-1 shows the total as a sum of the six major classes of materials. Almost 90 percent of total inputs go to providing energy, structures, and food. Inputs of water, if included, would raise the total many fold. Min- ing wastes (particularly for coal) are huge and represent another conse- quence of consumption mostly hidden from the public eye. The daily 50- kg quantity may be common to highly industrialized societies. In 1990, Japanese consumption also summed to a little over 50 kg per capita per day (Gotoh, 1994). The mix of materials consumed changes over time. For example, per capita U.S. lumber consumption has declined markedly in this century. At the turn of the century wood provided building materials for homes and factories, ties and rolling stock for railroads, utility poles for tele- phone and power lines, and fuel. Today a large fraction of harvested wood (approximately 40 percent including residues) goes to paper mills (Ince, 1994). Although drastic reductions in consumption are more the exception than the rule, wood is not unique in that both the level of consumption and how it is used in the economy have changed. A more aggregated account of consumption reveals wholesale changes in the amount of physical structure materials Americans con- sume. Figure 3-2 shows that in total tonnage per capita, reported con- sumption appears to rise over long cycles of economic growth and then to fluctuate during times of economic upheaval.
TRACKING THE FLOWS OF ENERGY AND MATERIALS 31 FIGURE 3-1 Daily per capita materials flow by mass (all values in kg): United States about 1990. Materials are here classed as energy fuels (i.e., coal, oil, gas), construction minerals, industrial minerals, metals, forestry products, and agricul- tural products. Data from Wernick and Ausubel (1995). Reprinted with permis- sion. FIGURE 3-2 Annual per capita consumption of physical structure materials: United States 1900-1991. Physical structure materials are here defined as con- struction minerals, industrial minerals, forestry products. Data from Rogich et al. (1993); U.S. Bureau of the Census (1975). Reprinted with permission.
32 ENVIRONMENTALLY SIGNIFICANT CONSUMPTION Are industrialized societies constrained to follow this path indefi- nitely? Do improvements in the standard of living necessarily translate to greater material consumption? Intensity of use (IOU) measures address this question directly. IOU measures show the evolution of individual materials used in the national economy by indexing primary, as well as finished, materials to gross domestic product (Malenbaum, 1978). Begin- ning with studies done in the late 1970s, researchers noted several com- mon patterns in the course of consumption of a material in the economy (Williams et al., 1987). Initially, the consumption of a particular material exceeds general economic growth. Growing markets and newly discov- ered uses for the material stimulate further growth. This rapid growth eventually saturates, and consumption of that material then tracks or lags the rest of the economy. Figure 3-3 illustrates this phenomenon at different stages for a variety of materials in the United States. One clear conclusion from the figure is that more dollars in the economy do not always mean more tons. Heavy materials such as steel, copper, lead, and lumber, all materials used for infrastructure, became less critical to economic growth over the course of this century. Paper seems to track economic activity in lockstep, conserv- FIGURE 3-3 Materials intensity of use: United States 1900-1990. Annual con- sumption data are divided by GNP in constant 1987 dollars and normalized to unity in the year 1940. Data for plastics are production data. NOTE: St: Steel; Lu: Lumber. Data from U.S. Bureau of the Census (1993-1994 and 1975); Modern Plastics 136(5):71-72(1959); plastics data from personal communication with Joel Broyhill, Statistics Department, Society of the Plastics Industry, Washington, D.C., August 20, 1993. Reprinted with permission.
TRACKING THE FLOWS OF ENERGY AND MATERIALS 33 ing its role through the national shift from manufacturing to information and services. The rapid growth of materials used as fertilizers shows the âgreenâ revolution that has raised agricultural yields. Finally, low-den- sity materials, such as aluminum, have outpaced economic activity in the second half of the twentieth century. This is spectacularly true with re- spect to plastics, a class of materials that, in addition to being lightweight, possess a host of properties that make them the material of choice for the manufacturer and the consumer alike. The types of material flows can be separated into the categories of elephants and fleas. Some of the bulk materials we have seen may be called the elephants. These high-volume material flows may cause little environmental impact per unit mass but can have profound long-range environmental consequences. Pumping oil, quarrying stone, and har- vesting feed each contributes to chronic global environmental problems, affecting atmospheric composition and land use. The fleas, materials generated in small quantities often as by-products of large-scale commer- cial production, can have more acute harmful effects. Consider that total annual U.S. dioxin releases are under 500 kg (Thomas and Spiro, 1994). Despite the small quantity released, environmental concerns about the effects of dioxin continue to demand the attention of both government and industry. According to the U.S. Environmental Protection Agencyâs inclusive definition of Toxic Release Inventory (TRI) production-related wastes, toxic chemicals totaled about 17 million metric tons in 1992, 0.3 percent of all materials consumption (INFORM, 1995). Concerns over this relatively small mass fraction dominate much of the current public environmental debate. Foreign trade in raw materials accounts for about 10 percent of U.S. materials flows. Table 3-1 shows that a few bulk commodities dominate trade. On a mass basis, agricultural products, coal, and chemicals domi- nate U.S. exports, whereas oil, oil products, and metals and ores dominate imports. The plentiful carbon that enters America, of course, exits as CO2 emissions. Agricultural trade surpluses require domestic land, chemi- cals, and minerals but feed many elsewhere. For many minerals the United States shall continue to rely on foreign sources. FORCES AFFECTING MATERIALS CONSUMPTION The simple arithmetic of a U.S. population of 400 million or more in 2100 will draw more materials into the economy (United Nations, 1992). Efficiency improvements might be able to maintain a constant total for the collective whole, in theory. However, in the United States more people means more individual consumers acting on their own. The average number of residents per American occupied housing unit halved since
34 ENVIRONMENTALLY SIGNIFICANT CONSUMPTION TABLE 3-1 Major Materials Flows in U.S. Foreign Trade Exports Imports Net Flow (million (million per Capita Category metric tons) metric tons) (kg) Agricultural products 135.5 14.9 (482.6) Coal 96.0 2.4 (374.5) Minerals 47.8 54.2 25.6 Metals and ores 27.0 76.4 197.8 Chemical and allied products 41.3 14.4 (107.6) Petroleum products 34.1 96.9 251.3 Timber products 16.4 18.4 8.0 Paper and board 6.2 11.9 22.8 Oil (crude) 5.6 307.4 1207.6 Natural gas 1.7 31.0 117.2 Automobilesa 1.2 5.9 18.8 TOTAL 412.7 633.8 884.4 Air transport 1.5 1.7 0.8 Waterborne transport 406.9 524.9 472.0 Trucks 151,000 (units) 766,000 (units) N.A. Other industrial and consumer products ? ? NOTE: N.A. indicates not applicable. Numbers in parentheses indicate net exports. aBased on an estimated average vehicle mass of 1.5 metric tons. SOURCE: U.S. Bureau of the Census (1975). the beginning of the century (U.S. Bureau of the Census, 1975, 1993, 1994). Besides the materials needed for additional structures, appliances and furniture enter these dwellings irrespective of the number of inhabitants. Thus the relationship of number of persons to materials consumed is not simply proportional, reflecting settlement patterns as well. This same relation holds true for energy consumption: the same number of people living in a larger number of residences consume more (Schipper, 1996). While American behavior drives expansion, historical development and technical innovations offer hope for contraction. The United States is a postindustrial country. The service sector continues to claim more of national economic activity, and the physical infrastructure of the country is largely in place. For instance, during the period 1970-1992, the surfaced road network in the United States expanded at only a third of the rate for the century (U.S. Bureau of the Census, 1975, 1993, 1994). Because of the
TRACKING THE FLOWS OF ENERGY AND MATERIALS 35 FIGURE 3-4 Volume ratio of pipe manufactured from plastic over all other mate- rials. Data from Hurdelbrink (1989). Repinted with permission. massive quantities consumed constructing roads and highways, slowing the rate is consequential to national consumption of materials like steel, asphalt, sand, and rock. Substituting lighter for heavier materials also puts downward pres- sure on national materials use. Replacing heavy copper cable with light fiber optics not only reduces the amount of mass consumed but also reduces the need for mining copper ore. Lightweight plastics now pro- vide the primary material for pipes, formerly made of steel and lead (Figure 3-4). The quantity of carbon steel in American automobiles fell drastically during the 1970s, while high strength steel alloys, plastics, composites, and aluminum continue to make up more of our cars (Figure 3-5). For some products the same utility can be supplied with less mass of product. Metallurgical advances allow for steel beams with smaller cross-sectional areas to support loads. Sweetening foods with high-fruc- tose corn syrup uses only a fifth the mass of sugar to produce the same result to our palate. The ubiquitous aluminum beverage can is today 25 percent lighter than in 1973 (personal communication, Jenny Day, Direc- tor of Recycling, Can Manufacturers Institute). In addition to smaller mass, the aluminum beverage can provide a model of a highly successful recycling system with a recycling rate exceeding 70 percent.
