Increased Automobile Fuel Efficiency and Synthetic Fuels

Increased Automobile Fuel Efficiency andSynthetic Fuels: Alternatives for Reducing

Oil Imports

September 1982

NTIS order #PB83-126094

Library of Congress Catalog Card Number 82-600603

For sale by the Superintendent of Documents,U.S. Government Printing Office, Washington, D.C. 20402

Foreword

This report presents the findings of an assessment requested by the Senate Com-mittee on Commerce, Science, and Transportation. The study assesses and comparesincreased automobile fuel efficiency and synthetic fuels production with respect to theirpotential to reduce conventional oil consumption, and their costs and impacts. Con-servation and fuel switching as a means of reducing stationary oil uses are also con-sidered, but in considerably less detail, in order to enable estimates of plausible futureoil imports.

We are grateful for the assistance of the project advisory panels and the many otherpeople who provided advice, information, and reviews. It should be understood, how-ever, that OTA assumes full responsibility for this report, which does not necessarilyrepresent the views of individual members of the advisory panels.

Director

Automobile Fuel Efficiency Advisory Panel

Maudine R. Cooper*National Urban League, Inc.

John FerronNational Automobile Dealers Association

Michael J. Rabins, ChairmanWayne State University

Donald FriedmanMinicar, Inc.

Herbert FuhrmanNational Institute for

Automobile Service Excellence

James M. GillThe Ethyl Corp.

R. Eugene Goodson**Hoover Universal, Inc.

Charles M. Heinen***Chrysler Corp.

Synthetic Fuels Advisory

Harvey O. BanksCamp Dresser McKee, Inc.

Ellen BermanConsumer Energy Council of America

Leslie BurgessFluor Corp.

Frank CollinsOil, Chemical and Atomic Workers

International Union

Thomas F. EdgarUniversity of Texas

Louis FrickE. 1. du Pent de Nemours & Co., Inc.

Bernard GillespieMobil Research and Development Co.

John B. HeywoodMassachusetts Institute of Technology

John HoldenFord Motor Co.

Maryann N. KellerPaine, Webber, Mitchell, & Hutchins

Paul LarsenGMC Truck and Coach Division

Robert D. NellConsumers Union

Kenneth OrskiGerman Marshall Fund of the United States

Howard Young*United Auto Workers

PanelHans Landsberg, Chairman

Resources for the Future

Robert HowellRefinery ConsultantChairman, Synfuels SubcommitteeSierra Club

Sheldon LambertShell Oil Co.

John L. McCormickEnvironmental Policy Center

Edward MerrowRand Corp.

Richard K. PefleySanta Clara University

AlIan G. PulsipherTennessee Valley Authority

Robert ReillyFord Motor Co.

Fred WilsonTexaco, Inc.

NOTE: The Advisory Panels provided advice and comment throughout the assessment, but the membersdo not necessarily approve, disapprove, or endorse the report for which OTA assumes full responsibility.

● Resigned from panel.● ● Formerly with Institute for Interdisciplinary Engineering Stud&, Purdue University.* ● ● Retired from Chrysler cOrp.ICurrently with LTV Corp.

iv

OTA Increased Automobile FuelSynthetic Fuels Project Staff

Efficiency and

Lionel S. Johns, Assistant Director, OTAEnergy, Materials, and International Security Division

Richard E. Rowberg, Energy Program Manager

Synthetic Fuels and Project Coordination Increased Automobile Fuel Efficiency

Thomas E. Bull, Project Director John A. Alic, Automobile TechnologyPaula J. Stone, Assistant Project Director Marjory S. Blumenthal, Cost Analysis andSteven E. Plotkin, Environmental Analysis Economic and Social Impacts

and Executive Summary Larry L. Jenney, Automobile TechnologyRichard W. Thoreson, Economic Impacts Richard Willow, Electric Vehicles

Contributors

David Strom Franklin TugwellDavid Sheridan, Editor

Administrative Staff

Lillian Quigg Edna Saunders Marian Grochowski Virginia Chick

Contractors

E. J. Bentz & Associates, Inc., Springfield, Va.Wright Water Engineers, Inc., Denver, Colo.Michael Chartock, Michael Devine, and Martin Cines,

University of Oklahoma, Norman, Okla.General Research Corp., Santa Barbara, Calif.

OTA Publishing Staff

John C. Holmes, Publishing Officer

John Bergling Kathie S. Boss Debra M. Datcher Joe Henson

Acknowledgments

OTA thanks the following people who took time to provide information or review part or all of the study.

John AcordMontana Department of Natural Resources

and Conservation

David AlberswerthWestern Organization of Resource Councils

Louis E, AllenWyoming State Engineer’s Office

Morris AltshulerU.S. Environmental Protection Agency

Marty AndersonTransportation Systems CenterDepartment of Transportation

Ray BaldwinPlanning Department, Garfield County, Colo.

Craig BellWestern States Water Council

Martin J. Bernard, IllCenter for Transportation ResearchArgonne National Laboratory

Don BischoffCalifornia Energy Commission

William BoehlyNational Highway Traffic Safety Administration

Charles BrooksU.S. Department of Energy

A. L. BrownEXXON Coal Resources, USA, Inc.

George ByronTransportation Systems CenterDepartment of Transportation

James ChildressNational Council on Synthetic Fuels Production

George ChristopulosWyoming State Engineer

David ColeOffice for the Study of Automotive TransportationUniversity of Michigan

George DavisU.S. National Committee for International

Hydrology Program

Clarence DitlowCenter for Auto Safety

C. Gilmore DuttonAppropriations and Revenue CommitteeKentucky Legislative Research Commission

Vernon FahyNorth Dakota State Engineer

Jack FitzgeraldU.S. Environmental Protection Agency

Gary FritzMontana Department of Natural Resources

and Conservation

Jack GilmoreDenver Research InstituteUniversity of Denver

Harris GoldWater Purification Associates

David GusheeCongressional Research Service

Jack HannonE. 1. du Pent de Nemours & Co., Inc.

Richard HildebrandU.S. Department of Energy

Frank von HippelCenter for Energy and Environmental StudiesPrinceton University

Edward HoffmanConsultant

Harriet HoltzmanEnvironmental Policy Institute

Darrell HowardTennessee Valley Authority

Richard HowardSouth Dakota Department of Water

and Natural Resources

Charles HudsonArgonne National Laboratory

James JohnDepartment of Mechanical EngineeringOhio State University

Arvid Joupi

Joseph KanianthraNational Highway Traffic Safety Administration

vi

Sarah La BelleArgonne National Laboratory

Jonathan LashNatural Resources Defense Council

Minh-Triet LethiOffice of Science and Technology Policy

Kevin MarkeyFriends of the Earth

David MarksDepartment of Civil EngineeringMassachusetts Institute of Technology

J. P. MatulaEXXON Research & Engineering Co.

Dennis McElwainAFL-CIO

C. E. McGeheeBattlement Mesa, Inc.

Gary McKenzieNational Science Foundation

Ron McMahanAbt/West

Gene MeyerShell Oil Co.

Carl NashNational Highway Traffic Safety Administration

Christopher PalmerNational Audobon Society

Laszlo PastorDRAVO Corp.

E. PiperEnergy Development Consultants, Inc.

James PosewitzMontana Department of Fish, Wildlife,

and Parks

Frank RichardsonNational Highway Traffic Safety Administration

Richard D. RidleyOccidental International Corp.

H. RobichaudANG Coal Gasification

Phil ScharreTennessee Valley Authority

William SchriverU.S. Department of Labor

Roger ShunU.S. Department of Energy

Tom SladekColorado School of Mines Research Institute

Catherine SlesingerGeneral Accounting Office

John H. SmithsonU.S. Department of Energy

Earl StakerUtah Deputy State Engineer

Dick StrombotneNational Highway Traffic Safety Administration

George TauxeScience and Public Policy ProgramUniversity of Oklahoma

David TundermannU.S. Environmental Protection Agency

Wallace TynerDepartment of Agricultural EconomicsPurdue University

Richard A. WaldeGulf Research and Development Co.

David WalkerColorado Water Conservation Board

George WangBechtel Group, Inc.

R. M. WhamOak Ridge National Laboratory

Mary Pat WilsonWESTPO

Edward P. YarmowichU.S. Department of Energy

vii

Contents

Chapter Page

I. Executive Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3

2.introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29

3. Policy . . . . . . . . . . . . . .. .. .. .. ... . . $........ . . . . . . . . . . . . . . . . . . . . . . 35

4. issues and Findings. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67

5. Increased Automobile Fuel Efficiency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105

6. Synthetic Fuels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157

T. Stationary Uses of Petroleum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183

8. Regional and National Economic Impacts . . . . . . . . . . . . . . . . . . . . . . . . . . . 195

9. Social Effects and Impacts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 219

10. Environment, Health, and Safety Effects and Impacts . . . . . . . . . . . . . . . . . . 237

11. Water Availability for Synthetic Fuels Development . . . . . . . . . . . . . . . . . . . 279

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 291

Chapter 1

Executive Summary

Contents

Page

Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3

Conclusions and Comparisons . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3Import Reductions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3costs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5Technological and Economic Risks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6Additional Bases for Comparison-Environmental Social and

Economic Effects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7

policy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8Stimulating Oil Import Reductions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8Dealing With Other Effects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10

Overview of the lmport Reduction Options . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10Increased Automobile Fuel Efficiency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10Synthetic Fuels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16Reducing Stationary Uses of Oil. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22Electric Vehicles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24

TABLES

Table No.

