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.
Appendix B Measures of Economic Growth and of Electricity Demand One focus of this work is a better understanding of the relationships between economic growth and electric ity demand . Economic growth is conventionally measured in terms of gross net tonal product (GNP), a measure of the output of the so-called productive economy, as distinguished from the residential sector, the output of which is not included in GNP. Yet a large part of electricity use occurs in households. Should electricity use then be cons idered here only for the productive economy to ensure the comparability of f igures? This treatment is usually not adopted because the income generated by the productive economy is used for household expend itures . In GNP each unit of electricity is valued at its price, for example, at 7 cents per kilowatt hour. Yet electricity certainly gives rise to costs external to its price, such as air pollution. It Night be argued that these costs ought to be subtracted somehow f rom GNP. On the other hand, some tasks, such as instant communication between cont inents, could not be accomplished at any price without using electricity. Thus, electricity's special characteristics give rise to various positive and negative effects external to its price, not captured by the GNP i nd icator . Similarly, conventional measures of electric service do not capture everything of importance for our purposes. Electric service is measured in terms of installed capacity to deliver power or of energy consumption. Electricity's full usefulness, however, lies in effective end-use service to customers--warmer homes, cooler office buildings, and many other applications--from electrically powered equipment. As discussed elsewhere in the report, a few of the nation's larger Futilities are making major expenditures to increase the effective delivery of electric services to their customers without increasing either installed capacity or energy consumption as usually measured. The expenditures are for conservation and load management technologies, rather than for new power plants. Because these expenditures are being made by public utility companies expressly to expand the ef fective use of electricity, the conventional measures of electric service that a utility renders do not tell us all we need to know about the service's 13;
136 value. In other words, conventional measures of electricity may understate the actual amount of effective service provided. We considered these measurement issues early and decided to rely on conventional measures of both electricity and economic growth because of the difficulty of doing otherwise and the need to conform with prior work. We simply recognize that the conventional measurements can sometimes be incomplete in signif icant ways. We must also distinguish two uses of the word "demand" to avoid confusion. In the economic sense demand is the quantity of a commodity or service wanted at a specific price and time. Thus electricity demand, as related to economic activity, is usually measured in units of electrical energy, conveniently kilowatt hours. This concept is not the one most natural to utilities. Utilities think of electricity demand as the instantaneous load on the utility system, measured in units of electrical power that must be supplied at that time, usually kilowatts. Utilities think of electricity consumption over time as energy, measured in kilowatt hours. We make this distinction because the vast majority of analyses that relate electricity consumption and economic growth focus on kilowatt hours. These studies