36 ENVIRONMENTALLY SIGNIFICANT CONSUMPTION FIGURE 3-5 Mass of carbon steel, high-strength steel, composite materials, and plastics in the average U.S. automobile: 1969-1989. Data from Wardâs Automotive Yearbook (1970â1989). Reprinted with permission. The combination of forces to reduce materials use in the industrial- ized countries drives a process that researchers have dubbed âdematerial- ization,â or aggregate reductions in the amount of material needed to serve economic functions (Wernick et al., 1996). Substitution of materials that require less mass to deliver a unit of a given service, a phenomenon formally named âtransmaterialization,â represents a central component of the proposed shift to lowered consumption. Developing nations can benefit from the knowledge-based shift to lower materials requirements. The dematerialization hypothesis main- tains that as nations launch into development later, their initial growth rates may be sharper, but consumption saturates at lower levels, as they can avoid the materials-intensive process of trial and error experienced by the earlier starters (Grubler, 1990). The potential for reducing materi- als use through recycling, or âmaterials recovery,â can also be studied in terms of mass transformations (Rogich, 1993; Wernick, 1994; Wernick et al., 1996; Allen, this chapter).
TRACKING THE FLOWS OF ENERGY AND MATERIALS 37 CONCLUSIONS Sustaining the U.S. economy requires consuming large amounts of materials. The mix of materials changes with time, and these changes matter from the perspective of environmental quality. The question of whether Americans will consume more or less materials in the future depends on demographic, economic, and technical variables difficult, if not impossible, to predict. One central question is whether increases in materials efficiency can keep pace with, or even triumph over, the forces driving increased consumption. Toxics and other harmful materials constitute a small part of total consumption but are currently linked to the large-scale production of goods. They pose threats to human health and environmental quality far exceeding their mass fraction of materials consumption. To what extent these nasty residuals, often unintended by-products of production, can be eliminated presents a further question. The demand for better performance, and hence greater sophistication in materials and goods, has lightened many products and is key to future trends in materials consumption and efforts in materials recovery. Re- search and development efforts must combine environmental objectives with consumption trends to reduce primary materials requirements, de- sign products for recovery, and find uses for so-called wastes. While technology may offer some solutions and help reduce the envi- ronmental impact of our consumption, changing human behavior will surely prove more difficult. Technological and economic solutions must recognize the deep behavioral forces driving human consumption to ef- fect positive change. ACKNOWLEDGMENTS I thank Jesse Ausubel and Perrin Meyer at The Rockefeller University and Paul Waggoner at the Connecticut Agricultural Experiment Station for their help in preparing this manuscript. REFERENCES Franklin Associates, Ltd. 1992 Characterization of Municipal Solid Waste in the United States: 1992 Update, Final Report. EPA/530/R-92/019. Prairie Village, Kan.: Franklin Associates, Ltd. Goeller, H.E., and A.M. Weinberg 1976 The age of sustainability. Science 191:683-689.
38 ENVIRONMENTALLY SIGNIFICANT CONSUMPTION Gotoh, S. 1994 The Potential and Limits of Using Life-Cycle Approach for Improved Environ- mental Decisions. Paper presented at the International Conference on Industrial Ecology sponsored by the National Academy of Engineering. May 9-13, Irvine, Calif. Grubler, A. 1990 The Rise and Fall of Infrastructures: Dynamics of Evolution and Technological Change in Transport. Heidelberg, West Germany: Physica-Verlag. Hurdelbrink, R. 1989 An analysis of materials selection criteria for synthetic polymer systems. In Pro- ceedings of the Industry-University Advanced Materials Conference II. Denver, Colo.: Advanced Materials Institute, Colorado School of Mines. Ince, P.J. 1994 Recycling of Wood and Paper Products in the United States. Paper presented at United Nations Economic Commission for Europe Timber Committee Team of Specialists on New Products, Recycling, Markets, and Applications for Forest Products. June 1994. INFORM 1995 Toxics Watch 1995. New York: Inform, Inc. Malenbaum, W. 1978 World Demand for Raw Materials in 1985 and 2000. New York: McGraw-Hill. NakiÄenoviÄ, N. 1996 Freeing energy from carbon. Daedalus 125(3):95-112. Rogich, D.G. 1993 Materials Use, Economic Growth, and the Environment. Paper presented at the International Recycling Congress and RECâ93 Trade Fair. Washington, D.C.: U.S. Bureau of Mines. Schipper, L. 1996 Life-styles and the environment: The case of energy. Daedalus 125(3):113-138. Thomas, V.M., and T.J. Spiro 1994 An Estimation of Dioxin Emissions in the United States. PU/CEES Report No. 285. Princeton, N.J.: Center for Energy and Environmental Sciences, Princeton Univer- sity. United Nations 1978 World Energy Supplies 1972-1976. New York: United Nations. 1992 Long-Range World Population Projections: Two Centuries of Population Growth 1950- 2150. New York: United Nations. U.S. Bureau of the Census 1975 Historical Statistics of the United States, Colonial Times to 1970. Washington, DC: U.S. Department of Commerce. 1993 Statistical Abstract of the United States, 113 ed. Washington, D.C: U.S. Department of Commerce. 1994 Statistical Abstract of the United States, 114 ed. Washington, D.C: U.S. Department of Commerce. U.S. Congress 1952 Resources for Freedom: Report of the Presidentâs Materials Policy Commission (Paley Report). U.S. House of Representatives Document No. 527. Washington, D.C.: U.S. Government Printing Office. Wardâs Communications 1970- Wardâs Automotive Yearbook. Detroit: Wardâs Communications. 1989
TRACKING THE FLOWS OF ENERGY AND MATERIALS 39 Wernick, I.K. 1994 Dematerialization and secondary materials recovery: A long run perspective. Journal of the Minerals, Metals, and Materials Society 46(4):39-42. Wernick, I.K., and J.H. Ausubel 1995 National materials flows and the environment. Annual Review of Energy and En- vironment 20:462-492. Wernick, I.K., R. Herman, S. Govind, and J.H. Ausubel 1996 Materialization and dematerialization: Measures and trends. Daedalus 125(3):171- 198. Williams, R.H., E.D. Larson, and M.H. Ross 1987 Materials, affluence and industrial energy use. Annual Review of Energy and Envi- ronment 12:99-149. World Energy Conference 1974 World Energy Conference Survey of Energy Resources. New York: World Energy Conference. World Resources Institute 1994 World Resources: A Guide to the Global Environment 1994-1995. New York: Oxford University Press.
40 ENVIRONMENTALLY SIGNIFICANT CONSUMPTION WASTES AND EMISSIONS IN THE UNITED STATES David T. Allen More than 12 billion tons of industrial waste are generated annually in the United States. This is equivalent to more than 40 tons of waste for every man, woman, and child in the country [U.S. Environmental Protec- tion Agency (E.P.A.), 1988a,b; Allen and Jain, 1992]. The sheer magnitude of the waste generation is cause for concern and drives us to identify the characteristics of the wastes, the manner in which the wastes are being managed, and the potential for reducing wastes. This summary provides a brief overview of the information available on waste generation and management. The sources and nature of industrial hazardous wastes, nonhazardous wastes, municipal solid wastes, and emissions of criteria and hazardous air pollutants will all be reviewed. INDUSTRIAL WASTES Industrial wastes can be in solid, liquid, or gaseous states. Most industrial wastes in solid or liquid form fall under the provisions of the Resource Conservation and Recovery Act (RCRA). The total mass of wastes generated each year that fall under the provisions of RCRA has been estimated to be in excess of 12 billion tons. Table 3-2 lists estimated rates of nonhazardous waste generation in the United States (U.S. E.P.A., 1988a). The total of more than 11 billion tons can be contrasted with an estimate of 0.75 billion tons of hazardous waste generation (U.S. E.P.A., 1991; Baker et al., 1992). Thus, hazardous wastes represent less than 10 percent of the total industrial waste mass, yet almost all of our data on waste generation and treatment focus on this small segment of the waste system. Figure 3-6 shows the industrial sectors responsible for the gen- eration of hazardous wastes. Chemical manufacturing clearly dominates. In contrast, the paper, mining, and electric power-generation industries dominate estimates of nonhazardous waste generation. Although they account for more than 12 billion tons of industrial waste generation, wastes regulated under RCRA do not represent the entire waste burden. Emissions to the atmosphere do not, in general, fall under the provisions of RCRA. Data on emissions into the atmosphere can be broadly classified into emissions of hazardous air pollutants and criteria air pollutants. The richest source of data on atmospheric emis- sions of hazardous air pollutants from industrial facilities is the Toxics Release Inventory (TRI). The TRI reports the releases and transfers of more than 300 chemicals (soon to be 600 chemicals) from manufacturing facilities. In 1991, more than 7 billion pounds of releases were reported
TRACKING THE FLOWS OF ENERGY AND MATERIALS 41 TABLE 3-2 Sources of Nonhazardous Waste Regulated under RCRA Estimated Annual Generation Rate Waste Category (million tons) Industrial nonhazardous wastea,b 7,600 Oil and gas wastec,e Drilling wasted 129â871 Produced watersf 1,966â2,738 Mining wastec,g > 1,400 Municipal wasteb 158 Household hazardous waste 0.002â0.56 Municipal waste combustion ashh 3.2â8.1 Utility wastec,i Ash 69 Flue gas desulfurization waste 16 Construction and demolition wastej 31.5 Municipal sludgeb Wastewater treatment 6.9 Water treatment 3.5 Very-small-quantityk generator Hazardous waste (<100 kg/mo)b,e 0.2 Waste tiresg 240 million tires Infectious wastec,l 2.1 Agriculture waste Unknown Approximate Total > 11,387 aNot including industrial waste that is recycled or disposed of off site. bThese estimates are derived from 1986 data. cSee Science Applications International Corporation, 1985. dConverted to tons from barrels: 42 gal = 1 barrel, â 17 lb/gal. eThese estimates are derived from 1985 data. fConverted to tons from barrels: 42 gal = 1 barrel, â 8 lb/gal. gThese estimates are derived from 1983 data. hThis estimate is derived from 1988 data. iThese estimates are derived from 1984 data. jThis estimate is derived from 1970 data. kSmall quantity generators (100-1,000 kg/mo waste) have been regulated under RCRA, Subtitle C, since October 1986. Before then, approximately 830,000 tons of small-quantity generator hazardous wastes were disposed of in Subtitle D facilities every year. lIncludes only infectious hospital waste. SOURCE: U.S. Environmental Protection Agency (1988a).
