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.
1 Introduction, Conclusions, and Recommendations INTRODUCTION From 1973 through 1982 a number of the general trends that characterized electricity use between 1960 and 1972 showed distinct changes. The annual percentage change in electricity price, adjusted for inflation, reversed direction, from a decrease of 3.8 percent per year to an increase of 4 percent per year. The price of a unit of electric energy fell from about 7 to about 3 times the price of an equivalent amount of energy in the form of natural gas and heating oil, because of the rise in the price of fossil fuels. The rate of growth in electricity use dropped from 7 to 2 percent per year; and although formerly it had exceeded real growth rate of gross national product (GNP), the growth rate fell to approximately the same pace as that of GNP. Such changes have led to great uncertainty about the future relationships between electricity use and economic growth. It is therefore necessary to examine the forces that now underlie electricity use and to ask whether basic changes have occurred or are occurring, either in kind or degree. Two Important Relationships Two important relationships between electricity use and the economy are addressed in this report. One is how electricity use, or demand, in the usual sense of economics, depends on various economic and technical factors. The second is how electricity, as an especially high grade of energy, may facilitate technological advances, and in turn stimulate the economy, by providing gains in productivity. Our report does not address the question of electricity supply--that is, which generation technolog ies in what combinat ions should be used to serve demand . Ordinary experience suggests that electricity use should depend on the general level of economic activity, the prices of electricity and its alternatives, public policy, the regulatory environment, and the development of novel applications, among other factors. Let us look briefly at each of these. 1
2 o General economic activity is usually represented by GNP. For detailed analysis it is often important to disaggregate GNP into sectoral components and to make other disaggregations by geography and demography as well. In particular, there are important regional differences in providing and using electricity. o Electricity prices depend mainly on investment requirements for all types of plant and equipment, interest rates, fuel costs, and allowed rates of return. All these cost elements have varied more since 1972 than they did previously, generally leading to higher electricity prices. o The price of fossil fuels rose even more dramatically than that of electricity during the 1970s. Thus, the price of electricity was relatively attractive compared to available alternatives, leading to a growth in electricity use. On the other hand, since the cost of electricity generation depends in part on the cost of fossil fuels, the price trends made increased efficiency of electricity use, and sometimes avoidance of use, more desirable than before. The price trends also led to the search for less expensive ways of producing with available technologies and for other methods of generation, such as cogeneration, wind power, and solar electric power, in the hope of finding less expensive alternatives. o Public policy, implemented in various ways at various levels of government, can also influence electricity consumption. Some policies act directly, such as those that control prices or that provide incentives for conservation. Other policies act indirectly, such as import restrictions and tax preferences for research and development. Thus the effect of public policy may be felt through factors such as prices, regulations, and the growth of new applications for electricity. It is certainly possible to model the effect of any particular policy, but the net effect of many may be hard to estimate. o Regulatory constraints influence electricity use, usually through their effects on price. Regulation may encourage consumption by keeping prices down, or it may impose added costs to satisfy requirements for operational safety and environmental protection. Regulation may create other kinds of barriers to use by limiting the siting and construction of generating capacity. o Novel applications for electricity, such as the electric furnace for steelmaking, are often developed as one aspect of ongoing technical change of all sorts. Such applications are adopted when they lead to greater value of output than their incremental cost. The new applications can increase electricity use through substitution for other fuels, as in induction heating, or they can decrease electricity use through greater energy efficiency, as in the replacement of arc welding by electron beam welding. Again, since 1973 several factors affecting electricity use have undergone noticeable changes. For orderly planning by many sectors of the economy, it is important to know whether these changes will modify long-term trends connecting electricity use and the economy. Is a different form for the relationship more appropriate by introducing new
