Energy and the economy (the economic impact of alternative energy supply-demand assumptions)


Material Information

Energy and the economy (the economic impact of alternative energy supply-demand assumptions) a study
Physical Description:
v, 30 p. : ill. ; 24 cm.
Kaufman, Alvin
Daly, Barbara
Farb, Warren E
Farb, Warren E. ( joint author )
Daly, Barbara ( joint author )
United States -- Congress. -- House. -- Committee on Interstate and Foreign Commerce. -- Subcommittee on Energy and Power
Library of Congress -- Congressional Research Service
U.S. Govt. Print. Off.
Place of Publication:
Publication Date:


Subjects / Keywords:
Energy consumption -- United States   ( lcsh )
Energy policy -- United States   ( lcsh )
Energía -- Consumo -- EE.UU
Economic conditions -- United States -- 1971-1981   ( lcsh )
Economic conditions -- United States   ( lcsh )
Condiciones económicas -- EE.UU -- 1971 ****
federal government publication   ( marcgt )
non-fiction   ( marcgt )


Includes bibliographical references.
General Note:
At head of title: 95th Congress, 2d session. Committee print 95-51.
General Note:
CIS Microfiche Accession Numbers: CIS 78 H502-18
General Note:
At head of title: 95th Congress, 2d session. Committee print. Committee print 95-51.
General Note:
Issued Apr. 1978.
General Note:
Reuse of record except for individual research requires license from LexisNexis Academic & Library Solutions.
Statement of Responsibility:
by Alvin Kaufman, Warren E. Farb, and Barbara Daly ; prepared at the request of John D. Dingell, Chairman, Subcommittee on Energy and Power, Committee on Interstate and Foreign Commerce, United States House of Representatives by the Congressional Research Service, Library of Congress.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 021526205
oclc - 04434685
lccn - 78602033
lcc - KF49
ddc - 333.7
System ID:

Table of Contents
    Front Cover
        Page i
        Page ii
    Table of Contents
        Page iii
        Page iv
    Letter of transmittal
        Page v
        Page vi
        Page 1
        Page 2
    Introduction and U.S. energy demand 1950-76
        Page 3
        Page 4
        Page 5
        Page 6
        Page 7
        Page 8
        Page 9
    The energy scenarios
        Page 10
        Page 11
        Page 12
        Page 13
        Page 14
        Page 15
    Economic impacts
        Page 16
        Page 17
        Page 18
        Page 19
        Page 20
        Page 21
        Page 22
    Appendix. Other energy models
        Page 23
        Page 24
        Page 25
        Page 26
        Page 27
        Page 28
        Page 29
        Page 30
Full Text

2d Session PRINT 9 51

(The Economic Impact of Alternative Energoy SupplyDemand Assumptions)



APRIL 1978 a

Printed for the use of the Committees on Interstate and Foi e9i'ce, United States House of Representatives, and Energy and Natural Resources,
Commerce, Science, and Transportation, United States Senate


JOHN D. DINGELL, Michigan JAMES T. BROYHILL, North Carolina
FRED B. ROONEY, Pennsylvania JOE SKUBITZ, Kansas
HENRY A. WAXMAN, California MARC L. MARKS, Pennsylvania
ROBERT (BOB) KRUEGER, Texas TIMOTHY E. WIRTH, Colorado PHILIP R. SHARP, Indiana JAMES J. FLORIO, New Jersey ANTHONY TOBY MOFFETT, Connecticut JIM SANTINI, Nevada ANDREW MAGUIRE, New Jersey MARTY RUSSO, Illinois EDWARD J. MARKEY, Massachusetts THOMAS A. LUKEN, Ohio DOUG WALGREN, Pennsylvania BOB GAMMAGE, Texas ALBERT GORE, JR., Tennessee BARBARA A. MIKULSKI, Maryland W. E. WLmLIASON, Chie) Clerk and Staff Director KENNETH J. PAINTER, First Assistant Clerk ELEANOR A. DINKINS, Assistant Clerk WILLIAM L. BURNS, Printing Editor
Professional Staff
LEWIS E. Bzu.r, Minority Counsel

DAVID E. SATTERFIELD III, Virginia SAMUEL L. DEVINE, Ohio (ex officio)
TIMOTIIY E. WIRTH, Colorado ANDREW MAGUIRE, New Jersey MARTY RUSSO, Illinois E)WARI) J. MARKEY, Massachusetts DO U G WALG REN, Pennsylvania ALBERT GORE, JR., Tennessee HARLEY 0. STAGGERS, West Virginia
(ex officio)
P3UM 14 POTTER, St, &aff Director and Counse


Letter of transmittal ----------------------------------------------- V

Introduction ------------------------------------------------------ 3
U.S. energy demand, 1950-76 --------------------------------------- 3
The energy 10
Base case ----------------------------------------------------- 12
Conservation case --------------------------------------------- 13
High electric 13
High oil 14
Some caveats ------------------------------------------------- 14
Economic impacts ------------------------------------------------- 15
Employment and 19

Personal 20
Foreign 21
1. Consumption and price of energy, 5-year averages, 1947-76---------- It
2. Energy consumption by sector, 1950-75 ---------------------------- 51
3. Electric generating productivity- - 6
4. Household ownership-cars and 8
5. Productivity of transportation 9
6. Energy productivity 10
1. Energy consumption, 1976 and 1990, by 12
2. Energy supply, 1976-90, by case ---------------------------------- 15
3. Summary economic impacts, 17
4. Employment and unemployment in the United States, 1977-90, by
energy scenario 19
5. Estimated fixed investment by sector, 1976-90, billions of 1972 dollars- 206. Estimated personal consumption expenditures, 1976-90, billions of 1972
dollars ------------------------------------------------------- 21
7. Exports and imports, 1990, billions of dollars ----------------------- 21
Appendix-Other energy models ------------------------------------ 2a

Digitized by the Internet Archive in 2013


Hon. JOHND. DINGELL, Washington, D.C., February 2, 1978.
Chairman, Subcommittee on Energy and Power, Committee on Interstate
and Foreign Commerce, U.S. House of Representatives, Washington,
DEAR MR. CHAIRMAN: In your letter of March 22, 1977, you asked that we prepare a study of alternative energy supply scenarios, the resulting energy prices, and the impact of those scenarios on the U.S. economy. In fulfillment of that request, I am enclosing a study entitled "Energy and the Economy." The delay in preparing the report was necessitated by a shortage of computer funds.
The report evaluates the impacts of five energy cases on the economy. These include a base case, as well as conservation, high electric and two high oil cases. The assumptions in each of these cases were tested for economic in pact using the Wharton energy model.
"Energy and the Economy" was prepared by Mr. Alvin Kaufman, Senior Specialist in Business Economics (Resources and Regulation) Dr. Warren E. Farb. formerly with our Economics Division; and Ms. Barbara Daly. Research Assistant (Resources and Regulation).
Sincerely yours,

(The Economic Impact of Alternative Energy Supply., Demand Assumptions)
(By Alvin Kaufman, Senior Specialist in Business Economics (Resources and Regulation); Warren E. Farb, Specialist in Macroeconomics; and Barbara Daly, Research Assistant in Resources and
Energy consciousness over the past few years has heightened in the United States. The President called in April 1977, for the country to deal with the situation with the fervor of the "moral equivalent of war." However, the debate continues in the Congress and elsewhere as to how this "war" is to be fought. The paths the country has before it are varied and reflect differing priorities. There are many choices to be made concerning the country's goals and also the way in which it sets out to reach these voals.
The Congressional Research Service in this study has outlined five possible energy scenarios which exemplify different options and priorities. Included are a reference case, a high electric use case, two high oil cases (limited imports and high imports), and a conservation case. Through the cases this paper attempts to assess the economic response to various methods of meeting our energy demand through 1990.
Projected end use energy consumption in 1990 ranged between a low of 96 quads in the conservation case to a high of 108 in the base case. Annual average growth rates thus ranged from 1.9 to 2.7 percent. Imported energy comprised 23 percent of total consumption in the high electric case and ranged up to 35 percent in the high oil import case. Prices for the primary sources of energy were, for the most part, determined exogenously.
According to these scenarios the high electric case is expected to generate the strongest domestic economy over the long run; reliance on energy conservation through high energy prices shows the weakest results. Reliance on imported oil would also result in a relatively weak domestic economy. As a group, the alternative forecasts indicate that as long as energy is available in the forms and quantities assumed in the simulations, and there are no unforeseen shocks to the system such as the 1973-74 OPEC price increases, the flexibility of the economy, as interpreted by the model, would minimize the economic differences flowing from the various energy source assumptions.
Unemployment levels indicated for W11 cases are substantially lower than today's 6-7 percent level. Unemployment can be maintained at a relatively low level, according to the model, due to the projected sharp decline in the labor force growth rate. This in itself can be disputed and is somewhat deceptive, as indicated in the paper.