42 ENVIRONMENTALLY SIGNIFICANT CONSUMPTION FIGURE 3-6 Industrial sources of hazardous waste. Data from Baker and War- ren (1992). through the TRI. The major industry categories contributing to this total are listed in Table 3-3. While TRI releases are more than three orders of magnitude less than the mass of wastes reported under RCRA, RCRA wastes are sent to treatment technologies that reduce waste volume and toxicity, whereas TRI releases are generally emitted directly to the envi- ronment. Thus, direct releases to the environment, reported through the TRI, may be as significant in affecting environmental quality as RCRA TABLE 3-3 Sources of Releases and Transfers of Toxics Total Releases Standard and Transfers Industrial Classifications (millions (% of Industry Code of pounds) total) Classification Food and kindred products 20 85 1.2 Tobacco products 21 3.6 0.049 Textile mill products 22 37 0.51 Apparel, etc.a 23 1.9 0.026 Lumber and wood products, necb 24 41 0.56 Furniture and fixtures 25 67 0.93 Paper and allied products 26 310 4.2 Printing, publishing, and allied industries 27 57 0.78 Chemicals and allied products 28 2,800 38 Petroleum refining and related industries 29 720 9.9 Rubber and miscellaneous plastic products 30 200 2.7 Leather and leather products 31 19 0.27
TRACKING THE FLOWS OF ENERGY AND MATERIALS 43 TABLE 3-3 Continued Total Releases Standard and Transfers Industrial Classifications (millions (% of Industry Code of pounds) total) Classification Stone, clay, glass, and concrete products 32 55 0.75 Primary metal industries 33 1,200 17 Fabricated metal products, necb 34 340 4.7 Industrial and commercial machinery and computer equipment 35 91 1.3 Electronic and other electrical equipment and components, necb 36 350 4.8 Transportation equipment 37 310 4.3 Measuring instruments, optical goods, watchesc 38 71 0.98 Miscellaneous manufacturing industries 39 34 0.46 Multiple manufacturing classifications Stone, etc. products/primary metal industries 32 and 33 37 0.51 Primary metals/fabricated metals 33 and 34 36 0.50 Chemical products/petroleum refining 28 and 29 35 0.49 Primary metals/electronic equipment 33 and 36 26 0.35 Chemical products/primary metals 28 and 33 25 0.35 Chemical products/rubber, plastic products 28 and 30 25 0.34 Food products/chemical products 20 and 28 20 0.28 Paper products/electronic equipment 26 and 36 17 0.24 Electronic equipment/transportation equipment 36 and 37 14 0.19 Fabricated metals/transportation equipment 34 and 37 11 0.15 Fabricated metals/electronic equipment 34 and 36 11 0.15 Primary metals/machinery 33 and 35 11 0.15 Other manufacturing industry combinations 20 to 39 170 2.3 Not classified as manufacturing No data, < 20, > 39 38 0.53 TOTAL 7,300 100 aFull description is as follows: Apparel and other finished products made from fabrics and other similar materials. bnec: not elsewhere classified. cFull description is as follows: Measuring, analyzing, and controlling instruments; pho- tographic, medical, and optical goods; watches and clocks. SOURCE: Toxic Release Inventory (1993).
44 ENVIRONMENTALLY SIGNIFICANT CONSUMPTION wastes, even though the mass of RCRA wastes dwarfs the mass reported through the TRI. Criteria pollutants are the other major category of air pollutants. The criteria pollutants include particulate matter less than 10 microns in diameter, sulfur dioxide, nitrogen oxides, carbon monoxide, ozone, and lead. Volatile organic hydrocarbons are not included in the list of criteria pollutants, but because of their role in forming ozone in the lower atmosphere, they are measured and controlled. The criteria air pollutants emitted by major industrial sectors are given in Figure 3-7. The emissions are on the order of millions of tons per year, comparable to the emissions reported through the TRI, but orders of magnitude lower than the wastes governed by RCRA. MUNICIPAL SOLID WASTES AND OTHER WASTE STREAMS Approximately 200 million tons of municipal solid waste are gener- ated annually in the United States. The main constituents are paper, yard wastes, food wastes, plastics, and metal. While much attention has been focused on these materials, the quantities of these waste streams are just a few percent of the overall waste material flows in the United States. In- dustrial wastes generated in the production of commodity materials and the generation of power and fuel are far more extensive than post-con- sumer materials. Agricultural wastes are also extensive but are not well characterized. IMPLICATIONS OF WASTE STREAM FLOWS Per capita waste generation in the United States is approximately 40 tons per year. This amount is roughly equivalent to each person generat- ing their body weight in wastes each day when water is excluded from the 40 tons (Wernick and Ausubel, 1995). In another paper in this vol- ume, Wernick reports that per capita material consumption in the United States is also approximately a body weight per day. Making detailed comparisons between the material use and waste generation is difficult because wastes are often poorly characterized, making the determination of flows of specific materials uncertain. Further, because the measured flows of wastes often contain significant quantities of diluting species such as water, even performing total mass balances on materials used and wastes generated is difficult. Because of these difficulties, which are de- scribed in more detail elsewhere (Allen and Jain, 1992), this paper only briefly summarizes the waste flow and comparison data and does not attempt to compare them in detail to material use data, although such comparisons should be the topic of future research. The available data on wastes clearly indicate that most materials used
5 4 3 2 Millions metric tons 1 TRACKING THE FLOWS OF ENERGY AND MATERIALS 0 Pulp and paper Chemicals Petroleum Stone, clay, glass Primary metals Other Particulates Sulfur oxides Nitrogen oxides VOCs Carbon monoxide FIGURE 3-7 Industrial sources of criteria air pollutants (United States, 1994). NOTE: VOCs = volatile organic compounds. Data from U.S. Department of Energy (1994). 45
46 ENVIRONMENTALLY SIGNIFICANT CONSUMPTION in the United States are from virgin resources, and only a relatively small fraction are recovered or reprocessed materials. Yet these 12 billion tons per year of wastes should not be ignored as a potential resource. Studies in what has come to be called industrial ecology or industrial metabolism are probing the material efficiencies of large industrial systems, searching for ways to improve material and energy efficiencies. Although it may be argued that low rates of material reuse are due to the inherently low value of waste streams, data on waste compositions tell a different story. For example, Allen and Behmenesh (1994) found that billions of dollars in metals are disposed of each year in hazardous waste streams, and a large fraction of this material is present at concentrations higher than those found in ores that are currently mined. The reasons for these lost oppor- tunities are complex (Allen, 1993), but it is clear that much more material- efficient industrial systems are feasible. ENVIRONMENTAL IMPACTS OF CONSUMPTION There are several lessons to be learned from waste and emission in- ventories. â¢ Waste flows are substantial and are dominated by the by-products of the manufacture of commodity materials and energy; post-consumer waste flows are relatively minor in comparison. â¢ Many of the waste streams contain valuable resources at concen- trations that should allow for economical recovery, yet a series of regula- tory, financial, structural, and technical barriers make the development of more efficient structures difficult. â¢ Information on waste stream flows exists in many disparate loca- tions. For some waste streams very little information is available. This lack of information is a major barrier to the formation of markets that could find uses for materials currently wasted. To be most useful in estimating the environmental impacts of con- sumption, the waste and emission data should be aggregated along prod- uct lines. The emerging discipline of Life Cycle Assessment attempts to accomplish this task. Consider Table 3-4, which is a compilation of the wastes, emissions, and raw materials used in generating a kilogram of polyethylene. These data reveal that emissions and wastes are generated throughout a productâs life cycle, from extraction of raw materials, through manufacturing, and in final product disposal. Further, the wastes and emissions fall into a variety of categories and will have a variety of impacts. In the face of this complexity, developing simple measures of the environmental impacts of consumption will be challenging. Despite