I. Projected Average Fuel Economy, 1985-2000 . . . . . . . . . . . . . . . . .2. Domestic Investment for Increased Fuel Efficiency . . . . . . . . . . . . . . . . . . . . .3. Consumer Costs for increased Automobile Fuel Efficiency, Without

Discounting Future Fuel Savings, Moderate Shift to Smaller Cars . . . . . . . . .4. Price of Synthetic Fuels Required To Sustain Production Costs . . . . . . . . . . .5. Two Comparisons of the Environmental impacts of Coal-Based Synfuels

Production and Coal-Fired Electric Generation . . . . . . . . . . .6. Summary of Annual Energy Requirements To Displace 2.55

Stationary Fuel Oil Use . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

FIGURES

Figure No.

l. Potential Oil Savings Possible by the Year 2000 . . . . . . . . . .

. . . . . . . . . . . . .MMB/D of. . . . . . . . . . . . .

. . . . . . . . . . . . .2. Estimated Investment Costs for the Oil Import Reduction Options . . . . . . . .

Page

1113

1417

20

23

Page

46

Chapter 1

Executive Summary

INTRODUCTIONIn 1981, U.S. oil imports averaged 5.4 million

barrels per day (MMB/D)–approximately 34 per-cent of its oil consumption and 15 percent of itstotal energy use. This is potentially a serious riskto the economy and security of the United States.Furthermore, recovery from the current recessionwill increase demand for oil and, although cur-rently stable, domestic oil production is likely toresume a steady decline in the near future.

Several options exist for reducing oil imports.However, even with moderate increases in auto-mobile fuel efficiency, moderate success at de-veloping a synthetic fuels industry and the ex-pected reduction in stationary use of fuel oil, U.S.oil imports could still be over 4 MMB/D by 2000,if the U.S. economy is healthy and has not under-gone unforeseen structural changes that mightreduce oil demand well below projected levels.

Only with vigorous promotion of all three op-tions and technological success can the Nationhope to eliminate oil imports before 2010.

Congress faces several decisions on how to re-duce the U.S. dependence on imported petro-leum. Two options, increased automobile effi-ciency and synthetic fuels, are particularly likelyto be subjects of congressional debates. First,Congress may want to consider new incentivesto increase auto fuel efficiency beyond that man-dated by the 1985 CAFE (Corporate Average FuelEfficiency) standards. Second, Congress will haveto decide whether to continue into the secondphase of the program to accelerate synfuels devel-opment under the Synthetic Fuel Corp. (SFC). Thepurpose of this report is to assist Congress in mak-ing these decisions and comparing these optionsby exploring in detail the major public and privatecosts and benefits of increased automobile fuelefficiency and synthetic fuels production. A thirdoption for reducing imports-increased efficien-cy and fuel switching in stationary (nontranspor-tation) oil uses—is examined briefly to allow anassessment of potential future levels of oil im-ports. Finally, electric-powered automobiles areexamined.

CONCLUSIONS AND COMPARISONSImport Reductions

In the judgment of the Office of TechnologyAssessment, increased automobile fuel efficiency,synthetic fuels production, and reduced station-ary (nontransportation) use of oil can significantlydecrease U.S. dependence on oil imports dur-ing the next two to three decades. indeed, reduc-ing oil imports as quickly as possible requires thatall three options be pursued. Electric cars areunlikely to play a significant role, however.

Although a precise forecast of the future contri-butions of the import reduction options is not fea-sible now, it is possible to draw some generalconclusions about their likely importance and toestimate what their contributions could be underspecific circumstances (see fig. 1).

First, increases in auto fuel efficiency will con-tinue, driven by market demand and foreign

competition. OTA believes that, with strong andconsistent demand for high fuel efficiency, thereis a good chance that actual average new-car fuelefficiencies would be greater than OTA’s low sce-nario in which average new-car fuel economy*was projected to be:

30 miles per gallon (mpg) . . . . . . . in 198538 mpg . . . . . . . . . . . . . . . . . . . . ., in 199043 mpg . . . . . . . . . . . . . . . . . . . . ., in 199551 mpg . . . . . . . . . . . . . . . . . . . . . . in 2000

with a moderate shift in demand to smaller cars.Although this scenario is based on modest techni-cal expectations, it is dependent on favorablemarket conditions. Domestic automakers are un-likely to commit the capital necessary to continue

*EPA values, based on 55 percent city, 45 percent highway. On-the-road fuel economy is expected to average about 10 percent less.

3

4 ● Increased Automobile Fuel Efficiency and Synthetic Fuels: Alternatives for Reducing Oil /reports

Figure 1 .—Potential Oil Savings Possible by the Year 2000a (relative to 1980 demand)

. .

Synfuels

High

n

Conservation andfuel switching instationary uses

of fuel oil

H i g h A d d i t i o n a l

Incf u

High

Low

‘- reased auto

u

potential

el efficiencysavings beyond

base case

n

Low Oil savingsincorporated in

base case

0

Low

SOURCE: Off Ice of Technology Assessment.

the current rapid rate of increase in efficiencyunless they improve their sales and profits.

If the industry is able to attain the fuel efficien-cy levels shown above, the United States wouldsave 800,000 barrels per day (bbl/d) of oil by 2000compared with the case where post-1 985 new-car efficiency remained at 30 mpg. The savingswould increase to at least 1.1 MMB/D by 2010because of continued replacement of older, lessfuel-efficient automobiles.

With a poorer economic picture and weakerdemand for high fuel efficiency, new-car efficien-cies could be 40 mpg or less by 2000, with cor-respondingly lower savings. Achieving 60 to 80mpg by 2000 would require not only favorableeconomic conditions and strong demand for fuelefficiency, but also relatively successful technicaldevelopment.

Second, substantial contributions to oil importreductions from production of synthetic fuelsappear to be less certain than substantial contri-butions from the other options. Potential syn-

fuels producers are likely to proceed cautiouslyfor the following reasons: 1 ) investment costs arevery high (even with loan guarantees covering75 percent of project costs); 2) there is a fairlysmall differential between the most optimistic ofOTA’s projected synfuels production costs andthe current price of oil; 3) investors are now un-ertain about future increases in the real price ofoil; and 4) there are high technological risks withthe first round of synfuels plants (possibly exacer-bated by the cancellation of the Department ofEnergy’s (DOE) demonstration program).

OTA projects that, even under favorable cir-cumstances, fossil-based production of synthetictransportation fuel could at best be 0,3 to 0.7MMB/D by 1990 and 1 to 5 MMB/D by 2000. Bio-mass synfuels could add 0.1 to 1 MMB/D to thistotal by 2000. In less favorable conditions—forexample, if the SFC financial incentives werewithdrawn—it appears unlikely that even the low-er fossil synfuels estimate for 1990, and perhaps2000, could be achieved unless oil prices increasemuch faster than they are currently expected to.

Ch. 1—Executive Summary ● 5

Achieving much more than 1 MMB/D of syn-fuels production by 2000 would require fortuitoustechnical success and either: 1 ) unambiguouseconomic profitability or 2) continued financialincentives requiring authorizations considerablylarger than those currently assigned to SFC.Achieving production levels near the upper limitsfor 2000 are likely to be delayed, perhaps by asmuch as a decade, unless there is virtually a “warmobilization’ ’-type effort.

Third, there are likely to be large reductionsin the stationary use of fuel oil (currently 4.4MMB/D) in the next few decades. With just cost-effective conservation measures, stationary fueloil use could be reduced significantly. Additionalconservation measures by users of electricity andnatural gas could make enough of these fuelsavailable to replace the remaining stationary fueloil use by 2000. Total elimination of stationaryfuel oil use by 2000 is unlikely, however, becausesite-specific factors and differing investor paybackrequirements will mean that a significant fractionof the numerous investments needed for elimina-tion will not be made.

Fourth, even a 20-percent electrification of theauto fleet—a market penetration that must beconsidered improbable within the next severaldecades—is unlikely to save more than about0.2 MMB/D. Electric cars are most likely to re-place small, low-powered–and thus fuel-efficient—conventional automobiles, minimizing poten-tial oil savings.

Plausible projections of domestic oil production—expected by OTA to drop from 10.2 MMB/Din 1980 to 7 MM B/D or lower by 2000—suggestthat oil imports could still be as high as 4 to 5MM B/D or more by 2000 unless imports are re-duced by a stagnant U.S. economy or by a re-sumption of rapidly rising oil prices. * Achievinglow levels of imports-to perhaps less than 2MM B/D within 20 to 25 years–is likely to requirea degree of success in the three major optionsthat is greater than can be expected as a resultof current policies.

*Rapidly rising oil prices are unlikely to occur simultaneously witha stagnant U.S. economy unless the economies (and oil import re-quirements) of Europe and others are thriving at the same time.

costs

Except for stationary fuel oil reductions, eco-nomic analysis of the options for reducing oil im-ports involves a comparison of tentative cost esti-mates for mostly unproven technologies that willnot be deployed for 5 to 10 years or more. Evenif costs were perfectly estimated for today’s mar-ket (and the estimates are far from perfect), dif-ferent rates of inflation in the different economicsectors affecting the options could dramaticallyshift the comparative costs by the time technol-ogies are actually deployed. Figure 2 presentsOTA’s estimates for the investment costs for alloptions except electric cars. The costs are ex-pressed in dollars per barrel per day, which is theamount of investment needed to reduce petro-leum use at a rate of 1 bbl/d. * In OTA’s judgment,the estimated investment costs (in dollars perbarrel per day) during the 1990’s of automobileefficiency increases, synthetic fuels production,and reduction of stationary uses of oil are essen-tially the same, within reasonable error bounds.If Congress wishes to channel national invest-ments preferentially into one of these options,differentials in estimated investment costs can-not provide a compelling basis for choice.

On the other hand, investments during the1980’s to reduce stationary oil use (from the cur-rent 4.4 to 3 MM B/D or less by 1990) and in-crease automobile fuel efficiency (to a 35 to 45mpg new-car fleet average by 1990) are likely tocost less than the 1990-2000 investments in anyof the options.