do not consider the availability and cost issues of having sufficient capacity to meet demand in the sense that utilities define it. To understand the relationship between electricity consumption and economic growth we have to keep in mind that each kilowatt hour consumed can have a different availability and cost to the utility (and possibly a different price to the consumer) reflecting the time and way it is produced. Availability and price will also vary with other features of the utility providing the electricity--its plant mix, fuel mix, reserve margin, and so on. Ideally we would look at the relationships both between growth in instantaneous electricity demand (kilowatts) and that in GNP and between electricity consumption (kilowatts hours) and GNP. Although the former has received relatively little attention f rom economists, it remains a central issue for the companies providing the electricity and for utility regulators. The distinction between kilowatts and kilowatt hours is important because electricity cannot be economically stored. Thus, providers of electricity, such as electric utilities, must have enough generating capacity available to meet the aggregate instantaneous demand on their systems that may result from the combined demands of each of their customers. The efficiency with which a utility's generating capacity is used is a function of its load curve, that is, the hourly distribution of aggregate demand. If there is a substantial difference between average power supplied over a long term and peak power capability, then the utility's plant may not be used very efficiently. Thus, it is important to consider both measures if one is interested in adequacy and efficient allocation of resources. The time pattern of growth in electricity demand (in kilowatts) is not the same as that of the growth in electricity use (in kilowatt
137 hours). The Edison Electric Institute's (EEI) 1983 Electric Power Annual Report shows that from 1979 to 1983 the average growth per year for summer peak demand in the United States was 2.9 percent per year and for winter peak demand 2.7 percent per year, while the average growth in electricity use per year during the same period wars only 1. 0 percent per year . These data show that, s ince 1979, peak demand has on average grown faster than electricity use. However, including estimated 1984 data would make a large difference, because they tentatively show an increase in kilowatt hour use of 4.5 percent, with only 0.7 percent increase in noncoincident summer peak demand (winter data are not yet available) . Without the ` 1984 data one might be inclined to infer that peak demand has begun to grow a good deal faster than electricity use, suggesting a deterioration in utility load curves and thus less efficient use of available generating capacity. Table B-1 does show ~ fairly steady decline in annual load factor since 1963. However, the 1984 data show an increase in load factor that is consistent with the greater increase in kilowatt hour use. It is too early to tell whether the divergence of kilowatt and kilowatt hour growth is a trend or not. Electricity demand, like economic growth, shows significant regional variation. Table B-2 is a summary of data from EEI's 1983 Electric Power Annual Report. The map in Figure B-1 shows the geographical boundaries of the electric reliability council regions, which are subsets of the National Electric Reliability Council (NERC) , used in EEI's planning and analysis. The summer peak demand compound growth rate for 1979 through 1983 varied f ram 0. 