3 variables? Or is only some adjustment needed, large or small, in the numbers that connect the customary variables? If we address these questions so as to permit informed judgments about past relationships and future prospects, we shall have gone as far as we can. The future values of the economic variables themselves as inputs to consumption models are largely unpredictable. To estimate them requires not only skill but also-.ucK. Understanding the relationships between electricity use and the economy, with regard to both consumption and productivity, is important in formulating public policy, in regulating the industry, and in managing individual firms. When the existing system of supply and demand works well, producing acceptable economic and social benefits, there is little need for government intervention beyond the usual activities of the state and federal regulatory system. A sound system of monitoring and evaluation is all that is needed. The existing system may, on the other hand, not work well. For example, the economic costs to a region, or to the nation, of a shortfall of electric power may be out of proportion to the cost of adding capacity, even though an individual utility may find it financially inadvisable to build a new plant. Also, though consumption may be satisfied, or even reduced, under a current set of conditions, aggregate productivity benefits from increasing efficiency or adding capacity may exceed the costs of these steps. In this context, recall that because of regional variations in electricity production and use, the adequacy of supply may vary regionally in ways not reflected by aggregate national data. On the other hand, it may prove more economical to reduce electricity use or to slow its growth, by increasing the eff iciency of its use and substituting other production factors, than to expand supply. In any case, public policies may help ameliorate problems in the system by means of legislation, regulatory changes, investment incentives, or stimulation of research. With regard to regulation, better knowledge of the relationships between electricity use and the economy, concerning both aggregate and particular end uses, should facilitate better decisions on many issues: rate design, capital investment, required reserve capacity, fuel contracts, cost recovery rules' and admissible research costs. With regard to individual f irms, the decisions of utilities as well as their suppliers would of course benefit from a better grasp of the interactions at work. At stake for the utilities and, by extension, their investors, are the consequences of allocating funds among additional plants, load management equipment, and conservation measures. Suppliers must anticipate the kinds and amounts of plant equipment and fuels that futilities will need. The Structure of the Task The committee's task, stated more fully in Appendix A, was to look at the role of "electricity" in "economic growth. " Both these terms are
4 familiar but, as in using many terms, we should strive for precise def inition and connotation. Economic growth is conventionally measured by GNP. Electricity service is measured either by installed capacity to deliver power or by energy consumption. None of these measures expresses everything of importance for the relationships at issue, as is discussed further in Appendix B. Nevertheless, we concluded that these measures were generally those appropriate for our study because of the difficulty of establishing other adequate ones and the desirability of relating our own to previous work. Another term used in the report is "electrification." Electrification means the adoption of processes and activities based on the use of electricity. The term connotes an application and associated equipment that use the special qualities of electricity, often for innovation. Electrification may increase or decrease electricity consumption, depending on such factors as whether there is a change from a nonelectrical to an electrical production technique, the amount of electricity consumed per unit of output, and the total units of output produced. The last can be substantially greater than before if product prices fall because of more eff icient production. The terms "productivity" and "productivity growth" are used in their usual economic senses. Productivity means output per unit of input, measured in appropriate units, whether for a single kind of input or for a combination of inputs. Productivity growth is the change in product ivity f ram one point in t ime to another, usually expressed as a percentage. "Productivity growth rate" is productivity growth per unit of time. Figure 1-1 i llustrates the system that we examined. The complexity of the f igure reflects that of the real situation and thus the complexity of any useful analysis. We sought to describe the relationships between the central elements in this diagram: electrif ication, productivity growth, GNP, and electricity consumption. We tried to summarize what is currently known about these relationships and to indicate some uncertainties. To do so we had to consider several additional factors. These factors include the prices of electricity and of substitute fuels and the costs of electrical and nonelectrical processes. We considered direct effects (for example, the price of electricity on electricity consumption) and also indirect effects (for example, the income effect of conservation and electrif ication, f reeing resources for other uses). When possible we tried to quantify these relationships when referring to the past; we could discuss them only qualitatively when referring to the future. The various chapters focus on different parts of Figure 1-1. Chapter 2 discusses the historical relationship between electricity consumption and GNP. Electricity consumption is first analyzed as a function of GNP. Gross product originating (GPO) and disposable personal income (OPI) are then used as measures in a finer analysis to reflect economic activity in different sectors. The central subject of Chapter 2 i s depicted in the right central part of Figure 1-1: the arrow between GNP and electricity consumption points to the right
5 Regu ration ~ Technolo~ | Electrification Productivity Growth Commercial ~ Price of Substitute Fuels Residential 1~ Income Eiectricity Using Devices ~;= ~ Product -~ Consumption ~ I nd ustria I SUPPLY D EMAN D FIGURE 1-1 Relationships affecting electricity and economic growth.