Electric vehicles are likely to be very expensiveto the consumer—costing perhaps $3,000 moreper vehicle than similar, conventional auto-mobiles or $300,000 to $400,000/bbl/d of oilsaved, (The latter is not strictly comparable to in-vestment costs for the other options. ) If batteriesmust be replaced at moderate intervals, whichis necessary today, the total costs of electric carswould escalate.

*This measure was chosen in order to avoid problems that arisewhen comparing investments in projects with different lifetimes andfor which future oil savings may be discounted at different rates.

6 • Increased Automobile Fuel Efficiency and Synthetic Fuels: Alternatives for Reducing Oil Imports

Figure 2.—Estimated Investment Costs for the Oil Import Reduction Options

Increased automobilefuel efficiency

I 45-80 mpgby 2000a

40-65 mpg , ,by 1995a-

Synfuels Reduction in stationaryuses of fuel oilin the 1990’s

35-50 mpgby 1990a

Estimated 1985 cost ofconventional oil and gas

exploration anddevelopment in the United States

SOURCE: Office of Technology Assessment.

Technological and Economic Risks

The general perception of the technologicaland economic risks of the import reduction op-tions is: 1 ) that the reduction of stationary oil usehas comparatively predictable costs and few tech-nological risks; 2) that synthetic fuels have severeeconomic and technological risks; and 3) that in-creased auto fuel efficiency has moderate eco-nomic and technological risks. OTA’s analysis in-dicates that these perceptions are correct onlyto a limited extent.

Ž Although the costs and technology of fuelswitching are well known and involve little risk,the success of retrofitting a given building toincrease its energy efficiency often cannot be

●

●

●

accurately predicted because of site-specificconsiderations that cannot be adequatelyquantified.The differences in risks between synfuels devel-opment and increased automobile fuel efficien-cy are less a matter of overall magnitude thanof timing.Synfuel production involves considerable tech-nical and economic risks for the first round ofcommercial-scale facilities, but once full-scaleprocess units have been demonstrated the riskfor future plants should drop substantially.Some increases in automobile fuel efficiencycan be implemented with negligible technolog-ical and small economic risks, but increasesto very high efficiencies do involve significant

Ch. l—Executive Summary Ž 7

technical and economic risks. Also, as thenumber and rate of changes in automobiles in-creases, there is increased risk that consumerswill not accept the automobiles and that insuffi-cient development and testing will lead to pooron-the-road performance and/or product re-calls.

Additional Bases for Comparison—Environmental, Social, and

Economic Effects

Increased auto fuel efficiency may reduce ve-hicle safety as cars are made smaller and lighter.But in all but extreme cases of vehicle size reduc-tion, improvements in vehicle design and in-creased passenger use of safety restraints havethe potential to offset any effects of reduced sizeand weight on the vehicle’s protection of its oc-cupants in a crash.

Continued pressure for increased fuel efficien-cy will dictate new plant investments which willreinforce the ongoing restructuring of the U.S.auto industry. This restructuring involves a shiftin manufacturing away from the traditional pro-duction centers to the Sun Belt and overseas, andstronger industry ties with foreign manufacturers.The composition and size of the manufacturingwork force may evolve towards a greater propor-tion of skilled workers but fewer workers overall.Increased sophistication and capital investmentmay be required for vehicle maintenance. A re-duction in the number of suppliers to the autoindustry may also result.

Large-scale synthetic fuels production wouldgenerate significant amounts of toxic sub-stances, posing risks of health damage to work-ers and possible risks to the public through con-tamination of ground waters or by smallamounts of toxics left in the fuels. There shouldnot be any technological barrier to adequatecontrol of these substances, but OTA concludesthat there are substantial reasons to be con-cerned about the adequacy both of proposedenvironmental protection systems and of the ex-isting regulatory structure.

Other important effects of synfuels produc-tion stem from the very large scale of both theindividual projects and, potentially, the industryas a whole. These may overwhelm the social andeconomic resources of nearby population cen-ters, especially in sparsely populated areas of theWest. At national production levels of a few mil-lion barrels per day, impacts from coal and shalemining and population pressures on wildernessareas and other fragile ecosystems can be sub-stantial even in comparison with major industriessuch as coal-fired power generation. On the otherhand, conventional air pollution problems fromsuch plants are likely to be considerably less thanthose associated with similar amounts* of coal-fired power generation.

Finally, although water requirements for syn-fuels are a small fraction of total national con-sumption, growth of a synfuels industry couldeither create or intensify competition for water,depending on both regional and local factors.Such competition is of special concern in the aridWest. Unfortunately, a reliable determination ofboth the cumulative impacts on other water usersand, in some instances, the actual availability ofwater for synfuels development is precluded byphysical and institutional uncertainties, changingpublic attitudes towards water use priorities, andthe analytical shortcomings of existing studies.

However, in areas where there are relativelyfew obstacles to transferring water rights (e.g., asis currently the case in Colorado), developersshould be able to obtain the water they needbecause their consumption per barrel of oil pro-duced is small enough to enable them to pay arelatively high price without significantly affect-ing the final cost of their products.

Electric vehicles, if they are ever produced inlarge quantities, could have an important posi-tive environmental effect-the reduction ofautomobile exhaust emissions and resulting im-provements in urban air quality.

*On a “per unit of coal used” basis.

9a-2 e 1 ~ - a 2 - 2 : I: IIJ 3


POLICYOTA’s analysis points to two conclusions that

may warrant congressional consideration ofchanges in current Federal energy policy.

First, current policies affecting investments inenergy conservation and domestic energy pro-duction are not likely to result in levels of oil im-ports below 4 MMB/D in 2000, if the U.S. econ-omy is healthy and has not undergone unfore-seen structural changes that might reduce oil de-mand well below projected levels. During thenext 20 years, OTA expects that, under thesepolicies, oil import reductions due to syntheticfuels production and decreased stationary andautomobile oil use will be partially offset by adecrease in domestic production of conventionaloil. Reducing net oil imports to 1 or 2 MMB/Dor less by 2000 is likely to require more vigorouspursuit of all options for reducing domestic con-sumption of conventional oil products. On theother hand, elimination of current conservationand synthetic fuels production policies couldcause imports to range from 5 to 6 MMB/D by2000 under these same economic conditions.

Second, current policies may not provide soci-ety with adequate protection from some of theadverse side effects of synthetic fuels develop-ment and increased automobile fuel efficiency.Of particular concern are possible reductions inautomobile crash safety (as the number ofsmaller, more fuel-efficient cars increases), inade-quate control of toxic substances from synfuelsdevelopment, and adverse socioeconomic effectsfrom both options.

Because of the large technical, economic, andmarket uncertainties inherent in the analyses ofoil displacement options, Congress may wish toemphasize flexible incentives with provisions forperiodic review and adjustment. A stable com-mitment to oil import displacement will be neces-sary, however, to maximize the effect of suchpolicies.

Stimulating Oil Import Reductions

The level of oil imports at the turn of the cen-tury will be determined by market forces, modi-fied by Government policy towards oil supply

and demand. The imposition of Federal policyon the workings of the private market generallyis justified on the basis of the market’s failure tovalue public costs and benefits. A particularly im-portant public cost of U.S. dependence on im-ported oil, for example, is the national securityproblem imposed by political instability in theMiddle East and the resulting potential for oil cut-offs. Although the precise magnitude of thesecosts is debatable, most people would agree thatthey are significant ($5 to $50/bbl depending onvarious circumstances) and that the private mar-ket generally does not take them into account.

Efforts to displace imports also have both publicand private costs. In addition to the potential sideeffects just mentioned, Government interferencein the oil marketplace can cause significant misal-Iocations of resources. Congress will have to bal-ance costs and benefits, which cannot be re-duced to common measures and which changewith time, in a complex tradeoff.

One policy option to displace imports is an en-ergy tax, either on oil imports or on oil in gen-eral. Both taxes have the advantage of encourag-ing alternatives to conventional oil consump-tion without predetermining which adjustmentswould be made. They could be used to provideconsistent price signals to the market—to assurethe auto industry, for example, that demand forfuel-efficient cars would continue and to assuresynfuels developers that they would receive atleast a constant real price for their products. im-posing a tax only on transportation fuels wouldsend the same signal to both the auto industryand to producers of synthetic transportation fuels,but this preferential treatment would be at theexpense of other conservation or synfuelsproduction investors.

All of these petroleum taxes also have a numberof other effects which must be considered. Forexample, a tax only on oil imports leads to anincome transfer from domestic oil consumers todomestic oil producers; and all oil taxes can leadto reduced international competitiveness of do-mestic industries heavily dependent on oil, suchas the petrochemical industry.

Ch. I—Executive Summary ● 9

policies can also be directed specifically at theautomobile or synfuels industries. The most effec-tive of these options will be those that directlyaddress the factors that shape, direct, and limitthe contributions that the automobile and syn-fuels industries can make to import displacement.

The critical factors that determine the pace ofincreased automobile fuel efficiency are con-sumer demand for fuel efficiency and the finan-cial health of the domestic auto industry. If theindustry is uncertain about demand, it will be re-luctant to make the expensive investments. Andwith continued poor sales, the industry will beless able to afford them.

Aside from energy taxes, Congress can main-tain and stimulate consumer demand for fuel ef-ficiency by a variety of measures that would raisethe relative costs to consumers of owning ineffi-cient cars. For example, registration fees (onetime or annual) and purchase taxes or subsidiesare incentives that can be directly linked to fuelefficiency. However, fuel-efficiency incentivesthat do not discriminate with respect to car sizewould tend to increase sales of small cars at theexpense of larger cars. Such discrimination mighthurt domestic manufacturers, which have beenmost vulnerable to foreign competition in thesmall-car market.