Most of the differences in the forecasts of economic growth occur after 1980 when the variances in the rates of investment spending are most pronounced. The average real growth rates of GNP to 1990 range only from 2.5 to 2.8 percent per year.
It must be noted that projetions, are based on an interpretation of the present, and how we view the future from within this framework. Thus the degree to which one can accept these results is based on the extent to which one agrees with the C RS assumptions and the structure of the Wharton energy model used to test these assumptions.

(The Economic Impact of Alternative Energy Supply-Demand Assumptions)
Severe winter weather in 1976-77, and the consequent shortages of natural gas, fuel oil and electricity in many areas of the United States has raised the energy consciousness of the country. This consciousness has been further heightened by the President's call for the "moral equivalent of war." One's perception of the need for and the kind of war, however, is dependent to a considerable extent on onels perception of the direction and velocity of future energy demand, as well as one's value system.
These perceptions color the assumptions upon which all forecasts are based. Assumptions are the product of insight and intuition heavily influenced by a vision of the future based on a conception of the present. In short, if times are hard, forecasts will tend to he gloomy; if times are good, they will tend to be overly optimistic. The above can be further summed up by noting that no one can forecast the future with confidence; there is always a strong element of risk and uncertainty. If we do sometimes turn out to be right, it is probably for the wrong reason(s). Despite the limited accuracy of longterm economic forecasts, they are a useful exercise in that they can provide a target or goal for policy formulation as well as an insight into possible reactions to various alternative proposals and assumptions.
It is in this spirit that we have developed several energy scenarios, and attempted to measure the impact of these on our economy. Before looking at where we may be heading, however, it may be useful to see where we have been.
U.S. Energy Demand, 1950-76
In 1973 the Arab nations imposed an oil embargo that did more harm to our national ego than to our economy. At approximately the same time OPEC raised oil prices to a level that is resulting in a massive income transfer from the energy consuming countries to the energy producing countries. The long-term impacts of this tax-like situation are still not clear. The short-run consequence of these actions, however, together with a coincident decline in economic activity, was a drop *in energy use from the record high in 1973. Current data indicate that energy consumption may have resumed the march upward, since consumption in 1976 was some 4.8 percent above 1975, although it is still 1 percent below 1973. The 4.8 percent rate of growth is substantially above the postWorld War 11 (1950-76) growth rate of 3 percent per year, but is in line with the 1965-70 rate of 4.7 percent annually.


Throughout the post-war period, consumption has grown at a relatively smooth upward pace with a shift to a somewhat steeper slope in the early 1960's (fig. 1). This increase was encouraged by rising real income coupled with declining real price. From the graph, however, it is apparent that real price and consumption of energy are closely related.
In order to more clearly illustrate this relationship, we have used 5-year averages to smooth out the annual fluctuations that have occurred in both energy use and price. For example, energy growth in the 1973-75 period was negative, but the average for the 5-year period, 1972-76, followed the past trend. During this same period, prices rose to record highs implying that consumption has not reponded in the latest 5-year period to the price signal.
Figure 1
Consumption and Price of Energy, Five Year Averages, 1947-76




80 to

60- ,. ,-Co nswfption

I 4T7 52 57 62 67 72 77
Source: Energy Perspectivu 2 U.S. Department of Intaor, Jun. 1976, p. 63 & 89.
The lack of response should not be taken as an indication that energy use will fail to react to rising energy costs, but rather as an indicator that it takes time to turn things around. Constantly declining energy prices over several decades has encouraged the creation of an inventory of relatively energy inefficient houses, cars, and equipment that will take years to replace or upgrade.
The initial impact of conservation measures, whether induced by price rises, taxes or government fiat, will be minimal. These impacts, owever, will tend to increase at a geometric rate over time as items of equipment are replaced with more energy efficient equipment.
Energy growth occurred in all economic sectors in the post-World War II period, but not at equal rates. For example, as shown in figure 2, the household-commercial and transportation sectors rose at 3.2 and 3.3 percent annually between 1950-76, while the industrial sector grew at 2.4 percent. As a consequence of these differential growth

rates the shares of total energy consumed by each economic sect or shifted during the period. The industrial sector, including nonfuel uses, accounted for 38 percent of energy use in 1950, but only 29 percent in 1976. The household-commercial and transportation sector maintained their shares at approximately one-fourth each througC-hout the period.
if one were to look. at net energy, however (that is total energy consumed less conversion losses), a somewhat different picture would emerge. The industrial sector would still have decreased its share, while the household-commercial and transportation sectors would increase from one-fourth each to approximately one-third during the 26-year period. These changes reflected the increasing electrification of the household-commercial sector and the increasing inefficiency of the transportation sector.

Figure Z

Energy Consumption bys Sector 1950W75




Year 70 2 7 7
Sce ri~or.ion \e~cie Z 'Np 6 7 5 06 esRlae31/7

elctict was cone atisotu au.Telse nurdi
converion o primry enrgy surcesto elctrictwreotedsp

1950ee 605an 70727476e o N f lctiit roue

Iyani oput intesea iurs consugemptio of 1.nernt her or.o Telerct was countedly atoitspoutputnvalue.hThe9losses incurredhinethe

forcain 19 perentofgtotalrenergycmpreen wit 13tereti 1950.6 This subtania inraerfete3h)tayelcrfcto.o h ain


'The flattening of the generating productivity curve during the past 10 years has been the result of thie imposition of strict environmental requirements, introduction of a new technology in terms of higher operating temperatures and pressures, and elongation of the planning Cycle. As these difficulties shake-out and as the technology for producing electricity continues to improve, there should be renewed impro'vement in conversion efficiency.
In the 1975-76 period, electric utility conversion efficiency has shown a substantial improvement, but we do not know if this will, continue, or if it is simply an aberration in the curve. Although one year cannot indicate a trend, it is likely that the rate of conversion Efficiency has resumed its upward march, albeit at a slower pace than before.
Figure 3

Electric Generating Productivity

120 Geneiating Produictivity/


Net Generaton


a 0 65 70 72 "N4 76
Sus:Energy Ppe~v Z. USD1. p. 73 &129.
The importance of improved conversion efficiency to overall energy use can best be appreciated by considering the primary energy needs of the electric utilities under alternative rates of conversion improvement. If electric consumption were to grow at an average rate of 5 percent per year for the next 15 years, close to 4 trillion kilowatt hours would. be required in 1990 compared with the present 2 trillion. This expansion in the demand for electric energy could be satisfied by a heat input of approximately 40 quadrillion Btu (Q) if the heat needed to produce a kWh (heat rate) declined 0.1 percent per year. Some 3 Q, or 1.5 million barrels of oil equivalent per day could be saved, however, if the heat rate improved at approximately half the historic rate, or 0.7 percent annually. There is a theoretical limit to the efficiency of this conversion process, ,and we may well be pushing hard against it. As a consequence, imnl)rovements such as indicated above may be hard to come by.

As the consumption of electricity rises, energy input reqjuirement~s wI also rise, but at a faster pace if generating efficiency doe> not improve because a larger quantity of the inputtwil be wasted.
If efficiency can be improved, this waste from conversion lo-sses- can be reduced. Losses, however, may well increase in t-'he future even with improved generating efficiency, due to our continuing need to turn to lower-grade sources of energy and because of increasing environm en t al restrictions. For example, the more we are forced to produce oil through secondary or tertiary recovery methods, tlhe greater the investment of energy in order to produce energy. A further shift from eastern coals to western coals will have a continuing serious impact on losses, since it takes approximately the same quantity of energy to produce a ton of western coal, but those coals have a lower heat content and thus yield less energy than most eastern coal. Using eastern coals wth scrubbers will also impose an energy penalty. Greater processing requirements in the future are also possible. For example, more extensive cleaning' of coal to remove pyritic sulfur will consume more energy. The production of fluids from coal carries with it greater conversion losses than at present. Output of low-lead gasolines means the use of more crude oil to yield the same usable energy. Thus, the outlook for the future is for a greater disparity between gross energy consumption and final demand despite improvements in electric utility generation. The conversion losses will be transferred from the electrical generating sector to the primary energy producing sectors, since the bulk of the losses will tend to occur back at the mine or wellhead rather than at the generating station. In short, barring seone kind of technical breakthrough we will have to run faster and faster to stay in the same place.
This situation is further compounded by the growth of appliance ownership over the past 13 to 14 years in the household-commercial sector, as well as by an increase in electric heating and air-conditioning. The proportion of households owning various appliances grcw dramatically in the 1960-74 period, with the exception of washing machines (fig. 4). The latter declined during this period primarily as a result of the trend toward single-individual1 households and young couples living in apartments. These types of households generally use common or commercial clothe s-washing facilities. In addition to the increased use of appliances is the fact that 49 percent of the houses built in 1974 were electrically3 heated and 48 percent were air-conditioned. This preponderance of electric space conditioning will result in greater future energy use since the higher efficiency at the point of use is often more than offset by conversion and transmission losses compared with other heating sources. This situation may be mitigated substantially by the widespread use of heat pumps rather than sistance heatingv. It should be noted, however', that the use of electricity in the household-commercial sector has increased close to sixfold between 1950 and 1975, and all indications point toward continued increases in electric usage in that sector of the economy.