6 because we consider how the level of economic activity affects electricity consumption. A later part of Chapter 2 discusses available evidence on how both the price of electricity and the price of nonelectric (substitute) fuels have affected electricity consumption. The influence of price links the left (or supply) side of the diagram with the right (or demand) side. Chapter 3 looks at the role of electricity in productivity growth. Productivity growth is one of three inputs to economic growth, the others being capital and labor growth. This chapter finds that the electrification of productive processes, as one type of technical change, may have special effects on productivity growth. As is shown in Chapter 4, these effects are due to the flexibility and high quality of electricity as energy in application. Chapter 3 demonstrates that the prices of both electricity and other fuels, along with their technical characteristics, do influence productivity growth in most industries. The relationships among these variables are depicted in the left and central parts of Figure 1-1. Chapter 4 looks at examples of the influence of electrification on economic activity, emphasizing further technical potentials. This chapter discusses the characteristics of various forms of electrification, their potential engineering and economic effects on general productive efficiency, and whether they may result in a net increase or decrease in electricity consumption per unit of output. This discussion is illustrated by some examples of technical change in different sectors of the economy. The examples show how electrification affects efficiency for the process and the firm and, by implication, how these gains can provide productivity growth in the aggregate. Chapter 4 is represented by the lower left corner of Figure 1-1. ~ In Chapter ~ we return to the subject of Chapter 2, but with attention to the future. Given the current uncertainty about whether the recent relationship between electricity use and GNP continues as before, we cannot foretell a precise future relationship between these variables. Furthermore, we are in no position to forecast the future growth rates of GNP and prices. Even so, we can consider the forces likely to influence the relationship between GNP and electricity use, for example, changes in the composition of national output, the prices of electricity and nonelectric fuels, and energy conservation. We also note that these forces have both direct and indirect influences. For instance, the individual consumer who realizes the benefits of conservation will enjoy greater disposable income. Similarly, if future electrification leads to greater overall productive efficiency and to lower electricity use per unit of output, this indirect effect on disposable income should lead to greater expenditures on goods and services. In turn, this would increase electricity demand according to the strong correlation that has held between GNP growth and the growth in electricity use. Thus, Chapter 5 addresses the possible future relationships among most of the elements of our diagram, but in a qualitative way.
7 These chapters together provide the information and analysis on which we based our conclusions. For the reader's convenience, our principal conclusions, along with statements of our rationale, appear below, and then our recommendations, which flow from these conclusions. CONCLUS IONS Electricity Consumption Electricity use and gross national product have been, and probably will continue to be, strongly correlated. Numerous representations of the relationships between economic activity and electricity use are possible.* We considered a number of these representations and discuss some in the report. The relationship we emphasize is based on considerations of simplicity and of adequacy in satisfying the committee's task. This relationship is at a high level of aggregation (between total electricity use and GNP) and takes a simple functional form (linear) such that one principal variable, GNP, is capable of explaining much of the variation in the other, electricity use, as they both change with the passage of time. In this century, the electricity use-economic activity relationship has been characterized by four well-defined periods. Within each period, the relationship has been linear and stable. The first period was prior to World War I and the second from the end of World War I through the 1920s. In the third period, from 1930 through the end of World World II, the linear relationship paralleled that of the 1920s. The fourth period began after World War II and may still be continuing, although the relationship holding after the 1973 Arab oil embargo is still in dispute. We cannot tell conclusively whether the relationship after 1973 simply reflects variations from the most recent trend line, such as have occurred before, or whether a fundamental change in the relationship is taking place. . . *The possibilities encompass various aggregations of data (by nation, sector, region, or household), various functional forms of the relationship (linear or another form of growth curve), the addition of potentially relevant variables (population and household data, prices, inventories of electricity-using equipment, labor force data, and time), and various transformations of variables from their "natural" units (for example, kilowatt hours and dollars) to other forms, such as annual percentage increases. What we know does not allow us to disentangle the individual roles of all the possibly relevant variables in the electricity-economic activity relationship. Moreover, obviously no one formulation is best for all purposes. For example, a linear formulation may best illustrate the long-term historical record, while certain logarithmic representations may best serve more specific analytical needs.