Congress can also choose policies aimed atauto production such as continuing to requiremanufacturers to improve fuel efficiency bymeans of stricter CAFE or similar standards thatwould ensure increased fuel efficiency even if de-mand for this automobile attribute is low. Thisregulatory route might reduce some risks to auto-makers by requiring all to make similar invest-ments. On the other hand, car sales may sufferif the costs of the fuel savings—either in highersticker prices or reductions in some desirable ve-hicle attributes–are higher than consumers arewilling to pay. Fuel-economy requirements arelikely to be perceived by the industry as exceed-ingly risky unless the requirements are accom-panied by measures to stimulate demand or toease the resulting financial burden on the auto-makers.

To help ensure that the fuel-efficient cars areactually bought and that the automakers can ac-

quire the capital needed for increasing fuel effi-ciency, Congress may also wish to directly pro-mote sales of fuel-efficient cars. A low-interest-rate loan program (with interest rates tied to fuelefficiency) is one potentially effective mechanism.Congress may also wish to consider awarding di-rect grants or loan guarantees for qualifying in-vestments in auto manufacturing facilities.

The factors that determine the pace of synfuelsdevelopment are the high degree of technical un-certainty and the continuing uncertainty aboutfuture oil prices. Both areas of uncertainty con-tribute to doubts about profits.

Current Federal policy maintains the valuableincentives associated with SFC, but reemphasizesDOE’s research, development, and demonstra-tion programs. The loan guarantee mechanismoffered by SFC significantly improves the proba-bility of financial success for a developer andprobably will be necessary to ensure even a fewhundred thousand barrels per day of synfuels pro-duction by the early 1990’s. Several major risksto synfuels investors remain, however. Cost over-runs couId nuIlify any potential profits becausedevelopers must base their product prices on themarket prices of competing fuels rather than onsynfuels production costs. It is also probable thatseveral first generation commercial-scale unitswill function poorly, and rapid expansion of theindustry may thereby be delayed.

Since the SFC program appears to be attract-ing the capital needed to build and demonstratea series of first generation commercial-scale pro-duction units, cancellation of DOE’s programsmay not turn out to be particularly harmful to syn-fuels development if the first plants performwell. However, cancellation of the demonstra-tion program probably will mean that fewer tech-nologies reach the stage where SFC support ispossible. Reemphasis of development programsmay also delay findings that would be useful infixing the technical problems that are likely toarise in the first commercial-scale units. To hedgeagainst the possibility of poor operation delay-ing expansion, Congress may wish to support de-velopment programs intended to demonstrate thetechnical feasibility of a variety of processes andto gain basic knowledge of and experience with

10 ● Increased Automobile Fuel Efficiency and Synthetic Fuels: Alternatives for Reducing Oil Imports

these processes. Although these demonstrationprograms support second and third generationprocesses, they will also provide engineering in-formation that may be useful for correcting tech-nical faults and reliability problems that may arisein first generation plants.

Dealing With Other Effects

An important effect of increasing automobilefuel efficiency is the potential for decreased auto-motive safety due to size and weight reduction.There may also be major employment-relatedside effects associated with the restructuring ofthe auto industry and the accompanying acceler-ated rates of capital investment by the industry.There are familiar policy instruments that can dealwith both of these effects. For the safety effects,Congress can choose among safety standards fornew cars, educational programs, and support ofsafety R&D. Employment effects may be easedby minimizing plant relocations (through taxbreaks or direct assistance to the industry), or byameliorating the effects of employment reduc-tions through aid to communities and affectedworkers and other individuals.

potential environmental and worker-relatedproblems associated with synfuels developmentare substantial, and there is cause for concernabout the adequacy of future regulation of thesynfuels industry. The Government can help to

assure that the private sector takes account ofthese problems. Specific areas worthy of congres-sional attention include: the environmental re-search and regulatory programs of the Environ-mental Protection Agency (EPA), DOE, the Officeof Surface Mining, and the Occupational Safetyand Health Administration, in light of recentbudget cuts and changes in program direction;the dismantling of DOE’s demonstration programfor synfuels technologies; and the progress of SFCin demanding appropriate consideration of siting,monitoring, pollution controls and occupationalsafety as a condition for financial assistance. Con-gressional options range from holding oversighthearings to increasing the resources of the envi-ronmental regulatory agencies and shifting theirprogram emphases by legislation.

To mitigate the socioeconomic effects on com-munities from synfuels development, Congressmay wish to consider several forms of growthmanagement assistance, including loan guaran-tees, grants, and technical assistance. Any newFederal initiatives in this area will be complicated,however, by continuing arguments about relativeresponsibilities of Federal, State, and local govern-ments and private industry. And new initiativesneed to be sensitive to the substantial differencesfrom location to location in the severity of im-pacts and the resources already available for miti-gation.

OVERVIEW OF

Increased Automobile Fuel

Automobile fuel efficiency can

THE IMPORT REDUCTION OPTIONS

Efficiency Projections of Fuel Economy

be increased Future oil savings from increased automobile

through a variety of measures, including: fuel efficiency depend, first, on the magnitudeand character of future auto sales. In the past few

●

●

●

●

●

●

●

reductions in vehicle weight;improvements in conventional engines,transmissions, and lubricants;better control of engine operating param-eters;new engine and transmission designs;reduced aerodynamic drag;improvements in accessories; anddecreases in rolling resistance.

years, consumer preferences for such fuel-econ-omy-related characteristics as vehicle size andperformance have fluctuated while new car saleshave dropped significantly. Both the long-termsales average and consumer preference for fuelefficiency will be critical determinants of the rateof penetration of fuel efficiency technology.

Second, in response to changing consumerpreferences and foreign competition, the rate of

Ch. I—Executive Summary • 11

change of vehicle technology has accelerated andold rules about how long it takes to put a newtechnology into place* are no longer valid. Thepresent rapid rate of replacement of capital equip-ment puts a great strain on the domestic auto in-dustry. During the next several years, competitiveforces will push toward continued rapid techno-logical change, but the financial weakness of thedomestic auto industry will pull toward slowertechnological change. The strength of future for-eign competition and consumer perceptions ofthe future price and availability of gasoline anddiesel fuel, among other factors, will influencethe balance of these opposing forces, and, conse-quently, whether rapid increases in fuel efficiencyof domestically produced cars continue.

Third, the efficiency increases are not fully pre-dictable. There can be discrepancies between testresuIts and the results obtained in actual use.Technical compromises that affect ultimate per-formance have to be made to allow better inte-gration with existing equipment, easier andcheaper production and assembly, and resistanceto extreme operating conditions and incorrectmaintenance procedures. Development prob-lems are not always solved satisfactorily; suchproblems could occur more frequently if techno-logical change accelerates.

OTA developed projections (table 1) of plausi-ble ranges of average new-car fuel economybased on varying expectations of the relative de-mand for different-sized cars and the effectivenessand rate of development and introduction of newfuel-economy improvements. As reflected inthese projections, both technology and vehiclesize are critical factors for future fuel savings. Mar-ketplace uncertainty is reflected even as early asthe 1985 projections—manufacturers’ plans andthe technology are already established, but the

*For example, previous assumptions were: 5 years to move frominitial production decision to introduction of a technology in amodel line; 5 to 10 years to diffuse the technology throughout thenew car fleet; and 10 to 15 years of production to pay for the invest-ment. The estimate of 5 years to move from initial production deci-sion to introduction may now be a bit too low, while increasedrate of change of vehicle technology would necessarily reduce theother two estimates.

Table 1 .- Projected Average New-Car FuelEconomy,a 1985.2000 (mpg)

1985 1990 1995 2000

No further shift towards smallercars beyond 1985 . . . . . . . . . . 30-34 36-45 39-5443-62

Moderate further shift tosmaller cars . . . . . . . . . . . . . . . 30-34 38-48 43-5951-70

Rapid shift to small cars . . . . . . 33-37 43-53 49-6558-78aBased on EPA city/highway (55/45 percent) cycle. Each of the mileage ranges(e g., 3034) reflects relative expectations of the performance and rate of develop-ment and deployment of new technologies. The lower value represents OTA’S“low estimate” scenario, the upper value represents a “high estimate” scenario.SOURCE: Office of Technology Assessment

Note on Table 1: Can We Do Better?

The projections in table 1 do not represent the teclmo/ogica/ limitof what could be achieved in this century. The most efficient auto-mobiles in each size class are likely to achieve considerably betterfuel economy than the average; for example, technologies are avail-able that probably can allow a new-car fleet average of 60 mpg by themid-1990’s with the same mix of vehicle sizes as today’s and adequatevehicle performance (compared with the table’s 1995 “same size mix”projection of 39 to 54 mpg). This ignores consumer preferences forvehic/e features that cor)f/ict with fuel economy maximization, how-ever. By the same argument, the 1981 new-car fleet average COUM havebeen 33 mpg if consumers had consistently chosen the most efficientvehicle in each of the nine EPA size classes and producers had beenable to meet the demand. Instead, the actual 1981 model average toJanuary 1981 was 25 mpg.

Interestingly enough, if consumers had chosen only the most fuel-efficient gasoline-powered automobiles in each size class, over 90percent of the vehicles would have been U, S.-manufactured cars orcaptive imports. The market problems of U.S. manufacturers in 1981cannot be traced primarily to an inability of U.S. manufacturers to pro-duce fuel-efficient cars, but depend on factors such as differencesin perceived value between American and imported automobiles.

The difference between average fuel efficiency, which is a func-tion of consumer preference, and potential fuel efficiency, whichassumes that every car in the fleet embodies the most fuel-efficientchoice of technologies available, is critical to understanding whyOTA’S projections may differ from other projections that apply a sin-gle choice of technologies to the entire fleet. The latter assumptionis realistic only if future consumers value fuel economy, relative toother automobile attributes, much higher than they do today.

projections still range from 30 to 37 mpg* (com-pared to the 1981 level of 25 mpg).