Figure 4
Household Ownership Cars E Appliances
Percent of U.S. Households

80 11974 !A

60 I

i I iiii1731973!!

Ak CW=he Dish- Befri- WashA Two or One or
Cwid- Dryer washie erator Machie Mare Mare
Cars ,'bhes SOURCES: St atsk.iAbarstac of U.S. 1975. p 406 & 1976 p. 424.
Figure 4 also illustrates that there has been a substantial growth in the proportion of households owning one or more vehicles between 1960 and 1974, as well as a virtual doubling in the percentage of households owning two or more cars. Not only do we have more automobiles, but the bulk of the automobiles currently being produced are equipped with energy-consuming options. In 1975, approximately 92 percent of the automobiles produced were equipped with automatic transmissions, 90 percent with power steering, 76 percent with power brakes, 73 percent with air-conditioning and 71 percent with V-8 engines. These numbers represent substantial increases over 1960. For example, in that year only 7 percent of the cars produced were air-conditioned.
Adding to the energy impact of these options was the installation of pollution-abatement equipment. These two trends resulted in a decline in the overall vehicle efficiency from 12.4 miles per gallon in 1960 to 11.9 in 1973 (fig. 5). During this same period, passenger car efficiency dropped 8 percent from 14.5 miles per gallon to 13.3. At the same time that efficiency was declining, miles (Iriven. doubled from 719 billion vehicle miles in 1960 to 1.4 trillion in 1975.

Figure 5

Productivity of Transportation Energy

Avg. Miles Per Gallon

Transportton Energy"'

13 Vehie Miles Traveled


1960 65 70 72 74 76
Since 1973 both vehicle arnd automobile efficiency have improved. This trend should continue as the Nation, moves toward the use of lighter and smaller cars in the future, although better mileage will continue to be offset to some extent by more miles driven. An indication of the importance of improved vehicle efficiency is available from the following computations. If the 135 million vehicles on the road today had an average efficiency of 20 miles per gallon instead of 12, gasoline consumption would be over 4 million barrels per (lay less. If the effiCienxcy was 16 miles per gallon, the savings would be close to 3 million barrels per day. It is likely that vehicle efficiency will continue to improve for at least the next several years as a result of the legal requirement for more efficient cars, coupled with the push for mass transit and other energy saving transportation options.
The result of the various trends discussed earlier was a rather substantial increase in energy productivity in the 1960-67 peiodl, followed by a decline to 1970 and a rise to new highs since that year (fig. 6). Productivity is the output derived from a unit of energy input. The productivity of gross energy has been improving at a slower rate compared with net energy since 1972, reflecting the increased importance of conversion losses from electric utility plants in the overall energy picture. The household-commercial sector, on the other hand, substantially increased its productivity since 1970, reflecting improved efficiency of appliances, the use of insulation, and other measures to cut energy usage i n response to higher energy prices. Improved energy efficiency in this sector results in a double bang for the buck due to its orientation toward electricity. A reduction in the need for electricity also reduces conversion losses.


Figure 6

Energy Produci vity Indices




1960 65 67 70 72 74 75
Soure: Congressional Resebrch Service
In summing up, it may be of importance-that gross energy productivity improved during a time when energy prices were declining. This would seem to indicate that improved energy efficiency maybe impacted more by factors other than price, such as the introduction of new technology and processes which tend to be energy efficient more by accident than by design. From the foregoing discussion it would appear that more bang for the Btu could be obtained by improving the efficiency of the vehicular fleet and the electric utility industry than by any other measure.

The Energy Scenarios
Having briefly recounted where we have been, we can look at where we may be going. In order to do so, five possible energy scenarios have been developed. These include a reference case, a high electric use case, two high oil cases, and a conservation case.
In order to test the impact of these scenarios on the energy sector and the U.S. economy, the Wharton Econometric Forecasting Associates, Inc. (WEFA) Annual Energy Model was used. This particular version of the WEFA model while based on their annul long-term forecasting model, is still in the development stage. The principal difference between the standard long-term model and the energy version is the addition of a detailed energy sector. To achieve this, the mining sector of the annual model, which includes most energy production, has-, been expanded to specifically forecast crude petroleum, natural gas lnd coal production. In addition, the output of electric and gas utilities, as iWell as oil refineries, has been segregated and a forecast prepared. This greater level of disaggregation is then carried through the economy allowing estimates of energy consumption to


be developed. The current form of the model does not permit estimation of energy use by intermediate users such as electric utilities, although the model estimates gross demand for energy alnd some final demands.
Most of the prices for primary sources of energy, such as natural gas, coal, and crude petroleum, are deterni ined outside of the model. Prices of secondary forms, such as electricity, gas, and oil refinery products are determined within the model but, as was necessary for the high electric alternative, can be nmod(lified to achieve specific results. In our case the different energy simulations were :leoJlpihed by making changes in the various energy prices from tllhoir baseline levels. In addition, it was necessary to change taxei on col tain energy inputs which in effect alter the prices for sonime of the alternative energy sources.
The price changes needed to meet our assumnption,- were coiI)uted by WEFA. These, as a consequence of the implicit a---fnl)tionl in the model, do not produce a constant relationship between pl ice changes and consumption, nor is there a one for one change. For example, in the case of petroleum a 5.6 percent annual average incre tse in p) ce between 1976 and 1990 results in a 2.6 percent p)er year' mni! increase in consumption in the base case; a 9.2 percent annual rise in price in the high electric case suppresses consumption to a 2.0 percent annual increase. In t he latter case this implies that a 10 percent increase in price will allow a 2.2 percent rise in consumpti),r v, while in the base case a 10 percent price rise permits a 4.6 percent rime in consumption.
Compared with the base case the relationship in 1990 would be as follows:

Percent increase from the
base case
Case Price Consumption
Conservation --------------------------------------------------------- 31 -7.3
High electric ... ---------------------------------------------------------- 59 -8.8
High oil- Domestic ......... .... . . . . ......... . .16 -3.1
High oil-Imports.... ... ----16 -1.6

These and other changes work through the model by affecting the supply and demand for the several energy products, and by affecting the relative prices among the various substitute energy commodities.
Additional adjustments to the model have been made as nece sary to be sure the results are consistent with available a priori knowledge. Where taxes are required to achieve a particular goal, the receipts are recycled through the economy to avoid sharp economic changes. Monetary policy assumptions have also been adjusted where needed to avoid unreasonable tightness or ease in the financial markets, as well as to influence the cost of capital and investment decisions.
The various assumptions made in each case together with the computed energy estimates (table 1) follow:

22-673-7-- 3



1990 estimated
High oil
Actual Conser- High
1976 Base nation electric Domestic Imports

Total consumption (Q) --------------- 74 108 96 1 113 97 97
Average growth rate 1976-90 (percent) ------------- 2.7 1.9 3.1 2.0 2.0
Percent of total consumption:
Coal -------------------------- 19 24 20 24 19 19
Natural gas .....--------------------- 27 18 17 15 17 17
Nuclear ------------------------ 3 9 11 17 11 11
Petroleum .......... ---------------------- 47 46 48 41 49 49
Other ......-------------------------- 4 3 4 3 4 4
Imports -------------------- 20 31 27 23 30 35
Electricity consumption (billion kilowatt-hours) ---------------------2,040 4,050 3,653 6,011 3,664 3,676
Electricity growth rate (1976-90)
(percent)............. ...........------------------------------ 5.0 4.1 7.5 4.2 4.3
Energy prices:
Price Index, 1976=100

Coal(perton) -----------------$20.00 165 233 170 233 233
Electricity (per kilowatt-hour)..... 2.90 188 198 154 191 186
Natural gas (per thousand cubic
feet) ----------------------$ 0.90 245 245 245 245 245
Petroleum, crude (per barrel) -. $13.50 214 280 340 248 248

Uncorrected for improved efficiency required to permit elecric prices to decline. When corrected total energy consumption would be 107Q.

Base Case.-This case is essentially a high energy growth case
except that electricity consumption was constrained to a 5 percent annual growth rate. It was further assumed that there were no supply constraints and that energy prices would rise at a relatively moderate pace over the 14-year period. Coal prices would be up 65 percent, electricity 88 percent, natural gas 145 percent, and petroleum 114 percent compared with 1976.
As a consequence of these price assumptions, total energy consumption is projected at 108 quads by 1990, indicating an annual growth rate of 2.7 percent.' The base case growth rate compares favorably with the 3 percent growth experienced during the post World War II period. The lower rate results from conservation induced by price, as
well as anticipated slower economic growth. The latter should occur because of demographic changes, the increased importance of services and environmental constraints.
The energy mix in 1990 is expected to be somewhat different than in 1976, although oil will continue to provide close to half our energy needs. Approximately two-thirds of our petroleum needs will be
imported, resulting in 31 percent of our total energy consumption
comprising imports, compared with 20 percent in 1976. The imports incluhdle crude petroleum, residual fuel oil, various other products, and a s ,na!l quant'iy of ilqu fiel natural gas (LNG). The major shifts in energ mix, however, are expected to includ(le a decline in the importance of natural gas, and increased use of coal and nuclear sources. Nonconventional sources, such as solar energy, are not expected to substantially contribute to our overall energy needs in the relatively short time frame of 14 years.