8 Historic trends in the electricity use-GNP relationship include the effects of a host of factors not explicitly identified in the linear equation representing it. Among those believed important are the prices of electricity and competing energy forms, the composition of national output, regional economic activity, technical change, conservation practices, and government policies. Only when there are major perturbations in the trends (not simply movements about the trend lines) of these underlying variables would changes in the basic electricity use-GNP relationship be expected. Even then, some effects may cancel each other (such as rises in both electricity and other energy prices). Lesser variations in these underlying variables produce temporary deviations from the electricity use-GNP relationship. Two forces believed capable of altering the trends of future electricity use-economic activity relationships are electrification and conservation. However, their potential effects, like those of the other underlying variables, are not readily quantified. Electricity and nonelectrical energy prices are generally acknowledged as factors determining electricity consumption. However, by far the most important contribution to explaining consumption in the past has been GNP. The observed departure on occasion of electricity consumption from the main trend line may be explained in part by taking price changes into account. Furthermore, there is an implicit dependence of electricity consumption on energy prices through the dependence of GNP in part on productivity g rowth, which in turn is shown to depend partly on energy prices. Productivity Growth Productivity growth may be ascribed partly to technical change; in many industries technical change also tends to increase the relative share of electricity in the value of output, and in these industries productivity growth is found to be the greater the lower the real price of electricity, and vice versa. Economic growth, or percentage change in GNP, results from growth in three factors: capital input, labor input, and productivity. Productivity accounts for increases in output in excess of the contribution of the first two factors. Productivity growth for the economy as a whole derives mainly from sectoral productivity growth and any reallocations of value added, capital input, and labor input among the sectors of the economy. The decline in the rate of U.S. economic growth since the early 1970s is associated with a decline in sectoral productivity growth rates rather than a reduction in the aggregate growth rate of capital and labor inputs or the reallocation of value added, capital input, or labor input among sectors. Chapter 3 shows that this sectoral decline in productivity growth is strongly associated with an increase in energy prices.
9 Sectoral productivity growth may be modeled as a function of the relative prices of major inputs--capital, labor, electricity, nonelectric energy, and materials--and the level of technology e One comprehensive model of this sort shows that technical change and reduced electricity prices have a related effect on many industries. First, for these industries technical change is electricity using; that is, it is found to increase the contribution, relative to those of other inputs to production, that a given change in electricity input value makes to change in output value and so tends to increase the relative share of electricity in the value of output. Second, for the same industries the productivity growth arising from the technical change increases as electricity prices decrease, and conversely. Such effects are found for nonelectrical energy in even more industries. The decline in the real cost of electricity, which resulted in part from dramatic increases in the thermal efficiency of electric generation, increased electricity use and stimulated productivity growth until the early 1970s. The reversal in the decline of electricity costs, combined with a rise in the prices of primary fuels after the international oil price increases of 1973 and 1979, has permanently reduced productivity in many industries from what it would otherwise have been. This result may be explained partly by the substitution of less efficient inputs for these energy inputs. Technical Change Technical change has made possible many new opportunities for exploiting the special qualities of electricity. In the past these changes were of ten associated with increased intensity of - electricity use, but in the future their net effect on that . intensity will depend on the balance between their increased penetration and the increased efficiency of these applications. Once generated, electricity has unique properties that make it an attractive form of energy. These properties include a highly ordered form (including the ability to be focused for efficient use and to produce very high temperatures), flexibility, and cleanliness of use. There is substantial potential in the major consuming sectors for further applications of electrical energy that take advantage of these special properties. We call such innovations electrification. There are several different forms of electrification: (1) changing either old or new processes so that they rely on electricity rather than on fossil fuels, or direct wind or water as a source of mechanical energy, or human labor--changes generally associated with an increase in the intensity of electricity use; (2) converting older electrotechnologies (such as motor drive in manufacturing) to advanced ones to meet end-use requirements better--changes that often increase economic efficiency and may either increase or decrease the intensity of electricity use; and (3) the rapid penetration of new activities that depend on electricity, such as the growth in the use of