TRACKING THE FLOWS OF ENERGY AND MATERIALS 47 TABLE 3-4 Raw Material Consumption, Emissions, and Energy Use Associated with the Manufacture of 1 kg of Polyethylene Category Unit Average Fuels, MJ Coal 3.28 Oil 3.58 Gas 12.38 Hydro 0.54 Nuclear 1.67 Other 0.21 TOTAL 21.66 Feedstock, MJ Coal < 0.01 Oil 33.87 Gas 33.02 Wood < 0.01 TOTAL 66.89 TOTAL FUEL PLUS FEEDSTOCK 88.55 Raw materials, mg Iron ore 200 Limestone 150 Water 24,000,000 Bauxite 300 Sodium chloride 8,000 Clay 20 Ferro-manganese <1 Air emissions, mg Dust 3,000 Carbon monoxide 900 Carbon dioxide 1,250,000 Sulfur oxides 9,000 Nitrogen oxides 12,000 Hydrogen chloride 70 Hydrogen fluoride 5 Hydrocarbons 21,000 Other organics 1 Metals 5 Water emissions, mg COD 1500 BOD 200 Acid as H+ 60 Nitrates 5 Metals 250 Ammonium ions 5 Chloride ions 130 Dissolved organics 20 Suspended solids 500 Oil 200 Hydrocarbons 100 Dissolved solids 300 Phosphate 5 Other nitrogen 10 continued on next page
48 ENVIRONMENTALLY SIGNIFICANT CONSUMPTION TABLE 3-4 Continued Category Unit Average Solid waste, mg Industrial waste 3,500 Mineral waste 26,000 Slags and ash 9,000 Toxic chemicals 100 Nontoxic chemicals 800 NOTE: The feedstock energy is the energy value of the fuels used as the raw materials for making polyethylene; the other fuels listed are consumed in the manufacturing and raw material extraction processes. COD: chemical oxygen demand; BOD: biological oxygen demand. SOURCE: Boustead (1993). the challenges, systems are emerging to perform product environmental impact assessments and these are described in a growing literature on Life Cycle Assessment (Society of Environmental Toxicology and Chem- istry, 1991; 1993a,b; 1994). REFERENCES Allen, D.T. 1993 Using wastes as raw materials: Opportunities to create an industrial ecology. Hazardous Waste and Hazardous Materials 10:273-277. Allen, D.T., and N. Behmanesh 1994 Wastes as raw materials. Pp. 69-89 in B.R. Allenby and D.J. Richards, eds., The Greening of Industrial Ecosystems. National Academy of Engineering. Washing- ton, D.C.: National Academy Press. Allen, D.T., and R. Jain, eds. 1992 Hazardous Waste and Hazardous Materials 9(1):1-111. Baker, R.D., and J.L. Warren 1992 Generation of hazardous waste in the United States. Hazardous Waste and Haz- ardous Materials 9:19-35. Baker, R.D., J.L. Warren, N. Behmanesh, and D.T. Allen 1992 Management of hazardous waste in the United States. Hazardous Waste and Hazardous Materials 9:37-59. Boustead, I. 1993 Ecoprofiles of the European Plastics Industry. Reports 1-4 (May). Brussels, Belgium: European Centre for Plastics in the Environment (PWMI). Science Applications International Corporation 1985 Summary of Data on Industrial Non-hazardous Waste Disposal Practices. Contract 68-01-7050. Washington, D.C.: U.S. Environmental Protection Agency. Society of Environmental Toxicology and Chemistry (SETAC) 1991 A Technical Framework for Life Cycle Assessment. Report from the Smugglerâs Notch, Vt., workshop held August 1990. Pensacola, Fla.: SETAC.
TRACKING THE FLOWS OF ENERGY AND MATERIALS 49 1993 Conceptual Framework for Impact Assessment. Report from the Sandestin, Florida, workshop held February 1992. Pensacola, Fla.: SETAC. 1993 Guidelines for Life Cycle Assessment. Report from the workshop held in Sesimbra, Portugal in March 1993. Pensacola, Fla.: SETAC. 1994 Life Cycle Assessment Data Quality: A Conceptual Framework. Report from the Win- tergreen, Va., workshop held October 1992. Pensacola, Fla.: SETAC. Toxic Release Inventory 1993 Toxic Chemical Release Inventory for 1991 (database). Bethesda, Md.: National Library of Medicine. U.S. Department of Energy 1994 Waste Generation in Industry, Draft, Office of Energy Efficiency, Washington, D.C. U.S. Environmental Protection Agency 1988a Report to Congress: Solid Waste Disposal in the United States, Vol. 1, EPA/ 530-SW-88-011. Washington, D.C.: U.S. Environmental Protection Agency. 1988b Report to Congress: Solid Waste Disposal in the United States, Vol. 2, EPA/ 530-SW-88-011B. Washington, D.C.: U.S. Environmental Protection Agency. 1991 National Survey of Hazardous Waste Generators, and Treatment, Storage, Disposal and Recycling Facilities in 1986: Hazardous Waste Management in TSDR Units. EPA/ 530-SW-91-060. Washington, D.C.: U.S. Environmental Protection Agency. Wernick, I.K., and J.H. Ausubel 1995 National material flows and the environment. Annual Review of Energy and the Environment 20:463-492.
50 ENVIRONMENTALLY SIGNIFICANT CONSUMPTION CARBON EMISSIONS FROM TRAVEL IN THE OECD COUNTRIES Lee J. Schipper This paper examines some of the forces driving increased emissions of greenhouse gases in developed countries, focusing on carbon dioxide and travel (a longer paper addresses other sources of emissions, Schipper, forthcoming). Figure 3-8 shows carbon emissions per capita in OECD countries in 1973 and 1991 from energy-using activities, allocating the emissions from production of electricity and district heating to the end uses of those energy forms in proportion to final use (Schipper, Haas, and Sheinbaum, 1996; Scholl et al., 1996; Schipper, Scholl, and Price, in press; Schipper, Ting et al., 1996; Torvanger, 1991). The figure illustrates well that it is possible to connect emissions to the activities where they arise. For many countries, per capita emissions from these activities actually fell between the years portrayed; normalizing by gross domestic product in each year shown would show a dramatic decline in every country. In- deed, absolute emissions for most countries shown were close to or lower than their 1973 levels in 1991, but absolute emissions have begun to rise since then. This analysis uses Laspeyres indices to measure how components of energy use changed (Howarth et al., 1993). The analysis decomposes or factors total energy use or emissions into a sum, over end-use sectors (such as modes of travel), of the products of subsectoral activity and energy or emissions intensity (i.e., energy or emissions per unit of activ- ity): E = Î£ Ai Ã Ei where E is energy use or emissions, Ai is activity in passenger-kilometers in mode i (including the driver for automobiles and light trucks), Ei is the energy or emisions per unit of activity in mode i, and Ai is calculated by multiplying A (total passenger-kilometers trav- eled) by Si (the share of activity in mode i). Letting one component follow its historic course while holding the others constant at their base year values shows how that component influenced energy use over time. This approach (see Schipper, forthcoming) shows that the growth in energy demand is shifting from producers (manufacturing and freight) to consumers (household comfort and mobility), as well as to services; these services act as both producers (i.e., insurance, banking, health) and con- sumers (personal services like shopping and leisure activities). This shift âfrom production to pleasureâ is barely discernible for the United States (Figure 3-9) because energy savings there were so great for household purposes and private cars, but it is very noticeable in Germany (Figure 3- 10) and other OECD countries. The shift depends on the rising impor-
TRACKING THE FLOWS OF ENERGY AND MATERIALS 51 FIGURE 3-8 Carbon emissions per capita in 8 OECD countries, 1972-1973 and 1991 by end use. NOTE: tcarbon = tons of carbon. SOURCE: Schipper, Ting et al. (1996). FIGURE 3-9 Evolution of energy use in the United States 1960-1992: From pro- duction to pleasure. SOURCE: Lawrence Berkley Laboratory calculations based on materials in Schipper, Ting et al. (1996).