How much fuel can be saved by improved fueleconomy? Assuming 30 mpg as a base and usingthe projections in table 1, continued develop-ments in automobile fuel efficiency could save0.6 to 1.3 MMB/D of oil by 2000. The lowervalue represents pessimistic expectations aboutthe advance of automobile technology and theshift towards smaller cars; the higher value repre-sents optimistic technological expectations andcontinued substantial shifts to small cars. Contin-

* Based on a weighted average of 55 percent EPA city test cycleand 45 percent EPA highway cycle, the formula used to measurecompliance with currently mandated CAFE requirements.


ued diffusion of these technologies into theoverall fleet could save 0.8 to 1.7 MMB/D by2010 with no further technological advances be-yond 2000. *

costs

OTA’s cost analysis of auto fuel-efficiency im-provements concentrates on investment costs intotal dollars as well as dollars per barrel per dayof oil saved. Estimates of the costs for associatedtechnology and product development are in-cluded in the investment costs,** because they

are part of the normal outlays needed to put anynew vehicle in production and represent a sizablefraction of the fixed costs (i.e., costs independentof production levels).

It is not possible to make highly accurate esti-mates of the investment costs (per barrel per dayof oil saved), due to the uncertainty associatedwith predicting actual efficiency increases thatwill be achieved. In addition, the cost of develop-ing technologies to the point where they can bereliably mass-produced has been highly variableand is difficult to predict.

Accurate cost estimation also is complicated bythe difficuIty of separating the cost of increasingfuel efficiency from the other costs of doing busi-ness. Increases in fuel efficiency are inextricablyintertwined with other changes in the car. For ex-ample, the engine redesign for fuel efficiency mayincorporate other changes, to improve otherautomobile attributes, at little additional cost. De-sign changes that increase efficiency may improveor degrade other attributes such as emissions orperformance.

If it is the industry’s judgment that consumersdo value fuel efficiency, the normal cycle of cap-ital turnover and vehicle improvement would re-sult in an increase in fuel efficiency automatical-ly. Unfortunately, the “normal” rate of fuel effi-ciency increase is not really predictable because

*Assuming 1.26 trillion vehicle miles (automobile only) traveledannually in 2000, 1.31 trillion in 2010, and an on-the-road fuel ef-ficiency 10 percent less than EPA rated fuel efficiency.

**In this context, development means all of the engineering activ-ities needed to prove a design concept and determine how it canbest be integrated into the vehicle system and mass-produced.

it depends on marketplace preferences and cor-porate strategies.

Because of the difficulty of separating out themarginal fuel efficiency investments from the“normal” investments, OTA’s investment costestimates (in dollars per barrel per day) in table2 are the total investments (including develop-ment costs) allocated to increasing fuel efficien-cy, divided by the total fuel savings rate expected.(See footnote c of table 2 for the details of thecost al location.) These investment rates may besomewhat lower than the marginal rates wouldbe because, in designing their “normal” invest-ment programs, manufacturers probably will se-lect those investments with the highest potentialpayoff in efficiency increase per dollars spent.

In any case, the range of investment rates forincreased fuel efficiency for each time periodoverlap the rates for investments in synfuels plants(see Synthetic Fuel section below), although the1985-90 fuel-efficiency rates would be lower thanthe synfuels rates if widespread expectations foroverruns in early synfuels investments are provedcorrect,

The total domestic capital investment associ-ated with increased fuel efficiency would beabout $25 billion to $70 billion between 1985 and2000, or less than $2 billion to $5 billion annu-ally during the period. This level of investmentcan be compared with recent and projected capi-tal investment by the industry* remembering thatpart of the fuel-efficiency investment could be in-cluded in “normal” capital expenditures if con-sumer demand for fuel efficiency is high enough.For the period 1968-77, annual capital investmentby General Motors (GM), Ford, and Chrysler aver-aged $6.68 billion in constant 1980 dollars. In-vestments by these companies rose to $10.4 bil-lion in 1979 and $10.8 billion in 1980, and areprojected by some analysts to rise to $12 billionper year during 1980-84. The ability of thedomestic industry to maintain their expectedschedule of capital expenditures is dependent on

*The two sets of figures are not fully analogous. A portion of thedomestic industry’s costs are for overseas investments, while a por-tion of the 1985-2000 fuel efficiency costs will be borne by outsidesuppliers rather than the major manufacturers.

Ch. I—Executive Summary ● 1 3

Table 2.—Domestica Investment for Increased Fuel Efficiency

Total investmentCar sales New-car fuel Efficiency investment (billion 1980$

Time (mill ion/yr) efficiency (mpg)b (thousand 1980 $/bbl/day) during time period)

1985-1990 . . . . . . . . . . . . . . . . . . . . . . . . . 8.6 38-48 20-60 8-291990-1995 . . . . . . . . . . . . . . . . . . . . . . . . . 8.8 43-59 60-140 9-201995-2000 . . . . . . . . . . . . . . . . . . . . . . . . . 9.1 51-70 50-150 7-18aASSUrneS 75 percent of all cars sold are manufactured domestically.bFleet average m~les per gallon at end of time period, with moderate shift in demand to W?’ Idler cars,CRepresents costs allocated t. fuel efficiency, including associated development costs of 40 percent of capital spending. lnVW3tm(3nt WaS allocated in the fOliOWing

way: 50 percent of the total engine investment is assumed to be for fuel efficiency; 75 percent of the total investment for conventional transmissions is for fuel effi-ciency; and all of the investment for advanced materials substitution, automatic engine on/off, and energy storage devices is for fuel efficiency.

dlncludes all capital costs associated with fuel efficiency, including fraction allocated tO other attributes, but excluding development costs,


a resumption of their former levels of sales andprofitability. There are already signs that U.S. automanufacturers are beginning to cut back onplanned investments in the face of continuedpoor sales and declining cash reserves.

If consumer demand for fuel efficiency is con-sistently strong, domestic manufacturers are likelyto respond by at least incorporating into their“normal” * rate of capital turnover as many fuel-efficiency features as possible. if capital turnoveris limited to its historical “normal” rates, then thefuel efficiencies shown in table 1 could still beachieved, but it would take longer to implementthe changes than is indicated by the schedulesshown in that table. In particular, implementa-tion of the low scenario could require 25 percentlonger (relative to 1985) than the schedule intable 1; and the high scenarios could require 45percent longer. Whether the high or the low sce-nario is eventually achieved, however, also de-pends on the success of technical developments.

If demand for fuel efficiency were high enough,however, the manufacturers would increase theirredesign/replacement rates. By adding $5 billionto $10 billion in capital expenditures during1985-2000, or $0.3 billion to $0.7 billion per year(5 to 10 percent above “normal”), capital turn-over can be speeded up to allow the low scenar-ios to be achieved on the schedule in table 1.Similarly, if technology developments are suc-cessful, the high scenario could be achieved asshown in table 1 with capital expenditures of $9

*Assuming “normal” capital turnover is: engines improved after6 years, on average, redesigned after 12 years; transmissions sameas engines; body redesigned every 7.5 years; no advanced-materialssubstitution.

billion to $23 billion above “normal” during theperiod 1985-2000, or $0.6 billion to $1.5 billionper year (10 to 20 percent above “normal”).

If future demand for fuel efficiency is not highenough to support these rates of change, in-creases in fuel efficiency will be further delayedunless required by new CAFE standards. On theother hand, CAFE standards without analogouslyhigh consumer demand for efficiency would re-quire the manufacturers to either defer expendi-tures for other improvements that might help carsales or to incur additional capital costs.

The consumer costs of increased fuel efficien-cy, measured in dollars per gallon of gasolinesaved, are speculative because the variable costs—mostly material and labor costs—are even moredifficult than investment costs to determine accu-rately. OTA’s analysis is based on alternative as-sumptions about the degree of change in materialand labor costs. A direct calculation of these costswould have been expensive and the results diffi-cult to defend because the source data is proprie-tary and highly dependent on judgments aboutthe success of adapting technologies to mass pro-duction. Table 3 shows the range of costs attrib-uted to fuel efficiency assuming that consumersvalue future gasoline savings as highly as today’ssavings (i.e., without discounting future savings*)and that manufacturers pass through the fullcosts. Conceivably, foreign competition couldforce the manufacturers to absorb part of thesecosts.

*The cost perceived by consumers would be about 2.5 times ashigh as those shown if the consumer discounts future fuel savingsat 25 percent per year, i.e., each future year’s savings during thelife of the car is valued at 25 percent less than the previous year’ssavings.

14 . Increased Autobile Fuel Efficiency and Synthetic Fuels: Alternatives for Reducing Oil Imports

Table 3.—Consumer Costs for Increased Automobile Fuel Efficiency, WithoutDiscounting Future Fuel Savings, Moderate Shift to Smaller Cars

Average new-car fuel Consumer costa ($/gal saved)efficiency at end of No variable cost High variable

Time period time period (mpg) increase costb increase1985-1990 . . . . . . . . . . . . . . . . . . . . 38-48 0.15-0.40 0.40-1.101990-1995 . . . . . . . . . . . . . . . . . . . . 43-59 0.35-0.85 1.10-2.601995-2000 . . . . . . . . . . . . . . . . . . . . 51-70 0.30-0.95 0.90-2.80aA~~Urn~~ a ~aPltal recovew factor per year of 0,15 times the capital investment allocated to fuel efficiencybAssumes variable cost increas~ IS twice the capital charges associated with the capital investments allocated tO fuel efficiency.


The consumer costs of fuel efficiency rangefrom values that are easily competitive with to-day’s gasoline prices to values that are consider-ably higher, depending on the efficiency gainsactually achieved, the success of developing pro-duction techniques that can hold down variablecost increases, and the value consumers placeon future fuel savings. Investments for increasedefficiency for the 1990-2000 model years will lookparticularly risky if the current soft petroleummarket continues for a few more years, or if automanufacturers have difficuIty holding down theirlabor and materials costs.