10 computerized techniques--changes that may either increase or decrease the intensity of electricity use. Some kinds of electrification increase electricity intensity (electricity use per unit of economic output) through wider application of electrical processes. Some kinds decrease electricity intensity through productivity gains. In the aggregate it is found that the increase in electricity intensity with GNP is relatively small because the two effects tend to be offsetting. Electrification can change not only the form of energy used but also the share and absolute quantity of other inputs, including labor, capital, and materials. In addition product quality and even manufacturing location can be affected. Technical change in the form of electrification has historically contributed to increased productivity and thereby to increases in GNP. We can expect this trend to continue. The Effects of Price Changes Electricity prices and alternative fuel prices affect electricity consumption in two ways: first, they directly affect the use of electricity and nonelectric fuels as input factors of production; second, they indirectly affect productivity growth and thereby economic growth. If electricity prices alone rise (for example, because of a rise in plant and equipment prices), electricity use will decrease in accordance with elasticity of demand with respect to its own price. This result will occur through improving the efficiency of electricity use and through substituting other inputs for electricity. A rise in the price of those fuels that compete with electricity, without a corresponding increase in the price of electricity, will increase electricity consumption because of elasticity of demand with respect to the prices of other fuels. If electricity prices rise because of a rise in primary fuel prices, a reduction in electricity use through its own-price elasticity will occur and be offset to some degree by an increase in the use of electricity as a substitute for primary fuels, that is, through cross-price elasticity. The numerical values of these elasticities have not been well established, but current estimates of price elasticities suggest that the two effects may cancel each other. Any increase in the real price of electricity will indirectly further decrease electricity use because it will lower productivity growth rates in many industries, in turn leading to a lower rate of economic growth. Reductions in electricity prices yield an opposite set of results, as indicated historically.
11 Conservation There is further potential for increasing the efficiency of electricity use, particularly in the residential and commercial sectors. Promising technologies have been identified for increasing the efficiency of electricity use. Their greatest promise is in the residential and commercial sectors, where there has been less investment in efficiency improvements than in the industrial sector since the Arab oil embargo. Energy price increases provide incentive for investments in conservation. The main constraints to such investment have been immaturity of the technologies, lack of information, lack of capitalization funds, inefficient electricity and fuel pricing, and doubts about the cost-effectiveness of such investment. Of particular interest for the residential and commercial sectors are potential improvements in building envelopes and lighting systems, which can be incorporated in new construction and retrofitted to existing buildings. Although these improvements may themselves reduce the intensity of electricity use, there may be other factors, hard to predict, that increase electricity consumption through new uses of electricity in production and household applications. In addition, many established uses of electricity, such as for air conditioning and electric space heating, still show potential market growth. The effects of residential conservation investments do not show up directly in sectoral productivity measures. However, their macroeconomic effect may be evidenced in a change in the composition of sectoral output and in changes in consumption f ram the income effect of reduced energy costs. In the commercial and industrial sectors those conservation measures that are cost-effective would appear in measures of sectoral productivity g rowth. Evidence of success in conservation and load management is provided by programs implemented by electric and gas ''tilities. Conservation and load management, if cost-effective, can also benefit economic growth by reducing the costs of electricity supply, and thus the price of electricity, through improving the efficiency of existing and new generating facilities in producing given levels of electric energy. The Composition of National Output Changes in the composition of national output toward less electricity-intensive goods and services have been offset by growth in the intensity of electr ic ity use within all the ma jor use sectors so that the combined ef feet on electric ity demand g rowth has not yet been great. However, if the trend toward a leveling off in sectoral electricity intensity growth that began in the late 1970s continues, future shifts toward less electricity-intensive goods and services are likely to dampen electricity demand growth relative to national output.