52 ENVIRONMENTALLY SIGNIFICANT CONSUMPTION FIGURE 3-10 Evolution of energy use in the Federal Republic of Germany 1960- 1992: From production to pleasure. SOURCE: National Energy Balances from the Federal Republic of Germany. tance of households and personal transportation over the other sectors and means that energy uses are spreading from the largest users (facto- ries) to the smallest users (households and the users of individual ve- hicles). In personal transportation, activity increased principally in the more energy-intensive modes of cars and air, and overall energy intensities fell only in the United States. Significantly, the energy intensity of automo- bile travel in the United States, which fell from the mid-1970s until 1991, has risen since then. The real, on-road fuel intensity of new cars and household light trucks is no longer lower than that of the fleet of the same vehicles, signaling the end of an era that had seen the rapid decline in the energy intensities of new vehicles. Scholl et al. (1996) analyzed the changes in travel emissions. Simpli- fying this work is the fact that travel depends almost wholly on oil prod- ucts, for which emissions vary little from one fuel to the next. As a result, changes in emissions depend principally on changes in energy use. Fig- ure 3-11 shows per capita emissions for travel by mode in 1973 and 1992 for Japan, the United States, and eight European countries aggregated. Because the differences among European countries, and changes in Eu- rope over time were relatively uniform, these countries are aggregated to simplify the description.
TRACKING THE FLOWS OF ENERGY AND MATERIALS 53 FIGURE 3-11 Carbon emissions from travel in 1973 and 1992. EU-8: West Ger- many, Denmark, Finland, Luxembourg, Norway, Sweden, France, and United Kingdom. NOTE: Lt: Light. SOURCE: Scholl, Schipper, and Kiang (1996). Figure 3-12 shows the behavior of per capita aggregate emissions over time. The dips in the evolution of aggregate emissions were caused mainly by declines in activity during periods of recession and higher fuel prices. Note that the predominant trend was toward greater per capita emissions, however, principally from automobiles, except in the United States Not surprisingly, aggregate carbon intensity (Figure 3-13), the ratio of emissions to aggregate activity in passengerâkilometers, increased over time as well, decreasing only in the United States; in Denmark and Italy, there were marginal decreases relative to 1973.1 In all but a few countries, it took more energy (and released more carbon) to transport a person one kilometer in 1992 than in 1973. Figure 3-14 compares the different effects. In all countries/regions, activity (greater domestic travel per capita) boosted emissions, from a low of 31 percent in the United States to 65 percent in Japan and about 40 1In this analysis the unit of activity, passenger-kilometer, is calculated for automobiles as vehicle-kilometer times load factor, or people per car.
54 ENVIRONMENTALLY SIGNIFICANT CONSUMPTION FIGURE 3-12 Travel carbon emissions per capita in 10 OECD countries. NOTE: tC = tons of carbon. SOURCE: Scholl, Schipper, and Kiang (1996). FIGURE 3-13 Aggregate carbon intensity of travel in 10 OECD countries. SOURCE: Scholl, Schipper, and Kiang (1996).
TRACKING THE FLOWS OF ENERGY AND MATERIALS 55 FIGURE 3-14 Impact of changes in components on carbon emissions from travel. SOURCE: Scholl, Schipper, and Kiang (1996). percent in European countries as a whole. Structural (modal) shifts to- ward cars and air travel increased emissions almost everywhere, above all in Japan, where automobiles only passed the 50 percent share of total travel in the late 1980s. Intensity changes reduced emissions by nearly 20 percent in the United States but had little effect elsewhere. This surprising result occurred because the load factors of automobile travel fell by 25 percent in every country, while the energy intensities of car use, in milli- joules per vehicle-kilometer, fell by less than 15 percent except in the United States, where they fell by 35 percent through 1991, at which time the decline stopped abruptly. Air travel energy intensity fell significantly in every country, but this change contributed marginally to total emis- sions, except in the United States. As a result of all of these changes combined, emissions per capita fell only in the United States, but in- creased in every other country, along with the share of carbon emissions from travel, as can be seen in Figure 3-11. The high level of carbon emissions from travel in the United States relative to the other countries shown in Figure 3-11 is a function of the distances Americans travel. In particular, Americans travel 60-100 per-
56 ENVIRONMENTALLY SIGNIFICANT CONSUMPTION cent farther per capita by car than do Europeans. Surprisingly, however, an average car trip in either region is between 12.5 km and 15 km (Schipper, Gorham, and Figueroa, 1996). Thus, it is the frequency of car travel, not âdistancesâ per se, that boosts Americansâ travel. ENERGY, EMISSIONS, AND LIFESTYLE Examining variation in carbon emissions for the same activities among industrialized countries suggests which factors drive emissions and might lie behind potential for future restraint. I examine recent trends in automobile characteristics and use these as a window into the role in emissions of âlifestyleââthat is, the bundle of activities in which indi- viduals engage (Schipper et al., 1989). A variety of indicators describe lifestyle: personal consumption expenditures, ownership of and access to energy-using consumer goods, time use, and distance traveled. Lifestyle âattributesâ include the sociodemographic characteristics, such as age distribution or employment status of individuals and families. Lifestyle âchoicesâ are activities that the population as a whole, sociodemographic subgroups, or individuals makeââfor example, choices on how much time to spend outside the home. These characteristics are not indepen- dent since families with small children may have to spend more time at home than families with no children. Some trends that increased energy use, such as increased numbers of women working (and driving to work) or smaller household size (which raised per capita area) can hardly be âfaultedâ for raising CO2 emissions, but other trends, such as the pur- chase of larger homes or the gradual movement of households away from cities, must be considered conscious decisions, at least in part. Income-driven lifestyle changes during the past decades have raised energy use for pleasureâi.e., for comfort and mobility. This is one way that lifestyle, as measured by the ownership and use of household equip- ment, travel, and visits to the service sector, continues to increase carbon emissions, even when those increases are less than proportional to in- come increases. Since the 1973 oil crisis, many energy uses have become less energy intensive, while lifestyles, continuing a long-term trend, have become more energy intensive. This is not a rebound effect: the largest energy savings and emissions reductions occurred in the household sec- tor, where the largest declines in energy intensities occurred. During this period no energy savings occurred in travel, except in the United States. The potential for technological change in transportation energy use (Schipper, Meyers et al., 1992; Schipper, 1993), is only being harvested slowly. Therefore at present the âstructuralâ changes in these sectors drive energy use, as people change. Schipper et al. (1989) demonstrated that much of this change can be measured by following expenditures of money and time. As Gershuny and Jones (1987) demonstrated, most of us
TRACKING THE FLOWS OF ENERGY AND MATERIALS 57 have more leisure time and spend increased amounts of that leisure away from home, consistent with what surveys of individual travel show (Schipper, Gorham, and Figueroa, 1996). As every person with a driverâs license has at least one car, the characteristics of these devices and their overall use become more important in determining energy use, unless new energy-intensive technologies appear. Unless energy prices are ex- tremely high, many of these choices will be made with little regard for energy prices. Although household energy uses appear saturated, no such trend is apparent for travel in the 1990s. In addition to increased car ownership, car characteristics and use have increasing importance to emis- sions. Figure 3-15 shows an important indicator of automobile characteris- tics that affects emissions, new car weight, in the United States and some European countries (Schipper, 1995). Although the weight of a United States car fell significantly in the late 1970s, weight increased after the early 1980s; in contrast, weight of new cars in Europe appears to have increased continuously. When the rising share of light trucks is added to the United States figures, the rebound is more dramatic. Nevertheless Americans occupied considerably lighter new cars in 1993 than in 1973. Needless to add, the engine size, or horsepower, in Europe increased continuously, while the same parameters dropped and then slowly re- bounded in the United States. Figure 3-15 contrasts with the data on space-heating intensity, which fell 25-50 percent in OECD countriesâso much so that even with modest increases in heated area, per capita space-heating energy use was lower in 1991 than in 1973 in almost all the countries we studied. But the analo- gous change in energy use for travel occurred only in the United States. To be sure, the ratio of fuel consumption to weight in new cars in virtually every OECD country has fallen continuously in all countries shown and, in fact, differs very little between these countries. Thus in a technological sense, cars are almost equally âefficientâ in all countries, but their test fuel consumption still differs significantly because of weight, power, and other features. Because cars are heavier now than in 1980, actual fuel consump- tion per kilometer has fallen very little, except in the United States. Figure 3-16 shows the use of cars in kilometer per capita per year; this reflects both the distances cars are driven and the number of cars per person. In the countries with the fewest cars (Finland, Britain, or Den- mark), yearly usage per car is very high, accounting for the small range of car use shown within Europe. Australia lies slightly above the European 2Multiplying the values in this figure by the load factor, 1.5-1.7 people per car, gives the travel from cars. Only a small part of the United States-Europe gap in either cars or car use per capita is filled by much higher use of bus and rail in Europe. In fact, the car accounts for 80-85 percent of all travel in the European countries shown and accounts for 55 percent of travel even in Japan.
58 ENVIRONMENTALLY SIGNIFICANT CONSUMPTION FIGURE 3-15 Average new car weight in Europe and the United States. Data excludes light trucks. Data from U.S. Department of Transportation, National Highway Traffic Safety Administration; Statistics Sweden; and European Associ- ation of Car Manufacturers. FIGURE 3-16 Per capita car use in 8 OECD countries. Data include household light trucks in the United States, Britain, and Denmark. Adapted from Lawrence Berkeley Laboratory.