STihee results are comparable to those of a 1976 CRS study that used the Data Resources Inc. (lDRI) energy model. See Kaufman, Farb, Daly. Ch. II, U.S. Energy Demand Forecast, 1976-90 in Project Interdependence: U.S. and World Energy Outlook through 1990. Committee Print 95-33, November 1977.


Conservation Case.-The conservation case assumes that the inp' ts of existing legislation such as the Energy Policy ad Conservat ion Act are carried through 1980 and that domestic petroleum prices follow the dictates of that legislation. Natural gas prices are maintained at a $1.45 per thousand cubic feet (nimef) through 1980, and then are permitted to rise at an annual '3.7 percent rate. ('oal prl es are assumed to move in correlation with oil prices to 1980 anl then remain virtually constant. Nuclear uei prices are assumed to increase at an annual rate of 13 percent; crude oil prices are postulated to rise, in real terms, at between 0.5 to 1 percent a year.
As a consequence of these assumptions, total energy use is estimated at 96 quads which indicates an annual growth rate of approximately 1.9 percent between 1976 and 1990. The energy mix in this case is somewhat different than that estimated for the base case. Petroleum contributes a somewhat higher proportion of the total than in the base case, but energy imports are a lower proportion. It is assumed that the higher oil prices will stimulate domestic output and depress demand resulting in a lower import requirement, although these sources will still supply over half of our petroleum.
Coal contributes approximately the same percentage of our needs as in 1976, but less than in the base case; natural gas declines reflecting the substantially higher prices and supply constraints, while the proportion of energy derived from nuclear sources increases dramatically compared with 1976. Consumption of electricity grows at barely more than 4 percent annually.
High Electric Gase.-The high electric case assumes that electricity consumption will follow the long-term trend and double every 10 years. The bulk of the incremental electric output will be generated by nuclear and coal units in this case. To achieve this result, crude oil prices were increased at an 8 percent annual rate while the real price of electricity was assumed to fall by approximately 1 percent per year. In addition oil and gas inputs to electric utilities were taxed in order to make these noncompetitive with other fuels.
In the high electric case, prices were adjusted to n1m-re conV 'rion to electric generation without regard to other potential factors afleeting energy prices, such as the impact of the -iie iLcreae in nIi 'ar generation on uranium prices Also, for the high electric cas, he Federal Reserve's discount rate wa: inrease( by one-halh of I ) .. (50 basis points) in or(ler to incre a the 'o t of capital. The hij r capital costs have the effect of making nuclear energy more exp.nsiv, preventing the system from over-stinulating nuclear ge er'ating capacity and maintaining more reasonable spreads between long- and short-termi interest rates. The model structure, however, internally allocates the necessary investment s)end(in for new electric generating capacity.
The result of these manipulations is an estimated energy con-m1)ption in 1990 of 113 quads indicating an annual average growth rate in excess of 3 percent. Thi" is achieved by a rather dramatic increase in the contribution of nuclear energy compared with the base case. The proportion of energy derived from coal is the same as in that case, but oil consumption is only slightly higher than in the conservation case. As a consequence, oil contributes 41 percent of our energy needs in this case, and energy imports are only 23 percent. Oil imports, in this case, are only 600,000 barrels more per day, or 0.4 percent, than in the conservation case.


In considering this case, however, it should be noted that the price assumptions imply an improvement in electrical technology that will permit the price of electricity to decline over the period. This implicit assumption, however, is not reflected in the energy inputs to the electric utility industry so that gross energy in the high electric case would be less than indicated above.
The price assumptions implicitly assume improvement in the heat rate at one-half the historic rate, or 0.7 percent per year. At that rate the primary consumption of energy would decline by 6 Q, so that total energy use in the high electric case would be 107 Q, or approximately equal to the base case. It would thus appear that the high electric case could result in reduced energy imports, and a total energy consumption no greater than that case. In relation to the conservation cae. energy imports would be less. If the 6 Q saving were prorated among all sources, imports in the high electric case would be 5 percent less than the conservation case. On the other hand if the full savings could be attained at the expense of imports, these would be 23 percent below the conservation case.
HigTh lOil Cases.-The high oil cases were broken into two parts. In the one instance it was assumed that imports were limited, and that a .ibsta-ntial portion of our oil requirements would be achieved through doiomestic production such as increased off-shore output, or through the production of shale oil or liquefaction of coal. In the second intan e, it was assumed that the bulk of our oil needs would be obtained through imports.
The WEFA model, however, does not have a mechanism to make suc a distinction. To overcome this difficulty the price assumptions in each instance were the same, bu.t in the high oil-domestic case the model was adjusted so that the incremental increase in energy supplie8 is produced domestically and reflected in the output of the mining sector. In the high oil-imports case the increment is assigned to imports.
The assumptions made in each case included removal of all gasoline taxes, imposition of an. en.vironmen-.tal tax on coal consumI)tion by electric utilities increasing at 10 percent a year, an.d a increase in domestic crude oil prices to reach $12 by 1980. Domestic crude prices were then held constant from 1980 to 1990.
As a consequence of these assumptions, consumption is estimated at 97 Q, or roughly equal to that forecast for the conservation case. This occurs because of the relatively highli eneri'y prices coupled with a shift from electric heating to oil heat. The latter results in savings in conversion losses at the generating plant, thus contributing to a reduced energy requirement. This is reflected in an electric consumption growth rate approximately e(Iqual to the conservation case.
'The composition of eergy use in. each case is identical, the only difference beitg the level of imports. In- either case, the level of imported celery is rather high, accounting for 30 to 35 percent of our

Close to half of our energy would be derived from oil, with coal roin from approximately on1e-fourth inl the reference case to onefifth in. the high oil cases. Grea1ter reliance would be placed on nuclear oeler-v in these cases than in the base case.
Sowi C(&areats.-The five cases as outlined are depen(.dent on our ability to expand coal and nuc!eir output sufHiietly to achieve the


stipulated goals (table 2). If this, cRn be a1Chieved, Oil 011tp1)1t wuol have to increase between 22 and :12 l wrcellt (dlepenldin o),n0 the v)y 1990 compared with 1976. Natural gas supply, includIing impOrte (,I I N G is postulated to decline up to 20 percent, and thus cannot b~e cwivted on to pick up any shortfall.
Hg;h ori
Conser- High - -1976 Base vation electric Domestic I c sports

Domestic supply:
Coal (million tons) ------------------------ 597 1, 114 824 1, 193 811 802
Natural gas (trillion tons cubic feet)----------- 19. 8 19.9 16. 3 16.6 16. 1 15.8
Doesi----------------18.9 16.7 13.8 14.1i 13.6 13.3
Imports------------------0.9 2.4 2.5 2.5 2.5 2.5
Nuclear billionn kilowatt hours) --------------- 190 938 957 1848 985 957
Petroleum (million barrels, per day)----------- -17.4 23.0 21.6 21.3 22.4 22.7
Domestic---------------------------- 10.3 10.7 11.0 10.7 10. 1 8. 1
Imors------------------7. 1 12.3 10.6 10.6 12. 3 14.6
Other (quads) --------------------------- 3.5 3.5 3.5 3.5 3.5 3.5

In none of the cases have we made any assumption in regard to solar, wind or other nonconventional energy sources. We feel the ability of such sources to contribute to our energy needs is severely limited over the timeframe with which wve are concerned. In no case, however, would we envision that these new technologies could contribute more than 3 to 4 Q in total by 1990 because of the lags in the system. We would regard that forecast as extremely optimistic.
Our case estimates call for an increase in coal production ranging between 34 to 87 percent over the 1976 level. Such increases appear to us to be difficult at best unless the institutional and economic problems inhibiting expansion of these industries can be solved.
On the nuclear side the economic issues revolve around the rising capital cost of such units and their availability. The major problem. however, may be the long lead time required to build such plants, plus strong opposition by some elements of the public. It can take from 10 to 1?. years to build a nuclear plant because of regulatory delays, environmental require mcnts, and law suits. In addition, low electrical load growth in 1975-76 caused the cancellation of plants. As a result, orders for new nuclear plants have declined.'
The impact, of the decline in new plant orders on our ablility to meet the nuclear capacity requirements postulated for 10990 is still uncertain. As of June 30, 1977, there were 67 nuclear units licensed to operate, 89 being built, 54 planned with reactors ordered, and 22 others planned without reactors ordered. Those 2.32 plants have a total capacity ofe 231 million kW. If all of these are in service by I f9D,0 and operat e at a 60 percent capacity factor, nuclear output would total 1,213 billion kWh in that year, or more than enough to meet the estimated requirement in all of our cases except the high electric case. Even if the 22 reactors planned but not ordered die on the drawing board, nuclear output (1,073 billion kWh) would still be enough in 19%0. to satisfy most of our case needs.
I Wall Street Journal, "Firms That Make Nuclear Power Plants Expect Slump in New
-Orders To Continue," Nov. 30, 1977, P. 24.