Another important measure of the cost to con-sumers of increased fuel efficiency is the increasein the price of new cars required to recover theindustry’s increased production costs. If themarket demand for fuel efficiency is strongenough to ensure that as much as possible of thecapital investment for fuel efficiency increases isincorporated into the normal capital turnover,and if the variable costs of production can be heldconstant, then the cost* of achieving OTA’s fuel-efficiency scenarios can be as low as $60 to $130per car during the 1985-2000 time period. Underthese conditions, an average of 35 to 45 mpgcould be achieved by 1990 without increasingnew-car costs.

If actual market demand for fuel efficiency isnot this high, automakers would be unlikely toincorporate a very high level of fuel efficiency in-vestments into their normal capital turnover. Also,the variable costs of production are likely to risesomewhat. The “upper bound” for added costs—assuming large increases in variable costs and

*Assumes a capital recovery factor of 0.15.

no market-driven investment for fuel efficiencyincreases beyond 1985—is $800 to $2,300 per carduring the 1985-2000 period, and $250 to $500per car to achieve 35 to 45 mpg by 1990. There-fore, the cost per car of increased fuel efficiencybeyond 1985 ranges from “clearly competitive”to “probably unacceptable. ”

Economic Impacts

The domestic automobile industry is in themidst of a massive investment program aimed atimproving the competitiveness of American auto-mobiles. These expenditures are associated withimportant structural changes in the industry; andaccelerating the rate of capital turnover (for in-creased fuel efficiency or other reasons) may ac-celerate some of these trends.

Manufacturers are closing older, inefficientplants and building new ones that incorporateextensive use of robots and other labor-savingtechnology to increase productivity. For a num-ber of reasons, including lower labor and othercosts, many of the new facilities may be built inthe Nation’s Sun Belt or overseas rather than inthe current North-Central auto manufacturingcenters, although recent labor concessions maychange this picture. Because of a shift in U.S. de-mand to smaller cars, which can be marketedmore universally, the incentive to produce in theUnited States is diminishing. Finally, because rap-id capital turnover is raising production costs ata time when consumer demand for automobileshas been sluggish, manufacturer profits havediminished and it has become harder for the firmsto secure capital at affordable costs.

American companies are forging more exten-sive ties with foreign manufacturers to design,

Ch. l—Executive Summary ● 1 5

produce, and market fuel-efficient cars, and aremoving towards producing more nearly standard-ized automobiles that can compete in internation-al markets. Current trends seem to be toward few-er separate automotive manufacturing and supplyfirms worldwide; only GM and Ford appear tobe reasonably certain of remaining predominant-ly American-owned.

Certain regions such as the industrial Midwest—and the Nation as a whole—will lose jobs ifthese structural changes continue. Job losses alsowould occur, however, if the process is inter-rupted, because the restructuring represents theindustry’s response to the conditions that causedits present market problems, and it clearly isaimed at regaining sales.

Social Impacts

As auto manufacturing and supply activities be-come more efficient and automated, there willbe important changes in the workplace environ-ment. Robots and other automated equipmentwill increasingly be used for the more routine anddangerous jobs, and skilled workers such as engi-neers and maintenance technicians should be-come a greater percentage of the smaller totalwork force. Shifting manufacturing overseas willreduce U.S. employment in primary manufactur-ing as well as in supplier companies. Althoughemployment losses may be larger in the supplierindustries, the effects in these industries will bedistributed over a larger geographical area. Em-ployment in related activities such as repair andservice will change to accommodate the newauto characteristics—e.g., repairs of plastic bodycomponents require adhesives, not welding—andthe increasing sophistication and capital invest-ment required for vehicle maintenance will placenew demands on shops and dealers.

Fuel efficiency increases also affect automobileowners by changing the physical attributes of thevehicle and the economics of owning cars. Forexample, a continued reduction in car size couldlead to increasing use of rentals for longer tripsor for occasional requirements for increased car-go-carrying capacity. Increases in the initial costof buying a car are likely to lead to a continua-tion of current trends of keeping cars longer,

resulting in a slower growth or reduction in new-car sales.

Environment, Health, and Safety

Increasing automobile fuel efficiency appearslikely to have a relatively benign effect on the nat-ural environment and public health, becausemost of the efficiency measures have few adverseeffects on auto emissions, emissions associatedwith vehicle manufacturing, etc. An important ex-ception may be any shift to widespread use ofdiesel engines, which could cause problems withvehicle particulate and nitrogen oxide (NOX)emissions. Also, the increased production of light-weight materials—particularly aluminum—maycause additional impacts, such as increasedenergy consumption in processing and increaseddemand for bauxite. On the other hand, signifi-cant downsizing of automobiles could allow ei-ther lower vehicle emissions or lower controlcosts to maintain current emission levels.

In contrast to their expected small effect on pol-lution levels, fuel conservation measures thatstress reducing vehicle size may have a signifi-cant adverse effect on vehicle safety. This is be-cause of the important role in crash survivalplayed by “crush space”* and other size- andweight-related factors. Even a relatively small de-cline in vehicle safety could cause hundreds oreven thousands of additional deaths and seriousinjuries per year.

There is no widely accepted estimate of themagnitude of this effect. The National HighwayTraffic Safety Administration has projected a10,000 per year increase in traffic deaths fromvehicle size reductions by 1990, but this is basedon a limited data set and a number of simplifyingassumptions. And a net increase in traffic deathsis not inevitable, since increased usage of passen-ger restraints and improvements in vehicle designcould more than offset the effect of moderate sizereductions.

*With a smaller “crush space” (thus, more rapid deceJerat;onof occupants in a crash), factors such as seatbelt and shoulder re-straint usage, better driver training and traffic control, and othersafety measures, become more important determinants of trafficsafety.


Synthetic Fuels

Production of a variety of fossil fuel-based syn-thetic fuels is planned or under development.

● Oil shale can be heated to release a liquidhydrocarbon material contained in the shale.After further upgrading a synthetic crude oilsimilar to high-quality natural crude oil can beproduced. This can be refined into gasoline,diesel and jet fuels, fuel oils, and other prod-ucts.

● Coal can be partially burned in the presenceof steam to produce a so-called “synthesis” gasof carbon monoxide and hydrogen, fromwhich gasoline, methanol, diesel and jet fuel,and other liquid fuel products (“indirect lique-faction”) or synthetic natural gas (SNG) canbe produced.

● Coal also can be reacted directly with hydro-gen (which is itself generated from a reactionof steam and coal) to produce a syntheticcrude oil (“direct liquefaction”). This oil canbe converted to gasoline, jet fuel and otherproducts in specially equipped refineries.

Projection of Synfuels Development

The principal technical deterrent to rapid de-ployment of a synfuels industry is the lack of prov-en commercial-scale synfuels processes in theUnited States. Shale oil, indirect coal liquefaction,and SNG processes currently are sufficiently de-veloped that the demonstration of commercial-scale process units or modules is being pursued,but these first units are likely to require consider-able modification before they can operate satis-factorily. Once these commercial-scale moduleshave been adequately demonstrated, full-sizecommercial facilities can be constructed (fromseveral modules). in contrast, direct coal liquefac-tion requires further development before com-mercialization and probably will not contributesignificantly to the synfuels industry before themid to late 1990’s. A major technical obstacle,the handling of high levels of solids in the proc-ess streams, is not now understood well enoughto allow developers to move directly to commer-cial- from small-scale units now in operation.

Normal planning, permitting, and constructionmay take 7 to 8 years for a large synfuels plant,with the last 5 years or so devoted to construc-tion. Consequently, a first round of commercial-scale plants conceivably could be operating bythe late 1980’s, although these would be quitevulnerable to delays and cost overruns. Beginninga second round of construction before the firstset of plants has been fully demonstrated wouldrisk additional costly revisions and delays.

In addition to scheduling constraints caused bytechnological readiness, shortages of experiencedmanpower (primarily chemical engineers andproject managers) could constrain the pace ofsynfuels development. On the other hand, prob-lems stemming from shortages of skilled crafts-men, construction materials, or specializedequipment probably can be averted because ofthe long Ieadtime before they are needed in largenumbers. However, some metals needed for cer-tain steel alloys are obtained almost exclusivelyfrom foreign sources.

Many variables affect the rate of development,and predictions are extremely speculative. It isOTA’s judgment that under favorable circum-stances, fossil fuel-based production of synthetictransportation fuels could be 0.3 to 0.7 MMB/Dby 1990, growing to 1 to 5 MMB/D by 2000,depending on the success of the first round ofsynfuels plants and the fraction of those plantsthat produce transportation fuels as opposed tofuel gases or fuel oils. Achievement of 0.3 MMB/Dby 1990 assumes that a sizable commercializa-tion program, such as that being pursued by theSynthetic Fuels Corp., is carried out, but thattechnical problems limit total production; 0.7MMB/D would require an increased number ofplant commitments within the next year or so,a virtually complete emphasis on liquid transpor-tation fuels, and a high level of technical successwith the first plants.

It must be stressed that even the “low” 0.3MMB/D production level maybe considered asoptimistic in light of current expectations of atleast short-term stability in oil prices, as well asremaining technical and environmental uncer-tainties. In addition, the dismantling of DOE’sdemonstration program may increase the per-ceived and actual technological risks of synfuels

Ch. l—Executive Summary ● 1 7

development. Thus, the goals of the NationalSynfuels Production Program, created by Con-gress in 1980—0.5 MMB/D by 1987 and 2MMB/D by 1992—appear unattainable withouta crash program that would involve extraordi-nary technical and economic risks and exten-sive Government intervention.

costs

The costs of synfuels are uncertain. First, thefactors that limit rapid deployment of the industryalso affect its costs. Technical uncertainties com-plicate cost evaluation, and long shakedowntimes and potential construction delays would bevery expensive at prevailing interest rates. Sec-ond, synfuels’ relatively high capital costs meanthat their total costs are especially sensitive to thetype of financing used, the level of interest rates,and the rate of return required by the investors.The present high level of uncertainty in capitalmarkets therefore translates into a high level ofcost uncertainty. In addition, the long construc-tion times associated with synfuels plants makethem vulnerable to hyperinflation. *

OTA has projected synthetic fuel costs basedon the best available cost estimates in the publicliterature and OTA’s previous oil shale study.1

These sources indicate that, if the potential forcost overruns is not considered, the capital invest-ment (in 1980 dollars) for a 50,000 bbl/d (ratedcapacity) synthetic fuels plant will range from $2.1billion to $3.3 billion, or $47,000 to $73,000/bbl/dof production (assuming the pIant produces at90 percent of rated daily capacity). Total plantinvestments for a 5 MMB/D synfuels industrywould thus be about $250 billion to $400 billion.Based on past experience, however, there is avery high probability that final costs will begreater than these ranges.