TRACKING THE FLOWS OF ENERGY AND MATERIALS 59 countries, and Japan lies far below. The United States value appears to be moving away from the levels elsewhere.2 The U. S. data are no surprise in a country with fuel prices lying at one-third to one-fourth of the other countries in this study. Whatever the exact coupling between fuel prices and new car characteristics or use, the result of Americansâ choices is three to four times the per capita carbon emissions from personal vehicles, mostly because of greater per capita driving but also because U. S. cars use 25-33 percent more fuel per km than those in Europe. Unlike central heating, house area, and appliance ownership, which appear to be saturating (Schipper, in press), there is no evidence of such saturation for travel. And while new homes or appliances are signifi- cantly less energy intensive than those they replace or supplant (Schipper, Meyers, et al., 1992), the same cannot be said of automobiles, (although it is true for aircraft). Thus, although the household sector significantly reduced its CO2 emissions in most countries studied, emissions from the travel sector were lower in 1991 than in 1973 only in the United States and started to rise after 1991 even there. Hence, travel is an area for concern. In the future, only very slow growth can be expected in the structural factors that formerly pushed up household energy use because of slowing trends toward larger homes, smaller households, and ownership of major energy-using equipment. The travel sector, however, differs. Although gradual aging of the population may leave more of us home more often, and less mobile, roughly 30-40 percent of all Europeans of driving age still do not drive. These are mostly older people; among those in the 20-35 age group, car use is almost universal. Therefore, car use is expected to increase in Europe. Moreover, increases in driving in the United States and Europe are mainly to visit the service sector or for free time and holidays. Liberalization of shopping hours in Europe may encourage more evening and weekend car use. And the characteristics of new cars in Europe and the United States continue to evolve in ways that are more fuel intensive, offsetting much or all of the effort to reduce fuel use through technology. The high fuel prices in Europe will probably keep a permanent wedge in per capita fuel use between the United States and Europe. But in contrast to the situation in the household sector, all indica- tors of energy use and CO2 emissions from travel now point up (Schipper, 1995). This analysis shows that travel is emerging as the primary leader of growth in carbon emissions in the wealthy, industrialized countries. Life- style changes driven predominantly by higher incomesâparticularly in- creased automobilityâhave consistently led to higher carbon emissions, and the trends in the travel sector show no signs of saturation. Because the energy intensity of travel is scarcely falling, coupling between life-
60 ENVIRONMENTALLY SIGNIFICANT CONSUMPTION styles and emissions in the travel sector may cause difficulties for govern- ments intent on restraining or even cutting emissions. It is critical to understand not only how efficiently energy is converted to energy ser- vices but also how the levels of services are growing. ACKNOWLEDGMENT The author is a Staff Senior Scientist with the International Energy Studies Group, Energy Analysis Program, Energy and Environment Divi- sion, Lawrence Berkeley Laboratory, and currently on leave to the Inter- national Energy Agency (I.E.A.), Paris. This work was initially supported by the U. S. Environmental Protection Agency and completed at the I.E.A. The opinions advanced are strictly those of the author. REFERENCES Gershuny, J., and S. Jones 1987 Time Use in Seven Countries, 1961 to 1984. Bath, England: University of Bath. Howarth, R.B., L. Schipper, and B. Andersson 1993 Structure and intensity of energy use: Trends in five OECD nations. Energy Journal 14(2):27-45. Schipper, L. 1993 Energy Efficiency and Human Activity: Lessons from the Past, Importance for the Future. Report presented at the World Bank Development Conference. May 3-4, Washington, D.C. 1995 Automobile use and energy consumption in OECD countries. Annual Review of Energy and the Environment 21. Palo Alto, Calif.: Annual Reviews Inc. in People in the greenhouse: Indicators of carbon emissions from households and press travel. In T. Dietz, ed., Environmental Impacts of Consumption. Schipper, L., S. Bartlett, D. Hawk, and E. Vine 1989 Linking lifestyles and energy use: A matter of time? Annual Review of Energy 14:273-320. Schipper, L., R. Gorham, and M.J. Figueroa. 1996 People on the Move: Comparison of Travel Patterns in OECD Countries. Paper prepared for the United States Department of Transportation. Berkeley, Calif.: Lawrence Berkeley Laboratory. Schipper, L., R. Haas, and C. Sheinbaum 1996 Recent trends in residential energy use in OECD countries and their impact on CO2 emissions. Journal of Mitigation and Adaptation to Global Changes 1(2):167-196. Schipper, L., F. Johnson, R. Howarth, B.G. Andersson, B.E. Andersson, and L. Price 1993 Energy Use in Sweden: An International Perspective (LBL-33819). Berkeley, Calif.: Lawrence Berkeley Laboratory. Schipper, L., S. Meyers, R. Howarth, and R. Steiner 1992 Energy Efficiency and Human Activity: Past Trends, Future Prospects. Cambridge, England: Cambridge University Press. Schipper, L., L. Scholl, and N. Kiang 1996 CO2 emissions from passenger transport. Energy Policy 24(1):17-30.
TRACKING THE FLOWS OF ENERGY AND MATERIALS 61 Schipper, L., L. Scholl, and L. Price in Energy use and carbon emissions from freight in OECD countries. An analysis press of trends from 1973-1992. Transport and the Environment. Schipper, L., M. Ting, M. Khrushch, F. Unander, P. Monahan, and W. Golove 1996 The Evolution of Carbon-Dioxide Emissions from Energy Use in Industrialized Coun- tries: An End-Use Analysis. Berkeley, Calif.: Lawrence Berkeley Laboratory. Scholl, L., L. Schipper, and N. Kiang 1996 CO2 emissions from passenger transport: A comparison of international trends from 1973-1992. Energy Policy 24(1):17-30. Torvanger, A. 1991 Manufacturing sector carbon dioxide emissions in nine OECD countries 1973-1987. Energy Economics 13(2). BIBLIOGRAPHY Fergesson, M. 1990 Subsidized Pollution: Company Cars and the Greenhouse Effect. London: Earth Re- sources Research. Greening, L., W.B. Davis, and L. Schipper 1996 Decomposition of Aggregate Carbon Intensity for Manufacturing: Comparison of Declining Trends from Ten OECD Countries for the Period 1971 to 1991. Unpub- lished paper submitted to Energy Economics. Holdren, J. 1992 Prologue: The transition to costlier energy. In L. Schipper and S. Meyers, with R. Howarth and R. Steiner, eds., Energy Efficiency and Human Activity: Past Trends, Future Prospects. Cambridge, England: Cambridge University Press. Howarth, R., and R. Monahan 1992 Economics, Ethics, and Climate Policy (LBL-33230). Berkeley, Calif.: Lawrence Ber- keley Laboratory. Howarth, R.B., L. Schipper, P.A. Duerr, and S. Stroem 1991 Manufacturing energy use in eight OECD countries. Energy Economics 13(2):135-142. Schipper, L., F. Johnson, R. Howarth, B.G. Andersson, B.E. Andersson, and L. Price 1993 Energy Use in Sweden : An International Perspective (LBL-33819). 1992 Intergovern- mental Panel on Climate Change Supplement. Geneva, Switzerland: Intergov- ernmental Panel on Climate Change. Schipper, L., and L. Price 1994 Efficient energy use and well being: The Swedish example after 20 years. Natural Resources Forum 18(2):125-142. Schipper, L., B. Richard, R. Howarth, B. Andersson, and L. Price 1993 Energy use in Denmark: An international perspective. Natural Resources Forum 17(2):83-103. Schipper, L., B. Richard, R. Howarth, and E. Carlesarle 1992 Energy intensity, sectoral activity, and structural change in the Norwegian economy. EnergyâThe International Journal 17(3):215-233. Schipper, L., R. Steiner, M.J. Figueroa, and K. Dolan 1993 Fuel prices and economy. Transport Policy 1(1):6-20. Schipper, L., F. Unander, M. Khrushch, M. Ting, and L. Peraelae 1996 Energy Use in Ten OECD Countries: Long Term Trends through 1991. Berkeley, Calif.: Lawrence Berkeley Laboratory.
62 ENVIRONMENTALLY SIGNIFICANT CONSUMPTION Sheinbaum, C., and L. Schipper 1993 Residential sector carbon dioxide emissions in OECD countries 1973-1989: A comparative analysis. Pp. 255-268 in The Energy Efficiency Challenge for Europe: Proceedings of the ECEEE Summer Study, Vol. II. Oslo, Norway: European Council for an Energy-Efficient Economy.