There is, however, a high probability that some of the units under construction may not come on line on the targeted date or that some of those planned and ordered may yet be cane e'lled or delayed. In that case, we could be hard pressed to meet our nuclear groal.
Coal then becomes the fuel of last resort, but thisindustry is faced with serious problems of its own. On the supply side these include questions concerning the availability of trained mine labor, the profitability of the industry and thus its ability to raise the needed capital, and environmental regulations. The latter impact the industry both at the mining and consumption ends of the business. On the ining side, surface mine and subsidence regulations may inhibit the development of additional capacity, or may introduce sufficient uncertainty to cause investors to wait for a clear signpost of what can be expected.
On the consumption side, many localities have literally regulated coal out of the market for environmental reasons. Whether these areas ,are now willing to permit this fuel back in is problematic. Further, the full effect of recent "clean air" legislation has not yet been felt, nor has the problem of sludge disposal from stack gas scrubbers been fully explored.
As a consequence of the above, there is a strong probability that neither coal nor nuclear will be able to fulfill its assigned role. In that case, oil imports would burgeon even beyond the substantial quantities postulated in the high oil-imports case.
Economic Impacts
The question of economic impacts resulting from variations in the energy sector is a difficult one. To some extent this interaction has been probed by the Energy Modeling Forum. A common set of assumptions and scenarios were tested on various models. The general conclusions of this study were as follows:
In the presence of constant energy prices, increases in economic activity produce similar increases in energy demands, although these may be moderated by trends toward less energy intensive products and services.
Higher energy prices or reduced energy utilization need not produce proportional reductions in aggregate economic output. There is a potential for substituting capital and labor for energy and the contribution of energy to the economy, relative to these factors, is small.
The models do show some substantial reductions in economic output resulting from higher energy prices. The magnitudes of these reductions are very sensitive to the substitution assumptions implicit in the models. Further, the impacts may be large for individual sectors of the economy.
The benefits of energy substitution may be lost in part if energy scarcity impedes capital formation. Reduced energy inputs may cause lower levels of investment and, consequently, reduce potential GNP. This indirect impact may be the most important effect of energy scarCity.2
A summary of the output of the various models and their problems is contained in appendix L
In our case, however, we were not probing the economic reaction to variations in energy output so much as the economic different methods of meetincr the demand. Thus, our five scenarios were all run through a single model. In reviewing the results one must keep in mind the fact that the, answers are conditioned by the energyeconomy relationship built into the model. The numbers derived from
2qnergy Modeling Forum, 1'1'nergy and the Exonomy," Institute for Energy Studies, Stanford Universit5,, E.NlFreport 1, September 1977, 1). Ili.

the computer should, therefore, be regarded in relative terms rather than as absolutes.
Of the five alternative energy simulations conidcilre(1 hiere, the highr electric case is expected to generate the strongest (loniestiC ecNom11y over the long run; the conservation case is expectedl to result in tile weakest economy Over the 1976 to 1990 period, tje average annual rates of growth of real GNP ranged from 2.9 to 3.2 1),-rcent per year, although the components showed greater variations (table ")). Reliance on imported oil, together with relatively strong increases in energy demand, would also result in a relatively weak domestic economy. As a group, the alternative forecasts indicate that as long as energy is available in the forms and quantities assumed in the simulations, and there are no unforeseen shocks to the system such as the 197.3-74 OPEC price increases, the flexibility of the economy is likely to minimize the economic differences flowing from the various energy source assumptions.

Conser- High Hg i
1976 Base vation electric Domestic Imports

Real GNP I---------------------------------------------1,364 1,943 1,900 1,962 1,903 1,901
Growth rate I (percent) ----------------------------- 3. 1 2.9 3.2 3.0 2.9
GNP: Energy indeX2 ------------------------ 100 98 108 99 107 107
Real personal consumption3--------------------.. 819 1,229 1,200 1,236 1,200 1,195
Manufacturing output 3------------------------.. 305 519 505 533 504 503
Real gross private domestic investment 3------------. ... 183 305 286 326 287 288
Net exports 4----------------------------------------------------. . . -9 -30 +14 -20 -8 -35
Federal deficit or surplus 4 ..-..-..--.--.--..-..-..--.--.--.-59 23 14 74 -5 -19
Unemployment (percent)----------------------- 7.7 3.3 4.6 3.3 4.5 4.7
Inflation rate (percent)'------------------------ 5.1 5.0 5.3 5.4 5.0 4.9
Savings rate (percent)------------------------- 6.6 5. 1 4.9 4.2 5.2 5.0
Bond rate (percent)--------------------------- 9.0 8.5 8.5 9.7 8.3 9.2
Commercial paper rate (percent) ------------------ 5.3 6.5 6.6 7.5 6.4 6.3
1 Average annual. 3 Billion 1972 dollars.
2 1976 equals 100. 4 Billion current dollars.

In this regard, CRS had earlier completed an analysis of impacts on economic inputs resu ,iting from energy conservation under various
assumptions pertaining to the flexibility of the economy.'
If the economy has sufficient flexibility to adjust to changes in energy use, an 18 percent drop in energy inputs from what would normally be expected would result in a 1.2 percent rise in capital and a 1.3 percent increase in labor inputs coupled to a decrease of 0.3 percent in other materials. As a consequence of these shifts GNP would only be 0.7 percent below the base. The economy would pay a price, however, in terms of greater inflationary tendencies because total productivity would be some 3 percent below the base, idctn economic ineficiency.
If the economy were relatively inflexible, however, total productivity would only be below the base by 1 percent, indicating relatively mild inflationary tendencies. GNP, in this case, would be 9 percent below the base, with capital and labor inputs 8.5 and 8.8 percent less, respectively.
a Kaufman, Alvin and Barbara Daly, "Alternative Energy Conservation Strategies: An Appraisal" CRS, 77-114S, 45p; and a revised report entitled "Alternative Energy Conservation Strategies, An Economic Appraisal."


The trade-off thus appears to be between economic efficiency and reduced economic growth. It should be noted that in the worst'case, unemployment would be catastrophic in approaching 14 percent of the labor force if wage rates remained constant.
Employment and Unemployment.-Even though GNP growth is low relative to historical rates, unemployment varies from a low of 3.3 percent of the labor force in the base case to 4.7 percent in the high oil imports case. By today's standards these rates may seem to be unachievable given the projected economic growth. There are, however, two reasons for the relatively low rate. One is the projected sharp decline by Wharton in the rate of growth of the labor force to less than 1 percent annually by 1990 compared with well over 2 percent currently. Just as the current economic situation is made more difficult by the rapid increase in labor supply, and has prompted many analysts to redefine "full employment" at or above 5 percent unemployment, beginningr in the 1980's the decline in the rate of new workers entering the jo makt a permit lower overall economic growth while still
achieving "full employment." This slower growth in the labor force will tend to lower the "full employment" rate of unemployment. While the rates of growth of the labor force used in this analysis tend to be lower than some, the direction of the change is clear, as is the magnitude of its potential impact on the required economic growth to achieve and maintain full employment. The slower rate of growth of the potential labor force will be offset by continued increases in labor force participation by women. This has been accounted for in the model.
A second reason for the low unemployment rates derives from the definition of labor force and unemployment. The labor force comprises all of those persons who are employed and unemployed. The latter are defined as those not working but seeking employment and available. As people become discouraged or unavailable, they drop out, of the labor force and are no longer counted as part of the unemployment
If we review table 4 we see that U.S. working age population (over 15 years old) remained const ant in all five energy cases, but labor force Varied by as much as 1 million people. If these persons are added ba,.ck into the ranks of the unemployed, the rate would be approx,matelvy 1 percentage point higher or 5.7 percent in the high import case. Em-drployment will vary between a low of 110 million persons in that cas e to a high of 112.7 million in the high electric case. The latter can produce 2.4 million more jobs than the conservation case, but, approximately the same employment level as the base case. TABLE 4.-EMPLOYMENT AND UNEMPLOYMENT IN THE UNITED STATES, 1977-90, BY ENERGY SCENARIO
[M'!illi1ons of people]
1977 1980 19F5 1990
Population ever 15 yr old---------------------------- 161.0 168.0 177.0 184.0
Civilian labor force:
Base----------------------97.2 103.7 110.7 116.5
Conservation--------------------------------- 97.2 103.7 110.4 115.6
High elect ic -------------------------------- 97.2 103.7 110.5 116.5
High oil D-)iweslic ----------------------------- 97.2 103.7 110.6 115.8
high oil -In1 pwts ------------------------------ 97.2 103.7 110.4 115.5
B'se- -------------------------------------- 90.6 98.2 106.8 112.6
Conservaticol---------------------------------- 90.6 97.7 104.9 110.3
High clectic-.------------------90.6 97.6 106.6 112.7
HIgh) (1 i l mti--------------90.6 98.2 105.6 110.5
High oil --imports ------------------------------ 90.6 97.8 104.6 110.0