7 percent per year in the MAIN region (Illinois, most of Wisconsin, and part of Missouri) to 5.9 percent per year in the ERCOT region (most of Texas) . The next greatest area of g rowth was the southeastern United States. Winter peak demand also g rew most rapidly in Texas and in the Southeast. Note that summer and winter peak load growth d if fer in most regions . (The line in this table designated "Entire United States" should be viewed with great caution. EEI notes that "U.S. totals are meaningless and should not be used for policy purposes.") Growth of electricity use also varied greatly by region, ranging from an increase of 3.6 percent per year in Texas to a decline of I.0 percent per year in the ECAR region. There is not a strong relationship between growth of peak demand and that of energy use except that in each region peak demand growth exceeded the increase in energy use over this period. However, recall that in 1984 energy use grew much faster than peak demand. Unfortunately regional data are not available for 1984. The aggregate 1984 f igure suggests that we should be cautious about inferring a relationship between peak demand growth and energy use demand growth between 1979 and 1983, since including 1984 data would signif icantly change any relationship inferred.
138 c - x :3 c JJ - .,, v ::) - . u 4 E" o UJ ~o c a rs: 3 ~ _. o ~4 .,. :d ~5 c ~ ~r O (D c~ ~ O P4 ~ ~o >' cr - ._. _ ~ 3 _ c 1 to m ~ U' C) ~s: ~ ~: ~ 0U 0 ~ ~m~ t4 t~ _ ~ 0 ~ 3 JJ p. c ~ c ~ c Q o ~ _ ~ eq ~ o ~ C :e C O . - E Ql :' ~ :~: 3 0 c c _ ~ _ ~ x ~ _ c =~ _ ~ _ v c ~ c ~ 0 ~ Z JJ c c ~ a' o _ _ _ Q. c a, ~ ~ O Q C) ~ c' c Y ~1 . - ~ ~ 3 `: ~ .,4 ~, _ O C c ~ O ~ 0 Z 3 81 ~ _ 3 v x: c - :~ v :~ . - ~ O -4 0 0- U E~ c O O C a, ~ "c ~ ~O ~4 c ~ c ~ a ~ ~, c ~ ~r a, 0 ~ :' a _ C C) C C 00 ~_ ~ ~ ~ic ~a E ~ v <' _ P. c u, .,, ~ o~ v ' ~ O O _ to 3 O 0 ~ t) · - ~ o, S C X aC ~:>, v ~ ~1 a _ ~ O 3 ~= 0 1 c ~ c ~ ~, a, u ~dJ c ~c ~o c o~ ~ ~ ~0 ~ ~J ~: ~ c ~v :> ~ ~ ~ ~ ~c ' a, ~ c o~ 0 a, ~ u' c~ 0 ~ a, O u: c :' ~ IJ ~ 3 c ~·~ o. a, 0 ~ O ~ ~. - . - 4 ~ ~O O u, s u' ~c 4J ~ :> ~ ~ ~ U ~0 c :~ a~ ~s c v u 3 tJ . - G1 Ql S ~5 C {0 _ ~ ~ O U) S l O O ~ :> _ ~1 0 3 0 ~ O ~ D' y ~ ~- - ._4 a~ ~ c ~a ~ ~ ~ :, ~ ~c c . - ~s ^ ~O x ~> ~ ~ o, e m_ .... C~ s . E" ~ ~ ~ C ~ c~, ., . ~? · ~ Qd ~ u: C a, ~ ~ ~ · _ ~ 0 ~ ~ ~ ~ ~ c ~ c e ~ ~ ~ ~ ~ ~ ' :' ~ ~ u' ~ e ~ ~ ~ ~ " a' ~ ~ E ,ca, 0 ~ c~- s ~ O t) U 3 1 v a ~ ~ ~ c ~ ~ e ~ on~ ~ ~ ~ ~ ~ 0 V V C ~ 0 C C ~ s ~ 3 c ~ .- ~ ,~ ,~ ' - ' - ~ ~ O c _ D ~ v ~ ~ o ~ ~ ~ ~ ~ ~ _ ,C!4 v c~ ~c ~ to (~ ·~ c) ~ ~ ~ s - c: ~1 ~ ~ ~ ^1 V ~ o1 _ ~ c ~ c co,, 0: s JJ ~ O ~ . 0 ~ ~ ~ LJ ~ .~4 ~ :>, :, 'E S ~ _ ._, O D ~ D~ ._~ ~S V ~ a, C _ _ D ~ 3 CO, C ~ ~ _ 0 - 3 S ~ ~ V _ ~ ~a C ~ ~ L. O ~ 'D O ns 0 c ~ ~Q ,4 a. ~J2 3 ~ c c 3 >, ~ v ~. c ~ 0 c) E _4 _ ~ D ~u] q ~ ~ ~ ~ ~V O U) C ' O a, v D ~_ C ~ ~U' C _t ~ ~ - c ~ a) · - ~ ~ O ~ ~JJ V ~ C ~ C ~' O C ~ O S ~ E v - c ~ O O E U) C - .,, . - X ~ C C C . - ~ ~ ~O E C u~ , . - ~· ~ v · o~ ~ · ~ ~ ~ c C JJ · - ~ LJ . - Q5 C Q' .. c ~ -- . - ~ . - ~ ~ v ~ ~ ~ ~ ~ ~c ., ~ ~-1 cl =1 ~ v u] c ~ - ~ JJ c) .-. v a, s v ~4 o
139 TABLE B-2 Compound Growth Rate in Peak Loads and Energy Requirements, by Region, 1979 to 1983 (Percent per YearJ Summer Peak Winter Peak Annual Growth in Reg ions Load Growth Load Growth Energy Requ i remeet s NPCC 2.1 0.9 1.1 MAAC 2.3 2.0 1.2 ECAR 1.7 1.7 -1. 0 SERC 4.0 2. 6 1. 7 MAIN 0.7 0.6 -0.8 SPP 5.1 6.1 3.2 ERCOT 5.9 8.5 3.6 MAPP 3.5 ~2.2 1.5 WSCC 2.0 2.5 0.6 Entire U.S. 2.9 2.7 1.0 aSee Figure B-1 for explanation of region acronyms. SOURCE: Edison Electric Institute, 1983 Electric Power Annual Report, pp . 7-10 (data f rom National Electric ity Reliabi lity Counc ill .