TRACKING THE FLOWS OF ENERGY AND MATERIALS 63 STRUCTURAL ECONOMICS: A STRATEGY FOR ANALYZING THE IMPLICATIONS OF CONSUMPTION Faye Duchin LIFESTYLES AND THE ENVIRONMENT Interest in personal consumption is of long standing in economics, and many related aspects of household behavior have been addressed in all the social sciences. Consumption can be viewed as the motor propel- ling an economy in that producers will fabricate only the goods and ser- vices that consumers want to buy. Very recently, environmental concerns have reinvigorated social scientistsâ interest in consumption. Most envi- ronmental degradation can be traced to the extraction of fuels and other materials and their transformation to produce, both directly and indi- rectly, the goods and services valued by consumers. Clearly, changes surrounding consumption would alter, and could alleviate, pressures on the environment. There are basically two ways in which such changes could be achieved. First, the technologies used to extract and transform materials could be improved in various ways. Second, consumption patterns could change. There are many efforts under way to develop technologies that are more efficient in their use of energy and materials and that generate less environmental damage than current practices. In this paper our con- cern is especially with consumption patterns. This paper describes a con- ceptual and methodological approach for situating consumption activi- ties within a broad socioeconomic framework. It brings together various pieces of work that I have carried out over the past few years and fills in the missing pieces to make a relatively complete and coherent frame- work. A book-length manuscript that elaborates the major aspects of this approach has recently been completed (Duchin, 1996). Economists are concerned with consumption by individuals, but there are two compelling reasons to think in somewhat broader terms. First, an individualâs consumption behavior is tightly linked with his or her em- ployment, in that earned income has to cover outlays for purchased goods and services. Consumption behavior is also related to other peopleâs employment and consumption: if everyone stopped buying cars, auto workers (and, by a domino effect, many other workers) would soon be without jobs and income. Second, people live in households (including, of course, one-person households) that generally contain one or more paid workers. At least a portion of the income they earn is pooled, based
64 ENVIRONMENTALLY SIGNIFICANT CONSUMPTION on various kinds of negotiations, to pay for both common purchases and those of financially dependent individuals. A householdâs lifestyle refers to the jointly determined work and consumption practices of its mem- bers. THE LOGIC OF STRUCTURAL ECONOMICS There is a vast amount of literature indicating that consumer demand for specific goods levels off at higher incomes. However, prospects for overall saturation are far more ambiguous. Following J.S. Mill, a number of economists have expressed the view that once population levels off in affluent societies, other forms of satisfaction might be preferred to further purchases of goods and services (Mishan, 1967; Scitovsky, 1976; Hirsch, 1977), especially when people become aware of the environmental impli- cations (Boulding, 1973; Daly, 1977). None of these authors, however, was able to provide an analytic framework for integrating these phenom- ena with other economic activities. The ânew home economicsâ initiated by the work of Gary Becker, on the other hand, established the impor- tance of the household as a decision-making unit within the analytic framework of neoclassical economics. Household decisions are portrayed as maximizing the householdâs âutilityâ subject to budget constraints; the treatment is analogous to that of business firms concerned only with maxi- mizing their short-term profit (Becker, 1981). Input-output economics provides a foundation for the description and analysis of household lifestyles that is both firmer and richer than neoclassical economics. However, this approach needs to be substantially extended in its coverage of both households and the physical environ- ment. Structural Economics provides this extended framework. Input-output economics describes the structure of an economy in terms of the interdependence among its different parts (Leontief, 1986). In the dynamic formulation, changes in structure result from technologi- cal changes, the accumulation of stocks of physical capital, and the deple- tion of stocks of resources. The framework consists of two simple but extremely flexible mathematical modelsâa model of physical intercon- nectedness and a corresponding representation of costs and pricesâand a highly structured database. In neoclassical economics, a money price needs to be associated with every variable, a network of parameters called elasticities govern auto- matic substitutions among inputs whenever prices change, economic ac- tions are limited to the operations of competitive markets, and a solution requires that all markets are simultaneously in âequilibrium.â The power of these assumptions is that they assure unique, optimal solutions to com- plex problems. However, the problems that are solved are arguably not
TRACKING THE FLOWS OF ENERGY AND MATERIALS 65 FIGURE 3-17 A structural table of an economy. NOTE: A structural table incor- porates elements of an input-output table (n sectors), social accounting matrix (o occupations; h categories of household), and natural resource accounts (r re- sources, w categories of wastes). See text. SOURCE: Duchin (1995). the most useful representation of actual situations. An input-output solu- tion has the important advantages that it is not restricted to money val- ues: substitutions of one technology for another are governed by sce- narios rather than by formal mathematical expressions, and scenarios can reflect competitive behavior or behavior that is strategic, civic, or ethically motivated. Because many fewer kinds of assumptions are built into the formal framework, more burden falls on the development of scenarios and the interpretation of alternative outcomes. Structural Economics situates the detailed inter-industry relationships within a broader social context of household activities, which in turn are entirely contained within the physical environment. Figure 3-17 shows the way in which a structural table extends an input-output table. The household and environmental portions of the table draw on social ac- counting (Stone, 1986; Keuning and de Ruijter, 1988) and natural resource accounting (Central Bureau of Statistics of Norway, 1992; de Haan et al., 1993; Lange and Duchin, 1994), respectively. Mathematical relationships that deal with these extensions in a realistic way have been developed in a number of recent studies. A structural analysis starts by defining the questions that will be
66 ENVIRONMENTALLY SIGNIFICANT CONSUMPTION addressed and selecting or developing the mathematical model. A de- scription of various input-output models can be found in Duchin (1988). Then classifications for industries, households, and resources are es- tablished, and the âtransactionsâ among them are quantified for one or more historical years. National Accounts can provide the bulk of this information. The base year data are expressed as stocks or flowsâe.g., tons of coal absorbed by the steel industry. The model is formulated in terms of both variables (representing the stocks and flows) and parameters; the latter quantify the relationships among variables. An example of a parameter is the tons of coal required, on average, to make a ton of steel in the United States in 1992, given the mix of technologies in use at that time and the relative importance of each. The mathematical equations specify the kinds of parameters that are required. One or more scenarios are built for each of the questions to be ex- plored. An example will demonstrate how a scenario translates the ques- tion into variables and parameters. As part of an analysis of development strategies for Indonesia, we were asked what changes would be needed in agricultural technology for Indonesia to remain self-sufficient in food over the next several decades, while upgrading the quality of the diet for a growing population and being obliged to take some of the most fertile land out of food production (Duchin et al. 1993). This scenario required assumptions about changes in diet (i.e., consumption parameters) and in the yields of new agricultural technologies (i.e., parameters for the agricultural sectors). The computa- tion would determine how much land would be required (i.e., endog- enous variables) to support these assumptions. Input-output case study methodology has been developed for struc- turing the data projections (Duchin and Lange, 1994). Case studies for Indonesia, focused on the use of land, water, and energy, were carried out for households, forestry, rice, other food crops, estate crops, livestock, pulp and paper, cement, chemicals, food processing, textiles and apparel, and basic iron and steel. The computations showed that even the most optimistic assumptions about the adoption of advanced agricultural tech- nologies could not satisfy the land constraints and other requirements; food will need to be imported. CATEGORIES OF HOUSEHOLDS Standardized Industrial Classification (SIC) schemes for goods and services produced on farms and in factories and offices are in wide use. These classification schemes have made it possible to share data, compare across studies and across countries, and cumulate results. Standard Occu-