Inflation.-It, therefore, is possible that in the 1980's, when t1le rate of labor force increase is once again down to (,lose to I percerit per year, not only will lower rates of economic growth be suitable, Init inflationary pressures associated with tight labor markets (under
5 percent unemployment) may subside as well.
The present estimates of inflation through 1990 are largely a re -ult of upward price pressures already built into the economic system, and additional pressures brought about by increasing energy cost ,. The energy prices (table 1) were developed to generate the partictilar energy mix desired and do not necessarily effect potential development costs. It is quite likely that the development of synthetic fuels (Syn fuels) will require considerably higher prices than implied here because of higher development costs'. For the purposes of our analysis, it is assumed that the current rate of investment in research and development in the syn fuel case will generate the necessary production at the assumed prices.
If energy prices increase more rapidly than assumed, the rate of inflation would undoubtedly be higher. Even more importantly, the rate of economic Yrowth could be slower, and there would be greater unemployment t9an would otherwise te the case, unless all-of the spending on energy is recycled back into the economy. As energy prices increase, the resulting economy would move toward that described by the conservation alternative.
As expected, the economy would suffer with the highest inflation rate in the high electric case, and the lowest in the high oil imports case. The conservation case is a close second to the high electric case. The latter results in more inflationary pressure because of the higher levels of economic activity and higher oil and natural gas prices required to produce the desired level and type of electric generation, despite a postulated decline in the price of electricity. The inflation forecast is also affected by the higher interest rates that are brought about, at least partly, by increased investment spending in the electric case. It is possible, however, that this could be offset by selected monetary and fiscal policies. If methods can be found to achieve conservation without having higher effective prices for energy, the inflation rate would be correspondingly lower.
Investment.-If the policies indicated above are successful in reducing interest rates, it is likely that investment spending would be somewhat higher and the overall level of economic activity greater than estimated. The projected higher inflation rate, together with the relatively low level of unemployment, also causes federal tax receipts to swell, resulting in a projected Federal budget surplus in excess of $70 billion in 1990 in the high electric case. As a result of this surplus, the model generates a relatively low rate of private saving so that the Government, in effect, is providing most of the savings in the economy. This also occurs, but to a lower degree, in the conservation and base cases. If Congress acts to maintain stable average tax rates, limiting the
-accumulation of a budget surplus, there would be some additional economic stimulus; but this stimulus would be largely offset by an increase in Ithe individual savings rate which would reduce private consumption.
It is clear from the model that the most important economic stimulus comes from the increased levels of investment in the high electric case (table 5). Even though the electric utility sector in that instance

requires more funds than in any of the other alternatives, the economy in general responds with increased investment in all areas; this requirement would be largely financed through higher corporate profits until 1985, and after that date by Government saving.
TABLE 5.-ESTIMATED FIXED INVESTMENT BY SECTOR, 1976-90 IBillions of 1972 dollars]
High oil
Conser- High
1976 Base vation electric Domestic Imports
Fixed investmenL ----------------------------177 298 278 318 280 279
Residential -----------------------------55 58 59 60 58 57
Nonresidential ... ........ ...-------------------------- 122 240 219 258 222 222
Commercial and other ..... ..--------------------------- 60 60 58 60 60
Transportation and communication ------------------ 32 32 30 32 32
iManufacturing, durables ------------------------- 50 47 50 48 48
Manufacturing, nondurables ----------------------- 35 33 34 34 34
Utilities ............ .......------------------------------------ 43 28 66 30 31
Electric --------------------------------- 39 24 62 26 27
Gas and water.... ----------------------------- 4 4 4 4 4
Farming and mining --------------------------------35 34 35 33 33

The lowest investment occurs in the conservation case, with the major drop in the utilities sector. The two high oil cases tend to be in the middle, again as a result of lower utility investment compared with the base case. In this regard, we would note that investment in the high oil domestic case would be close to that in the electric case with the mining sector absorbing substantial sums for syn fuel development. As a result of the way in which the case was computed, this impact does not occur.
The investment distribution through the various sectors tends to be mrearkably similar in each case except for the utility sector.
Personal Consumption.-Personal consumption expenditures would be highest in the high electric case (table 6) and lowest in the high oil-imports case. Regardless of the case selected, however, services are the major expenditures sector, followed by nondurable goods. The importance of the services sector results, to some extent, from heavy expenditures for electric utility services. From the point of view of the homeowner most of the benefits of increased personal consumption expenditures are likely to be eaten up by higher household operating expenses. In 1990, high electric assumptions would result in approximately 4.5 percent of the total personal consumption expenditures being spent on heating, gas and electricity, compared with 3.3 percent in 1977; these expenditures would account for 4 percent in the base case. In the conservation and high import cases the heat-gas-electric percentage would only increase to about 3.5 percent. The higher level of overall economic activity projected under the high electric assumptions should generate sufficient personal income so other categories of consumption are not lowered by the higher heating and utility expenditures. This level of consumption spending, however, could not be maintained if the personal saving rate was not unusually low as an offset to the large Federal budget surplus.

High oil
Conser- High
1976 Base va~ion electric Domestic Imports
Personal consumption expenditures --------------- 819 1, 229.0 1, 200. 0 1,236.0 1, 200.0 1, 195.0
Durable goods -------------------------- 118 204.0 199.0 204.0 199. 0 198.0
Nondurable goods ------------------------- 334 480.0 472.0 478.0 471.0 469.0
Services ------------------------------ 37 545.0 529.0 554.0 529.0 528.0
Heating oil and coal ------------------------------ 5.5 5.4 4.8 5.6 5. 5
Electric utility services --------------------------- 38.0 32.0 45.0 33.0 33.0
Gas utility services ----------------------------5.2 4.8 4.7 4.8 4.7

Foreign Trade.-In all of the cases considered in this paper, with the exception of the conservation case, the balance of trade as measured by net exports is in favor of other countries measured on a current dollar basis (table 7). Only the conservation case has a positive balance, and this despite the fact that the high electric case has a lower energy import requirement. The conservation case results in the highest level of total exports and lowest total imports of any of our five cases.
[Billions of dollars]
Case Exports Imports Net exports
Base ----------------------------------------------------------- 540 570 -30
Conservation ---------------------------------------------------- 548 534 -14
High electric ---------------------------------------------- 547 567 -20
High oil-domestic ----------------------------------------------- 544 552 -8
High oil-imports ------------------------------------------------ 540 575 -35

This occurs because total imports, and to a lesser degree exports, are more closely alined with overall economic activity than with the volume of oil imported. Further, the electric case involves greater imports of finished manufactured (zoods which could result in enhancement of the U.S. standard of living as a result of increased foreign trade.
Even though the high electric case is expected to generate the fastest increase in the total value of imports of the five alternatives, any negative implications are lessened by the fact that this increase is generated by a stronger domestic economy than in the other cases and not because of increased oil payments. In this simulation, exports increase at a somewhat slower rate leading to the development of a trade deficit from 1988 onward. In the long run, however, the international payments system would adjust, reducing the rate of increase in manufactured imports and increasing exports. The net result would be for a stronger U.S. economy in the long run.
The major trade deficit occurs in the high oil imports case. In this instance, as a result of the large amounts of U.S. dollars going abroad to pay for oil imports, foreign consumers are able to make increasing claims on U.S. production. While this automatic adjustment process


can keep the U.S. economy running at reasonably full employment, the long term benefits accrue to the oil exporters. These benefits axe all the larger because the terms of trade (relative to prices of traded goods) have moved to the disadvantage of oil importers.
The above estimates are, of course, dependent on the assumptions fed into the model as noted earlier. In the case of the external sector, however, this may be even more so in the sense that additional variables are involved. For example, if the pattern of prices assumed are different then the answer will be different. This is particularly critical, since imported energy prices are not set by he market, but by a cartel operating under political as well as economic guidelines.


In the preceding pages, CRS-using the Wharton Annual Energy model-has atte to assess the impact of alternative energy
futures on our economy. We thought that it would be a useful addition to the CRS paper to describe the way several of the other models work and what they are equipped to do regarding energy. The short summary of various models w1lich follows is based almost solely on "Modeling Energy-Economy interactions: Five Approaches." I This book contains papers presented in May 1977, at a Joint National Meeting of the Institute of Management Sciences and the Operations Research Society of America. The modelers whose works are presented here were encouraged to the extent possible to follow a set of common assumptions regarding resource availabilities, the substitutability of other factors of production and goods and services for energy, and possible constraints on the use of coal and nuclear power. For the most part other assumptions are based on the CONAES (Committee on Nuclear Alternative Energy Systems) Modeling Resource Group's base case GNP assumptions.
Four scenario runs were made through each model as described below:
(1) Ue: Unconstrained development of resources and 0.75 elasticity of substitution;
(2) Ce: Constrained development of resources and 0.75 elasticity of substitution;
(3) Ui: Unconstrained development of resources and 0.25 elasticity of substitution; and
(4) Ci: Constrained development of resources and 0.25 elasticity of substitution.
In the unconstrained cases, domestic oil and natural gas are exhausted but large quantities of electricity are produced by coal and nuclear power plants. In the constrained cases, the use of coal and nuclear power is subject to substantial constraints.
The models dealt with here are all long range models "exploring rather gradual changes over decades." This long range feature can be seen as a plus or minus depending on one's point of view. Admittedly the long range (to 2000 and beyond) future is almost impossible to foresee (especially as one considers the difficulty of forecasting even the short-range future today), but the models are useful in slowing the long-range consequences of possible actions we might take in the near future based on the best available information.
The choices taken regarding energy policy should at least be geared in a certain direction; the models, by testing the various scenarios, help to indicate the path the policymakers, should take, based on a future as seen from the perspective of today..
'Hitch, Charles J., ed., "Modeling Energy-Economy Interactions: Five Approaches," Resources for the Future, Washington, D.C., 1977, 303 p. Research Paper R-5.