For example, an extrapolation from recent costoverruns in the chemical industry widens the sin-gle plant (50,000 bbl/d) range to $2.3 billion to

“Hyperinflation in construction costs of major capital projectsis a relatively recent phenomenon, however, and some industryanalysts consider it a temporary aberration.

1 U.S. Congress, Office of Technology Assessment, An Assessmentof Oi/ Sha/e Technologies, OTA-M-1 18 (Washington, D. C.: U.S.Government Printing Office, June 1980).

$4.7 billion (excluding direct liquefaction, forwhich cost estimates are less reliable), or about$50,000 to $110,000/bbl/d. Other related invest-ments (e.g., coal mining) raise the total to $50,000to $125,000/bbl/d. The investment costs per bar-rel per day of production may be further inflatedby performance levels below the 90-percent de-sign factor, although presumably this will be aproblem only with first generation plants.

The actual selling price of synthetic fuels willbe determined in the marketplace by the pricesof competing fuels regardless of the costs of pro-duction. Using the projected synfuels productioncosts, however, OTA calculated the price thatservice stations would have to charge in orderfor the synfuels producer to attain a required re-turn on investment. Table 4 displays these“prices” for a few alternative combinations of fi-nancing and real* return on investment.

Based on these estimates, it is clear that com-panies that must bear the full investment burdenof a new synfuels plant are unlikely to invest insynthetic fuels production unless: 1) they viewthis investment as one of low risk and worthyof a low expected return on investment, 2) theyexpect fuel prices to rise very sharply in the fu-ture, or 3) they are willing to take a loss or lowreturn to secure an early market share. The firstalternative is not credible for the first genera-tion of commercial plants.

With the large (75 percent of project costs)loan guarantees that are possible under the En-

*The real rate of return is the nominal rate of return minus theinflation rate.

Table 4.—Price of Synthetic Fuels Required ToSustain Production Costsa (1980 $/gal of

gasoline equivalent)

Real returnPrice (pretax) Financing equity investment

$0.80-$1 .10... 100% equity 50/0$1.30-$1.60, . . 1000/0 equity 10%0$1.70-$2 .40... 100% equity 15 ”/0$0.80-$ 1.10... 25°/0 equity, 75°/0 debt 10 ”/0aA~~umPtiOns: no cost Overruns, $1 .20/million Btu coal, 5 percent real interest

rate on debt financing, $.20/gallon distribution cost including retailer profit. Formore details, see ch. 8, table


18 . Increased Automobile Fuel Efficiency and Synthetic Fuels: Alternatives for Reducing Oil Imports

ergy Security Act, * however, investments in syn-thetic fuels appear to be attractive even at 1981fuel prices, but only if current capital cost esti-mates for synfuels plants are correct and thereare no cost overruns. Most industry experts,however, consider the chances of substantialcost overruns to be high. A cost overrun in plantinvestment of so percent would increase the nec-essary price of synfuels by 20 to 30 percent, orelse reduce the return on investment.

OTA’s cost analysis implies that significant lev-els of investment in synfuels production are un-likely at this time without the kinds of financialincentives offered by SFC. Of course, a further—and currently unexpected—rapid escalation in oilprices could change this conclusion.

Uncertainties associated with the cost estimatesare too large to allow in-depth comparison of thecosts of the various synfuels. Also, in OTA’s opin-ion, significant reduction of these uncertaintiescannot be achieved by further study but will re-quire actual plant construction. There are indica-tions, however, that shale oil and methanolfrom coal could be the least expensive optionsfor producing transportation fuels. Shale oilplants are less complex technically than othersynfuels processes and shale oil is relatively easyto refine, thereby suggesting a lower cost. Metha-nol’s high octane and burning characteristicsmake it more efficient than gasoline in speciallydesigned engines. But materials-handling prob-lems for oil shale, engine technology develop-ments that could offset methanol’s efficiency ad-vantage, and unforeseen requirements for proc-ess changes could negate these apparent advan-tages.

Economic Impacts

Development of a fossil fuel-based syntheticfuels industry could create a major new economicactivity in the United States, particularly in areaswith large reserves of coal or oil shale. There arepotential drawbacks, though. For example, be-

*The loan guarantees not only allow synfuels developers to bor-row money at somewhat lower interest rates than without them,but the 75 percent debt level is considerably higher than the in-dustry average of about 30 percent. Also, in some cases, the loanguarantee may be necessary to secure any debt capital at all.

cause of synfuel plants’ long lifetimes and con-struction Ieadtimes, a liquid fuel supply industrybased largely on synfuels would be less able thana natural petroleum-based industry to respondquickly to changing market conditions. Rapidsynfuels deployment would create a risk that un-foreseen market changes could leave the UnitedStates with an outdated, idle, capital-intensiveindustry.

Development of the industry will have otherimportant consequences. For instance, becauseof the large capital, technical and marketing re-quirements, and the high risks, small companiesare unlikely to enter the market except as partsof consortia. This contrasts sharply with the largenumber of small-scale producers currently in-volved in oil and gas development, althoughownership concentration in the oil and gas indus-try will grow in any case as the more easily recov-ered resources are depleted.

Rapid deployment of a synfuels industry couldlead to temporary shortages of equipment, mate-rials, and personnel, which in turn can lead toconstruction bottlenecks and local inflation.However, long-term inflationary effects are notexpected to be large because, in general, theIeadtime is sufficient to expand production capac-ity and labor supply. An important exception maybe the supply of experienced chemical engineersand project managers. If shortages of these per-sonnel develop, poor project management or im-proper plant design could lengthen constructionschedules, delay plant startup, and increase costsfor chemical plants and oil refineries as well assynfuels plants.

The financial requirements for rapid growth arevery large. For example, the rate of investmentrequired to achieve 5 MMB/D of synfuels by 2000is likely to be greater than $30 billion per year*after the first few years, about as much capitalas was spent for all U.S. oil and gas explorationand development in 1979. Making this large acommitment to synfuels would likely divert someinvestment capital away from conventional oiland gas exploration and development; and this

*OTA’S analysis of reducing stationary uses of fuel oil was doneprimarily to provide a reference point and was less extensive thanits analysis of synfuels and increased automotive fuel efficiency.

Ch. 1—Executive Summary ● 1 9

could reduce conventional domestic oil produc-tion below the 7 MMB/D assumed for 2000.

Social Impacts

The principal social consequences of develop-ing a synthetic fuels industry stem from shiftinglarge numbers of workers and their families in andout of local areas as development proceeds.These population shifts disproportionately affectsmall, rural communities such as those that pre-dominate in the oil shale and some of the coalareas. High population growth rates can lead todisruptions and breakdowns in social institutions;systems for planning, managing, and financingpublic services; local business activities; and la-bor, capital, and housing markets. Whether thegrowth rates can be accommodated depends onboth community factors (e.g., size, location, taxbase, management skills, and availability of devel-opable land), and technology-related factors(e.g., the type of synfuels facilities, the timing ofdevelopment, and labor requirements).

On the other hand, communities should realizesocial benefits from synfuels development, e.g.,increased wages and profits and an expandingtax base. A significant portion of these benefitsmay not be realized, however, until after the plantis built. In the meantime, the community mustmake significant expenditures and the overall im-pact depends substantially on the existence of ef-fective mechanisms to provide the “front end”resources needed to cope with rapid growth.


The production of large quantities (2 MMB/Dor more) of liquid synthetic fuels carries a signif-icant risk of adverse environmental and occupa-tional health effects, some of which are quite de-pendent on the effectiveness of as yet unprovencontrol measures.

The industry will cause many of the same kindsof mining, air quality, solid waste disposal, wateruse, and population effects as are now associatedwith coal-fired electric power generation andother forms of conventional coal combustion.Table 5 shows the amount of new coal-fired pow-er generation that would produce the same ef-

fects as a 50,000-bbl/d coal-based synthetic fuelsplant, and also directly compares the effects ofthis plant with a 3,000-MWe coal-fired power-plant. In general, the emissions of combustion-related pollutants, especially the acid rain pre-cursors sulfur dioxide (SO2) and NOX, and thewater use of the synfuels plant are significantlylower than for a powerplant processing thesame amount of coal.

To place these effects into perspective, actualcoal-fired generating capacity in the United Statesis about 220,000 megawatts (MW), and about200,000 MW are expected to be added by 1995.In comparison, SO2 and NOX emissions from a2 MMB/D coal-based synfuels industry would beequivalent to emissions from less than 25,000MW of power generation, and water use wouldbe equivalent to that of 30,000 MW or less if con-servation practices were followed.

On the other hand, a 2-MMB/D industry(equivalent in coal consumption to 110,000 to160,000 MW of coal-fired electric generatingcapacity) would mine hundreds of millions oftons of coal each year, with attendant impactson acid drainage, reclamation, subsidence andoccupational health and safety, and would havesubstantial population-related impacts such assevere recreational and hunting pressures onfragile Western ecosystems.

Oil shale development using aboveground re-torts has the added problem of disposing of largequantities of spent shale. Although successfulshort-term stabilization of shale piles has beenachieved on a small scale, uncertainty remainsabout the long-term effects of full-scale develop-ment. The major concern about shale disposalas well as in-situ shale processing is the poten-tial for contamination of ground waters.