TRACKING THE FLOWS OF ENERGY AND MATERIALS 67 pational Classifications also exist, although they are less widely used. Classification schemes for households are more fundamental than those for occupations but are at a much earlier stage of development. Anthropologists and sociologists have provided detailed, qualitative description of specific categories of households; see Wilk and Lutzen- heiser (Chapter 4, this volume) for recent developments. Economists have established classifications that cover the entire society, but they are usually in terms of income categories only. An exception is the work done within the social accounting framework. Most Social Accounting Matrices (SAMs) have been constructed to examine the distribution of income in developing countries. A particularly detailed SAM (a flow table similar in structure to the industry and household portions of Figure 3-17) is the one for Indonesia, where households are classified according to urban or rural location, ag- ricultural or nonagricultural nature of the work of the âheadâ of the house- hold, and economic status, for a total of ten categories. This SAM also distinguishes four occupations and whether or not the workers are paid (Central Bureau of Statistics of Indonesia, 1990). The most promising kind of household classification scheme is one developed for consumer research and marketing based on a direct exami- nation of detailed data (by Jonathan Robbin; described in Weiss, 1988). Observing that people who share a âzip codeâ tend to have similar life- styles, Robbin built a database about U.S. household practices in each of these small areas; he included detailed information from the Census of Households, automobile purchase lists, credit card information, voting records, social values from surveys carried out at the Stanford Research Institute, and a host of specialized, private surveys. Robbin discovered that 34 variables accounted for almost 90 percent of the variation among neighborhoods. Each zip code was rated on these variables and assigned to one of 40 clusters, for which Robbin created a descriptive name. The resulting classification, which is widely used by corporations and politi- cal candidates to customize their messages for specific markets, is shown in Table 3-5. Research scientists may well be able to improve on these categories for the kinds of purposes envisaged in this paper. At the present time, my colleagues and I are designing classification schemes and building structural tables for several developing countries (Indonesia, the Dominican Republic, and Namibia) in collaboration with local researchers and the national statistical offices. The classification schemes are obviously very different from that shown for the United States in Table 3-5 but have been stimulated by its example. After this type of work has been done in several countriesâwith attention to using similar nomenclature for similar lifestylesâsome categories are likely to emerge that are common to a variety of societies. The most important
68 ENVIRONMENTALLY SIGNIFICANT CONSUMPTION TABLE 3-5 Household Classifications and Characteristics for the United States in 1987 ZQ Cluster Description 1 Blue Blood Estates Americaâs wealthiest neighborhoods includes suburban homes and one in ten millionaires 2 Money & Brains Posh big-city enclaves of townhouses, condos and apartments 3 Furs & Station Wagons New money in metropolitan bedroom suburbs 4 Urban Gold Coast Upscale urban high-rise districts 5 Pools & Patios Older, upper-middle-class, suburban communities 6 Two More Rungs Comfortable multi-ethnic suburbs 7 Young Influentials Yuppie, fringe-city condo and apartment developments 8 Young Suburbia Child-rearing, outlying suburbs 9 Godâs Country Upscale frontier boomtowns 10 Blue-Chip Blues The wealthiest blue-collar suburbs 11 Bohemian Mix Inner-city bohemian enclaves Ã la Greenwich Village 12 Levittown, USA Aging, post-World War II tract subdivisions 13 Gray Power Upper-middle-class retirement communities 14 Black Enterprise Predominantly black, middle- and upper-middle- class neighborhoods 15 New Beginnings Fringe-city areas of singles complexes, garden apartments and trim bungalows 16 Blue-Collar Nursery Middle-class, child-rearing towns 17 New Homesteaders Exurban boom towns of young, midscale families 18 New Melting Pot New immigrant neighborhoods, primarily in the nationâs port cities 19 Towns & Gowns Americaâs college towns 20 Rank & File Older, blue-collar, industrial suburbs 21 Middle America Midscale, midsize towns 22 Old Yankee Rows Working-class rowhouse districts 23 Coalburg & Corntown Small towns based on light industry and farming 24 Shotguns & Pickups Crossroads villages serving the nationâs lumber and breadbasket needs 25 Golden Ponds Rustic cottage communities located near the coasts, in the mountains or alongside lakes 26 Agri-business Small towns surrounded by large-scale farms and ranches 27 Emergent Minorities Predominantly black, working-class, city neighborhoods 28 Single City Blues Downscale urban singles districts 29 Mines & Mills Struggling steeltowns and mining villages 30 Back-Country Folks Remote, downscale, farm towns 31 Norma Rae-ville Lower-middle-class milltowns and industrial suburbs, primarily in the South 32 Smalltown Downtown Inner-city districts of small industrial cities 33 Grain Belt The nationâs most sparsely populated rural communities
TRACKING THE FLOWS OF ENERGY AND MATERIALS 69 % U.S. Median Home % College Household Income Value Graduate 1.1 $70,307 $200,000+a 50.7 0.9 45,798 150,755 45.5 3.2 50,086 132,725 38.1 0.5 36,838 200,000+a 50.5 3.4 35,895 99,702 28.2 0.7 31,263 117,012 28.3 2.9 30,398 106,332 36.0 5.3 38,582 93,281 23.8 2.7 36,728 99,418 25.8 6.0 32,218 72,563 13.1 1.1 21,916 110,669 38.8 3.1 28,742 70,728 15.7 2.9 25,259 83,630 18.3 0.8 33,149 68,713 16.0 4.3 24,847 75,354 19.3 2.2 30,077 67,281 10.2 4.2 25,909 67,221 15.9 0.9 22,142 113,616 19.1 1.2 17,862 60,891 27.5 1.4 26,283 59,363 9.2 3.2 24,431 55,605 10.7 1.6 24,808 76,406 11.0 2.0 23,994 51,604 10.4 1.9 24,291 53,222 9.1 5.2 20,140 51,537 12.8 2.1 21,363 49,012 11.5 1.7 22,029 45,187 10.7 3.3 17,926 62,351 18.6 2.8 21,537 46,325 8.7 3.4 19,843 41,030 8.1 2.3 18,559 36,556 9.6 2.5 17,206 42,225 10.0 1.3 21,698 45,852 8.4 continued on next page
70 ENVIRONMENTALLY SIGNIFICANT CONSUMPTION TABLE 3-5 Continued ZQ Cluster Description 34 Heavy Industry Lower-working-class districts in the nationâs older industrial cities 35 Share Croppers Primarily southern hamlets devoted to farming and light industry 36 Downtown Dixie-Style Aging, predominantly black neighborhoods, typically in southern cities 37 Hispanic Mix Americaâs Hispanic barrios 38 Tobacco Roads Predominantly black farm communities throughout the South 39 Hard Scrabble The nationâs poorest rural settlements 40 Public Assistance Americaâs inner-city ghettos National Median NOTE: The source document does not report the year for which the data apply or the price unit. The household percentages are based on 1987 data, but the values appear to be for 1986 in current prices. The table shows a median household income of $24,269; this com- pares with figures of $23,618 for 1985 and $24,897 for 1986, according to the U.S. Bureau of the Census, Statistical Abstract of the United States (1994), Table No. 707. The ZQ (zip qual- lifestyle changes in the developing countries surround those households whose work is unregistered and untaxed, and who are largely not reached by social services. Their ways of life are being rapidly altered by urban- ization and industrialization. The objective of scenario analysis in this context is to anticipate the nature and magnitude of these changes in terms, for example, of the future demand for education, health care, sani- tary facilities, or small loans. REFERENCES Becker, G.S. 1981 A Treatise on the Family. Cambridge, Mass: Harvard University Press. Boulding, K. 1973 The economics of the coming spaceship earth. Pp. 253-263 in H.E. Daly, ed., Economics, Ecology, Ethics. San Francisco: W.H. Freeman. Central Bureau of Statistics of Indonesia 1990 Social Accounting Matrix for Indonesia, 1985. Jakarta, Indonesia. Central Bureau of Statistics of Norway 1992 Natural Resources and the Environment 1991. Oslo, Norway. Daly, H. 1977 Steady-State Economics. San Francisco: W.H. Freeman.
TRACKING THE FLOWS OF ENERGY AND MATERIALS 71 % U.S. Median Home % College Household Income Value Graduate 2.8 18,325 39,537 6.5 4.0 16,854 33,917 7.1 3.4 15,204 35,301 10.7 1.9 16,270 49,533 6.8 1.2 13,227 27,143 7.3 1.5 12,874 27,651 6.5 3.1 10,804 28,340 6.3 $24,269 $64,182 16.2 ity) index, based on income, home value, education, and occupational status, measures socioeconomic rank. aThe upper census limit for home values is $200,000+; the figures for Blue Blood Estates and Urban Gold Coast are estimates. SOURCE: Duchin (1995), based on Weiss (1988) pp. 4, 5, 12, 13. de Haan, M., S. Keuning, and P. Bosch 1993 Integrating Indicators in a National Accounting Matrix Including Environmental Ac- counts. Netherlands Central Bureau of Statistics, No. NA-060. Duchin, F. 1988 Analyzing structural change in the economy. In M. Ciaschini, ed., Input-Output Analysis: Current Developments. London: Chapman and Hall. 1995 Global Scenarios about Lifestyle and Technology. Paper prepared for the Sus- tainable Future of the Global System conference, United Nations University, To- kyo, Japan. in Household Lifestyles: The Social Dimension of Structural Economics. Paper press prepared for the United Nations University, Tokyo, Japan. Duchin, F., C. Hamilton, and G. Lange 1993 Environment and Development in Indonesia: An Input-Output Analysis of Natu- ral Resource Issues. Final report for Indonesian Ministry of Planning. U.S. Agency for International Development and Canadian International Development Agency. Duchin, F., and G. Lange 1994 The Future of the Environment: Ecological Economics and Technological Change. New York: Oxford University Press. Hirsch, F. 1977 Social Limits to Growth. London: Routledge and Kegan Paul. Keuning, S., and W. De Ruijter 1988 Guidelines to the construction of a social accounting matrix. Review of Income and Wealth. Series 34. 1(March): 71-100.
72 ENVIRONMENTALLY SIGNIFICANT CONSUMPTION Lange, G., and F. Duchin 1994 Integrated Environmental-Economic Accounting. Natural Resource Accounts, and Natural Resource Management in Africa. Washington, D.C: Winrock Interna- tional Environmental Alliance. Leontief, W. 1986 Input-Output Economics, 2nd ed. New York: Oxford University Press. Mishan, E.J. 1967 The Costs of Economic Growth. New York: Penguin Books. Scitovsky, T. 1976 The Joyless Economy. New York: Oxford University Press. Stone, R. 1986 Social accounting: The state of play. Scandinavian Journal of Economics: 453-472. U.S. Bureau of the Census 1994 Statistical Abstract of the United States. Table No. 707. Washington D.C.: U.S. Department of Commerce. Weiss, M.J. 1988 The Clustering of America. New York: Harper and Row.