The ETA-Macro model is designed to estimate the extent of two-. way linkage between energy and the U.S. economy. It tests the relative strength of the linkage as it responds to new technologies and other changes in energy supply (real or contrived). The degree to which one accepts or rejects this model depends on one's opinion of the "elasticity of substitution" between energy and other inputs to the economy. The proponents of this model believe energy demand to be relatively inelastic. Thus if one feels cost effective conservation is relatively easy, one would probably not utilize this model.
ETA-Macro is a single integrated model based on two submodels:(a) ETA, a process analysis for energy technology assessment; and (b) a macroeconomic growth model providing for substitution between capital, labor and energy inputs.
For an overview of ETA-Macro see figure 1.
This model assumes the following:
(1) The impending exhaustion of oil and gas resources-and a transition to new supply technologies over the next 20-50 years;
(2) Price induced conservation-that is, the possibility of substituting other economic inputs in place of energy; and
(3) The effects of rising energy costs upon the accumulation of physical capital in future time periods.2
In order to focus upon the energy sector, the economy i s generally described in highly aggregative terms. Within the energy sector only two categories are distinguished-electricity and nonelectric energy.
For the economywide results this model depends up on f our inputs: K, L, E,Nrespectively capital, labor (measured in efficiency units), electric arid nonelectric energy. Assumptions regarding these inputs are as follows:
(a) There are constant returns to scale in terms of these four inputs;
(b) There is a unit elasticity of substitution between one pair' of inputs-capital and labor;
(c) There is a unit elasticity of substitution between the other pair of inputs-electric and nonelectric energy;,and
(d) There is a constant elasticity of substitution between these two pairs of inputs.3

fUELS, et.)



SIbid., 2.
*Ibid., p. 5.


The four sets of assumptions being tested in the models presented here revolve around various degrees of a nuclear growth policy. The ETA-Macro model results are as follows:
Under base case assumptions prior to the year 2000 (for the most part provided by the CONAES Modeling Resource group) the macroeconomic effects would be negligible, however after that year they are projected to be substantial. An annual GNP loss in 2010 of $100 billion is possible. It must be noted that although this number is large absolutely, it is a relatively small amount of projected GNP (3 percent).
According to this model the crucial economic parameters for U.S. energy policy are the elasticity of substitution between energy and other inputs-together with the rate of growth of the labor force and of its productivity at constant energy prices. These are the parameters that appear to dominate the long terra growth picture.4
This model looks at the big picture, the macro impacts. It is not suitable for analyzing specific proposals for energy conservation within specific sectors. Other models are designed to give that kind of delineation.
The ETA-Macro model in this exercise thus aids us in assessing the pluses and/or minuses of following or not following a nuclear growth policy. Is it reasonable to accept slower economic growth as the price of eliminating possible nuclear hazards? The options as perceived through the ETA-Macro model are presented, it is left to the policymakers to decide.

DESOM-LITM (Brookhaven/DRI)
The second model to be discussed is also composed of two separate models: the Data Resources, Inc. (DRI) Long Ternm Interindustry Transactions Model (LITM), and the Brookhaven National Laboratory (BNL) Dynamic Energy System Optimization Model (DESOM). In this system "a small macro-model (four sectors plus six energy sectors) is used to drive a larger interindustr model which in turn determines the many end use demands in DESOM." 5 Figure 2 depicts the processes of this model.
The BNL DESOM focuses on the end use demands for energy attempting to find the cheapest way of meeting them. The model contains only limited ability to substitute other inputs for energy, but gives a detailed picture of the appropriate fuel mix for a given level of end use demands. In this study, the DESOM model is linked to the DRI-LITM model, which, in turn, models the substitution which occurs between energy, capital, labor, and four types of materials as a result of changes in relative input prices.6
Like the other models this model has many exogenous plug-ins resulting possibly in the output being determined more by the person making the run than the model itself.
Of the models discussed in this appendix, this one seems to have departed most from the CONAES assumptions. The "alternative energy form" concept which refers to a possibly unpredicted energy source (in amount or kind), and associated EMF values are not employed, since the DESOM-LITM model incorporates:
'Ibid., p. 28.
5 Ibid., p. 285.
a Ibid., p. 280.


Variables representing most of the potential alternative energy supply and conservation technologies, including a wide variety of synthetic fuel, solar, and conservation technologies (heat pumps, electric cars, etc.). Also, the "elasticity of substitution of energy for nonenergy" is not parameterized upon, since, in the context of the system of models used, the elasticity is determined endogenously.7
As other models also show, the quantity of energy consumed during the early period (thru 2000) does not differ greatly among the four scenarios, reflecting the long life of already existing capital equipment and end use devices. However in the post 2000 period great divergence among the scenarios results.
Lester Lave in the same book criticizes this model as perhaps trying to do too much with too many unanswered explanations. Lave also appears to be hesitant regarding the way in which the micro model seems to predominate the macro model'

Net Inots of EnwyNCw Tochno ,y Impt eem- DRI Otpt ad Final D1m4e (4)
WW Produce WIvit Tra d*La tmltidsr
/ Conventona Enm" Sectwo Tras"etn MoelSt"e and Aggrgat Emag
/ Capeakion eofDomesf CWauP1ea (BIftd)

Leet Coat Min at Energwy Sup- BNL

~ ,wi(miro Emissions
o! mleiocome. n .. .

Resources for the Future
The RFF/SEAS (Resources for the Future, Strategic Environmental Assessment System) is a system of interlinked models, the core of which is the University of Maryland's dynamic input-output model of the U.S. economy, INFORUM. This model is one which requires the modeler to make many assumptions not only in the beginning but throughout the run. In this way the model becomes highly subjective. The model is large, depicting 185 sectors. The end focus of this model is the enviro;nme ti and it produces monetary estimates of environmental quality for each year.
In all cases run through this model net energy consumption grows less rapidly than GNP and the net Btu/GNP ratio falls. Most of the results are what would probably be expected. Coal production inI Ibid., p. 46.


creases quite rapidly in the unconstrained ca,:es. The shift away from electricity to other fuels is most prominent in the cotistrained cases due to the assumed ban on new nuclear development and limits on coal production.
Four different measurements of environmental impacts have been developed: pollution residuals, land disturbed, pollution abatement costs and pollution damage costs. These results are summarized through a welfare index by subtracting per capita environmental damage estimates from per capita consumption figures. Table A presents these computed welfare indices as they appear in the text of the RFF paper.


Scenarios 1975 1985 2000 2020 2025

A. Unadjusted estimates:
U75 ------------------------------------------------- 2 947 4,156 6 236 ---------- 9 714
U25 ------------------ ------------------------------- 21947 4,110 6 134 ---------- 9 513
C75 ------------------------------------------------- 2 947 4,128 6,202 ---------- 9 390
C25 ------------------------------------------------- 2:947 4,044 6,114 ---------- 9: 063
B. Estimates adjusted for labor productivity losses:
U75 ------------------------------------------------- 2,947 4,110 6,136 8,874 9,559
U25 ------------------------------------------------- 2,947 4,102 6,122 8,820 9,494
C75 ------------------------------------------------- 2,947 4,083 6,102 8,567 9,183
C25 ------------------------------------------------- 2,947 4,035 6,090 8,440 9,027
C. Estimates adjusted for labor productivity losses and savings constraint (Elasticity =0.3):
U75 ------------------------------------------------- 2,849 4,182 5,631 8,729 9,503
U25 ------------------------------------------------- 2,849 4,167 5,555 8,473 9 203
C75 ------------------------------------------------- 2,849 4,159 5,507 8,177 8:845
C25 ------------------------------------------------- 2,849 4,100 5,275 7,911 8,570
D. Estimates adjusted for labor productivity losses and savings constraint (G N P from model):
U75 ------------------------------------ ------------- 2,849 4,065 4,553 7,759 8,560
U25 ------------------------------------------------- 2,849 4,051 4,410 6,942 7,575
C75 ------------------------------------------------- 2,849 4,017 4,210 6,412 6,963
C25 ------------------------------------------------- 2,849 3,892 3,845 4,952 5,229

These estimates have been scaled to eliminate the minor differences in the GNP projections that occurred in the model runs due to targeting.