140 ~ ; ~ r -~ ;: ~ ~ I 1 I r~ a: C: · - r-, . - o C: · - . ~ rat · - a: J pi 8 a: · - ,~ . - ~ c. 03 .,, o ~ · - ¢5 s · - o Z ~ o s ., o . . in o . - .,' ~ a, ,0 . 1 ~ m c - V U] Ed · - 0 ~ ~ O ~ 3 O · - P4 O O ~ 3 .,' ~S 'o · - ~ ~ 3 0 O U] 1 7:5 ._, a: IJ a: ·` .- _ :5 o ·_4 , ~ a: 1 ·~4 ~: ~: · - ~2 x · E~4 ~u c · - ~ o u] ~ cn s ·' ~ · - o ~ z ~ ~5 ~ o : c) z ~ · - · - c.) ·` · - ~ Q O O ~ ~ C a: · - 3 O ~ ~ · - C ~ · - a) ~ ~ ~ ~ O <: ~ O ~ C) ·,. o 1 ·,. e4 ~: £ ·` o 3 z . C U. s~ ~ U] U] ,~ C O ,~ .~ U2 p: 3 ~Q C) ·' U] 3 ·,- C ~ O O O o ~
141 As Table B-3 shows, the NERC regional councils predict that growth in kilowatt hour use will equal or exceed peak demand growth in most regions over the next decade. They also project that on average annual load factors will remain constant over the next decade, at about the 1984 level. (The predicted value is 61.7 percent for 1-984 to 1993 from 1983 Electric Power Annual Report, p. 24; the predicted value is 61.8 percent for 1984 from the EEI Statistical Yearbook. ~ Thus, the decline in load factor between 1979 and 1983 is expected by the counc its to be arrested but not signif icantly improved over the next decade. Although this is clearly the intent of the councils, recent history suggests that it may not be easy to improve the growth of kilowatt hours over kilowatts. However, the interest expressed by various utilities in load management techniques suggests that there will be conside rable emphasis placed on this goal over the next decade. What does all this suggest about the relationship between GNP and electricity demand? It shows that the different concepts of demand have different growth predictions and that major variations in growth by reg ion have been the norm and should be expected in the future. It also argues tor disaggregating GNP growth by region to characterize past trends in the relationship between electricity use and GNP growth and to predict future trends better. Electricity can be transported, but there are limits to the distances over which it is efficient to do so. Centers of economic development may shift over time, but substantial lead times and capital expenditures are required. EEI predicts summer capacity -reserve margins in 1993 that a^re - higher in the areas where less demand growth is likely to occur and that range f rom 15. 5 percent in the ERCOT region to 26. 3 percent in the WSCC region. It predicts winter margins in 1993 f rom 24.4 percent in the SERC region to 37 percent in the SPP region. Interregional power sales can help equ ilibrate much of the mismatch of supply and demand and wi 11 he lp maximize eff icient general use of resources.
142 TABLE B-3 Proj ections of Growth in Annual Load and Energy Requi remeets, by Region, 1984 to 1993 (Percent per Year) Summer Peak Winter Peak Annual Growth in Regions Load Growth Load Growth Energy Requirements NPCC 1.7 1.8 1.9 MAAC 1.3 2.1 1. 8 ECAR 2.4 2.5 2.4 SERC 2.9 2.4 3.0 MAIN 1.8 2.3 2.3 SPP 2.7 3.1 3.0 ERCOT 4.0 4.1 4.0 MAPP 2.4 2.7 2.9 WSCC 2.6 2.2 2.5 Entire U.S. 2.5 2.5 2.7 resee Figure B-1 for explanation of region acronyms. SOURCE: Edison Electric Institute, 1983 Electric Power Annual Report, pp. 20-24 (data f rom National Electricity Reliabi lity Counc ill .