12 Looking at the sectoral composition of national output is one means of analyzing structural changes in the economy. Gross product originating (GPO) in producing sectors is often used to measure and compare their output. Employment figures are also widely used, but they are not as useful as GPO in analyzing the relationships of electricity use to other factors since they account for only one of the inputs to sectoral output. Since 1950 the share of GPO in the commercial sector has increased steadily, while that in the industrial sector as a whole has declined. This decline is almost entirely due to a decrease in the relative importance of agriculture, mining, and construction as components of the industrial sector. The share of manufacturing GPO remained fairly constant over the entire postwar period, although within manufacturing there has been a shift toward less electricity-intensive industries. The electricity intensity of the industrial sector is about three times that of the commercial sector, so that shifts away from industry, all other things being equal, would lead to a decline in electricity intensity for the total economy. However, there were large increases in average electricity intensity in all three of the major consuming sectors after World War II, which more than counteracted the negative influence on overall electricity intensity of the shift from industrial to commercial output. Almost all the growth in average sectoral electricity intensity occurred prior to 1973; by 1983 industrial and commercial sector electricity intensities were back near their 1973 values, while residential electricity intensity remained stable from about 1977. It is uncertain whether recent declines in sectoral electricity intensity growth represent the beginning of a new long-term trend or a response to short-term influences. Regional Differences Valid conclusions about electricity demand drawn from national data do not necessarily pertain to regional circumstances; there are significant regional differences in such factors as economic output, prices, electricity supply mix, availability of generating capacity, climate, and regulatory environment. With regard to economic activity, the regional factors important to electricity consumption include overall level of output, industry mix, labor and resource availability, and the relative importance of a region's commercial and industrial sectors. With regard to energy use, important regional factors include electricity and nonelectric energy prices, electricity supply mix, climate, and regulation. Shifts in demographic characteristics and regional activity may alter national electricity use patterns, although probably gradually and in a small way. National policy decisions should be sensitive to important regional differences.
13 RECOMMENDAT IONS The principal focus of this study is a better understanding of the complex relationships between electricity use and economic growth. Two important conclusions underlie the recommendations that follow. there has been a strong correlation between the use of First, _ electricity and the magnitude of GNP. Second, the recent research described in Chapter 3 was judged sufficiently significant to put forward with some confidence its thesis that there is a strong connection between electricity and productivity growth. 1. The relationship between electricity and productivity is so important that it should be considered in developing federal and state energy and economic policies. Productivity growth is central to solving many problems facing the United States, ranging from the federal deficit to the balance of trade. Consequently, all possibilities of ~ , ,_ _ ~ growth, including attention to electricity supply and use, should be evaluated and pursued in accord with their promise. * * * * * At ~ ml1 1 at i no nrn~lil~t. i wi EN 2. To foster increased productivity, policy should stimulate increased efficiency of electricity use, promote the implementation of elect~otechnclogies when they are economically justified, and seek to lower the real costs of electricity supply by removing any regulatory impediments and developing promising technologies to provide electricity. The findings of this report establish a connection between electricity and productivity growth. ~_ . . ~. . The two factors that must coexist to realize the productivity growth associated with electricity are technical change and favorable electricity supply conditions. In addition, cost-effective increases in the efficiency of electricity use will themselves not only increase productive output for a given input , . . . .. , . ~ . . , ~ These points suggest that federal and state policies that promote lowering the real costs of electricity supply and use, through research and development or through more efficient pricing by regulatory authorities, will benefit productivity growth. or elects 1C lay out also tree Income tor other Purposes. * * * * * 3. Further research should be undertaken to identify and quantify the forces affecting the relationships between electricity and economic growth in view of their critical importance, complexity, and regional diversity.
14 The strong and persistent relationship between electricity use and GNP requires that close attention be paid to the adequacy of electricity supply to sustain a high future rate of economic growth. The adequacy of electricity supply can be maintained not only through new generation facilities but also through efficiency improvements that use existing generating capacity better. Although favorable electricity supply conditions of themselves will not assure economic growth, a lack of adequate supply would almost certainly constitute a serious impediment to such growth. In making this point we are keenly aware of the need to learn more about the correlations and the causal relationships between economic growth and electricity use. As pointed out above, well directed policy, regulation, and management decisions rest on such knowledge. It should be systematically sought and better established. Otherwise progress toward greater economic efficiency, innovation, and competition may suf fer .