As these numbers indicate:
The more difficult of the energy scenarios generate significant losses only in the cases where,,,. savings constraint is applied. Without such a constraint the economy appears flexible enough to take quite sizable shocks to the energy sector in its stride.8
The RFF modelers note that their model restrictions may be severe-the demand for energy is probably too high, there is the possibility of a relaxation or, constraints to coal use in the long run, stockpiling of oil imports may be looked upon as a sounder policy than restricting imports, etc. Thus smaller aggregate welfare losses are indeed possible.
However, the modelers also spagest that by emphasizing aggregate losses one overlooks the energy inI_eractions as they affect individuals. These include such things as income and wealth distributions among persons, factors of production and geographic regions. Other factors ,not considered in the model involve safety, security, to-i-id some environmental consequences which are difficult, if not impossible, to quantify. The RFF modelers themselves admit that the very items the model can not deal with. quantitatively, such as safety and national security, may vN-ell ru.le the day" when it comes to decisior-s b-\,polie.vmakers.

8 Ibid., p. 171.


The IEA (Institute for Energy Analysis) general equilibrium two sector energy demand model as developed by David B. Register and. James A. Edmonds provides lone term forecasts of ener v demand and GNP. According to the modelers, this model can be used to study questions such as: the economic impact of the transition from inexpensive energy sources to more expensive energy sources; the consequences of modifying the historical relationship between growth in the GNP and growth in energy; or the impact of large increases in the price of energy on economic growth."
The IEA model is based on five theoretical assumptions:
A. 1. The economy can be characterized by two sectors: energy and materials.
A. 2. The two fundamental factor inputs to the economy are capital and labor.
A. 3. Energy is used only as an intermediate good.
A. 4. The materials sector is homogeneous of degree one in its inputs of capital, labor, energy, and materials, and is assumed to be perfectly competitive (that. is, a doul-Aing of all inputs leads to a doubling of all outputs).
A. 5. The energy sector is not homogeneous of degree one but is characterized by decreasing returns to scale, and is assumed to be imperfectly competitive.10
The authors refer to the model as a parametric (versus econometric) model because the theoretical structure of the model and the values of' the parameters which control the model have been determined by assumption and not by econometric methods. The major assumptions. plugged into the model are then just that-assumptions. Models treat the assumptions as fact-thus the source of error is not the model but the degree to which we cannot see the future.
According to the results of this model, reduction in energy demand (based on the given assumptions) does not require an equal reduction in GNP. These runs also indicate that "substantial" energy conservation is possible in the elastic case. "Since fewer resources are needed to produce energy the elastic case provides more GNP for both the free supply and constrained supply assumption.""
The IEA model gives great attention to demographic changes. Ilt is, this that tends to lead to lower growth rates for energy use in this. versus the other models.
The authors conclude that the transition from cheap energy to expensive energy will have a substantial but not catastrophic impact on GNP, even for the case with tight coal and nuclear constraints. Furthermore, the impact is substantially reduced if the economy can freely substitute one factor of production for another.
Lester Lave, in his critique finds this model, like ETA-Macro, "aggregates the 95 percent of the economic activity (which is other than energy) into a few sectors, while displaying detail on the energy portion." Lave also finds fault with the energy.pricine mechanism. "* * The authors decided to treat energy pricing simply and to. set price equal to average (versus marginal) cost * It means that energy prices are biased downward * and thus that energy use is likely to be biased upward." He concludes that one must approach results of such a model with great caution."
Ibid., P. 199.
11) Id. P. 200.
Ibid., p. 20
Ibid., pp. 285-6.


29 3 1262 05602 3814

The Scenarios

The table below presents a brief characterization of the fouir s-cenarios run by each model. Primary energy use is shown in quwkprices of petroleum, natural gas and electricity Ii dollars per11 Million
B tu or index numbers. Total consumption in the ecoanis 1"Shown InI
billions of dollars (using various years as a base). Altho1IghJ thle
numbers appear to be comparable, Hitch notes this is not always
possible and is (lone only at the reader's own risk.
As was noted earlier there is little variance ini primary energy use
across the models and scenarios through the year 2000; however,
this changes considerably between 2000 and 2020 since the constraint,
on resources is binding. Perhaps more important is that total consumption in the economy in billions of dollars varies little across
the four cases even when extended through 2020. This seems to imply
a great flexibility in the economy when it is given sufficient time

2000 2020
1975 U F CE UI C1 U 1 CEm U1 C1

Primary energy use:
ETA-Macro-------------------------- 107 104 126 118 151 119 192 165
DESOWM-LITM------------------ 71--------------- 139 112--------------- 195 134
lEA------------------------- 71 134 116 149 141 ---------------RFF ------------------------ 77 124 112 171 148 203 148 292 232
Price of liquids:
ETA-Macro ------------------ 0.80 2.70 2.80 3.30 3.30 5.00 5.00 5.00 5.00
DESOM-LITM---------------- 2.26 --------------- 3.25 4.27 --------------- 4.95 11.46
lEA (index number) ------------------ 2.87 2. 67 2.59 4. 10 -------------RFF------------------------- 3.64 5.10 5.70 5.10 9.47 5.17 9.47 5.17 947
Price of natural gas:
ETA-Macro---------------- 0. 80 2.70 2. 80 3. 30 3.30 5. 00 5. 00 5. 00 5. 00
DESOM-LITM---------------- 2.50 --------------- 3.52 4.54 --------------- 5.24 12.58.
l EA (index number) ------------------ 1.79 2.84 2.17 5.42 -------------RFF------------------------- 1.29 4.54 4.30 4.54 8.84 4.54 8.84 4.54 8.84i
Price of electricity:
ETA-Macro ------------------ 3. 22 6. 36 6. 80 6. 48 9. 88 6.68 11.87 7. 21 11. 75,
DESOM-LITM---------------- 10.51 --------------- 10.06 16.24 --------------- 10.49 15.37
lEA (index number) ------------------ 1.27 1.39 1.56 1.65 ---------------RFF------------------------- 7.91 8.30 9.50 8.30 15.61 8.70 15.61 8.70 15.61
ETA-Macro 1----------------------------- 2, 795 2, 790 2, 763 2, 757 4, 375 4, 304 4, 244 4,109
DESOM-LITM ----------------- 770 --------------- 1,808 1,727 --------------- 2,928 2,297
1EA (1967 dollars) -------------- 490 1,511 1,457 1,484 1,365-------------RFF ------------------------ 678 1, 663 1, 668 1, 649 1, 633 2, 652 2, 586 2, 650 2, 492
RFF with savings constraint (Elasticity equals 0.3)-------------- 678 1, 530 1, 512 1, 500 1, 419 2,609 2,471 2, 548 2,336
RFF with savings constraint (GNP
from model) ----------------- 678 1,248 1, 171 1, 200 1,044 2, 324 1, 952 2, 096 1, 466
C/pop less environmental damage-RFF ----------------- 2,947 6, 136 6, 102 6, 122 6,090 8, 874 8, 567 8, 820 8, 440,

I Includes private consumption plus government
Uz: Unconstrained development of resources and 0.75 elasticity of substitution. CE: Constrained development of resources and 0.75 elasticity of substitution. Ui: Unconstrained development of resources and 0.25 elasticity of substitution. Ci: Constrained development of resources and 0.25 elasticity of substitution.

Electricity prices, as one would expect, show substantial differ-.
ences between the constrained and unconstrained cases when drawn,
out to 2020. This indicates that:

The variation in resource availability is more important than the variance in the elasticity of substitution. NVhen the constraints on coal, oil shale, and nuclear reactors are lifted, the value of the elasticity of substitution makes little difference. Thus if we can learn either to mine and burin these fuels cleanly, or learn to live with a polluted environment, there is little indication of an energy problem in the United States by 2020. However, the combination of constrained resources and a low elasticity of substitution indicates an energy problem by 2020.
The models and scenarios indicate there are two main questions
the first is the elasticity of substitution, and the second is the extent to which energy is expected to be curtailed. Without a more definite handle on elasticity, the credibility of basing decisions on model results is subject to question. To a degree the elasticity problem is a product of an uncertain regulatory framework. If this were more definite, the choices would be clearer, and the substitutions that had to take place in the long run, would. This seems to imply that the economy could adjust in good stead over the long run, given free market price signals. By not allowing the economy sufficient time to adjust as mandated by 4(correct" price signals, Government policies which a pear to some our best interests now, may prove detrimental in Ye long run. The bottom line is that the economy must have. sufficient time to adapt to changing situations in order to avoid possible severe consequences.
The foremost conclusion of the model comparison exercise 111as been the relative agreement that energy and economic activity need not be so directly tied as long as the economy and society has sufficient leeway and time to adjust.
NOTE.-This appendix is presented as a brief outline of the models, to aid the reader in becoming more aware of other energy-economy models. It is a "state of the axt" assessment. The book "Modeling Energy-Economy Interactions: Five Approaches" as cited earlier, gives a much more comprehensive view of the models we speak of here.
23 Ibid., p. 290.