Despite the relatively moderate level of emis-sions per unit of production, an intense concen-tration of synfuels development within relative-ly small areas may yield air quality problems andviolations of existing air quality regulations. Suchconcentration is more likely with oil shale, be-cause of the concentrated resource base. As aresult, air quality restrictions may limit oil shaledevelopment to under 1 MMB/D unless there arechanges in the restrictions or improvements in

20 c Increased Automobile Fuel Efficiency and Synthetic Fuels: Alternatives for Reducing Oil Imports

Table 5.—Two Comparisons of the Environmental Impacts of Coal-BasedSynfuels Production and Coal-Fired Electric Generationa

A. Coal-fired generatingcapacity that wouldproduce the same B. Side-by-side comparison of

impact as a 50,000 bbl/d environmental impact parameters

coal-based synfuels 3,000 MW(e) 50,000 bbl/dType of impact plant (MW(e)) generator synfuels UnitsAnnual coal use . . . . . . . . . . . . . . 2,500 -3,600b 6.4-15.0 5.3-17.9 million tons/yrAnnual solid waste . . . . . . . . . . . (2,500-3,600)±C 0.9-2.0+ 0.6-1 .8+ million tons/yrAnnual water use: acre-ft/yr

Current industry estimate. . . . 640-1,300 25,000 5,400-10,800Conservation case . . . . . . . . . . 400-700 3,400-5,900

Annual emissions: tons/yrParticulate. . . . . . . . . . . . . . . . 120-2,800 2,700 100-2,500Sulfur oxides . . . . . . . . . . . . . . 90-500 27,000-108,000 1,600-9,900Nitrogen oxides . . . . . . . . . . . . 60-300 63,000 1,600-7,800

Hourly emissions: Ib/hrParticulate. . . . . . . . . . . . . . . . 90-2,200 880 30-800Sulfur oxides . . . . . . . . . . . . . . 70-40 8,800-35,200 500-3,200Nitrogen oxides . . . . . . . . . . . . 60-300 20,500 500-2,500

Peak labor . . . . . . . . . . . . . . . . . . . 4,100-8,000 2,550 3,500-6,800 personsOperating labor . . . . . . . . . . . . . . 2,500 440 360 personsa in example A the powerplant uses the same coal as the synfuels plant, new source performance standards (NSPS) aPPIY, SUlfUr oxide emissions assumed to be

0.6 lb/lOC Btu~ In B, NSPS aiso appiy but sulfur oxide emissions can range from 0.3-1.2 lb/104 Btu. In both cases, the synfuels piant parameters represent a rangeof technologies, with a capacity factor of Xl percent and an efficiency range of 45 to 65 percent; the powerplant is a baseload plant, with a capacity factor of 70percent, efficiency of 35 percent. The major data source for this table was M. A, Chartock, et al., Erw/rorrrnertta/ Issues of Syrrtfret/c Triwrsportatlorr Fue/s from Coa/:8ac/rgrourrd Report University of Okiahoma Science and Public Policy Program, contractor report to OTA, July 1961.

b In other words, the amount of coal–and thus, the amount of mining–needed to fuel a 50,000 bblld synfuels plant iS the same as that required for a 2,500-3,600 MVV(e)powerplant.

c A synfuels plant will have about as rllu’ti asfl to dispose of as a coal-fired powerplant using the same amount of COal. It may have 18SS scrubber sludge, but it maYhave to dispose of spent catalyst material that has no analog in the power plant thus the +.


control technology. In most cases, however, therestrictions do not involve possible violations ofthe health standards, but rather visibility or otherstandards.

Aside from these effects, synfuels developmentcreates a potential for occupational, ecological,and public health damage from the escape oftoxic substances formed during the conversionprocesses. These include cancer-causing organiccompounds, chemically reduced sulfur and nitro-gen compounds, and inorganic trace elements.The occupational risks, generally acknowledgedas the most serious, are mainly associated with“fugitive” emissions and leaks from valves, gas-kets, etc., and with the handling of fuels and plantcleaning. The major ecological and public healthrisks are associated with contamination of surfaceand ground waters—from inadequate treatmentof wastewaters, leakage from holding ponds orsolid-waste landfills, and disruption of aquifers bymining operations—as well as with spills and ex-posure to contaminated fuels. Fugitive emissionsand leaks from the plants also pose some risk to

public health, but at a far lower level than to theplant workers; a potentially important impact ofthe public’s exposure to these substances, how-ever, is likely to be discomfort from their odor.

The risks associated with these toxic sub-stances, although possibly the most serious of syn-fuels’ potential environmental risks, are not quan-tifiable at this time. However, it does appear pos-sible to differentiate, at least tentatively, amongsome of the basic process groupings in terms oftheir comparative risk. Direct processes (e.g.,Exxon Donor-Solvent, SRC ll) appear to presentthe greatest risk because of their comparativelylarge number of potential sites for fugitive emis-sions, high production of toxic hydrocarbons, andabrasive process streams. Indirect processes usinglow-temperature gasifiers (such as Lurgi) maybeintermediate in risk because they produce rela-tively large quantities of toxic hydrocarbons. In-direct processes using high-temperature gasifiers(e.g., Koppers-Totzek, Shell, Texaco) appear tobe the cleanest group of coal-based processes.Finally, if the risks from spent shale are excluded,

Ch. I—Executive Summary ● 21

the risks from toxics yielded by oil shale processesprobably are no worse than those of indirect,high-temperature coal processes.

There is little doubt that it is technically feasi-ble to moderate a synfuels industry’s adverse im-pacts to the satisfaction of most parties-of-interest.in OTA’S judgment, however, despite substan-tial industry efforts to minimize adverse im-pacts, the environmental risks associated withtoxic substances generated by synthetic fuelsproduction are significant and warrant carefulGovernment attention.

Current development plans and existing legisla-tion call for strong measures to reduce many ofthe potential adverse environmental impacts fromsynfuels plants through intensive application ofemission controls, water treatment devices, pro-tective clothing for workers, monitoring for fugi-tive emissions, and other measures. Virtually allof these measures have been adapted from con-trols used with some success in the petroleumrefining, petrochemical, coal-tar processing, andelectric power-generation industries. Synfuels in-dustry spokesmen are confident that the plannedcontrols will adequately protect worker and pub-lic health and safety as well as the environment.

Spokesmen for labor and environmental orga-nizations are far less confident, however, andthere remain important areas of doubt concern-ing the adequacy of environmental management.The full range of synfuels impacts–especiallythose associated with the toxic substances createdor released in the conversion processes—may notbe effectively regulated. Existing regulations donot cover many of these toxic substances, andextending regulatory controls to provide full cov-erage will be difficult. Critical stumbling blocksare the large number of separate compounds thatmust be controlled, and the recent reductions inthe budgets of Federal environmental agenciesand reemphasis of their synfuels research pro-grams. Detecting synfuels environmental dam-ages and tracing them to their sources—a key re-quirement in establishing and enforcing controlstandards—may be difficult because many of thedamages will occur slowly and the relationshipbetween cause and effect is complex.

Another important concern is the possibilitythat the industry’s environmental control effortsmay not be sufficient to avoid environmental sur-prises. Federal Government personnel are con-cerned that many developers are focusing theircontrol programs on meeting immediate regula-tory requirements and are reluctant to commitresources to studying and controlling currentlyunregulated pollutants. Also, despite pollutioncontrol engineers’ optimism that all synfuelswaste streams are amenable to adequate cleanup,there are still doubts about the reliability of pro-posed control systems. These doubts are aggra-vated by differences in process conditions andwaste streams between synfuels plants and therefineries, coke ovens, and other facilities fromwhich the proposed controls have been bor-rowed, and also by a lack of testing experiencewith integrated control systems.

Water Availability for Synfuels Development

When aggregated nationally, water require-ments for synfuels development are small (pro-ducing 2 MMB/D oil equivalent requires onlyabout 0.2 percent of estimated total current na-tional freshwater consumption). Nevertheless,these requirements may have significant impactson competing water uses. In each of the riverbasins where major coal and oil shale resourcesare located, there are hydrologic as well aspolitical, institutional, and legal constraints anduncertainties involving water use (e.g., conflictsover the use of Federal storage, Federal reservedwater rights including Indian water rights claims,interstate and international compacts and treaties,State water laws). In addition, existing water re-source studies vary in the extent they considerwater availability factors and cumulative impacts.

Given the uncertainties that surround the ques-tion of water availability generally, only limitedconclusions about possible constraints on futuresynfuels development can be drawn. This is espe-cially true in areas where institutional rather thanmarket mechanisms play a dominant role in ob-taining and transferring water rights. Where effi-cient markets do exist, however, water is not like-ly to constrain synfuels development because de-

22 “ Increased Automobile Fuel Efficiency and Synthetic Fuels: Alternatives for Reducing Oil Imports

velopers can afford to pay a relatively high pricefor water rights.

In the major Eastern river basins where coalreserves are located (the Ohio, Tennessee, andUpper Mississippi River Basins), water should beadequate on the main rivers and large tributar-ies, without new storage, to support planned syn-fuels development. In the absence of appropriatewater planning and management, however, lo-calized water shortages could arise during ab-normally dry periods or from development onsmaller tributaries.

In the West, competition for water already ex-ists and is expected to intensify with or withoutsynfuels development. In the coal-rich UpperMissouri River Basin, the magnitude of the legaland political uncertainties, together with the needfor major new water storage projects to average-out seasonal and yearly streamflow variations,make it impossible to reach an unqualified con-clusion as to the availability of water for syn-fuels development.

In the Upper Colorado River Basin, whereboth oil shale and coal are located, water couldbe made available to support initial synfuels de-velopment—as much as a few hundred thou-sand barrels per day of synfuels production by1990—but political and legal uncertainties in thebasin make it difficult to determine which sourceswould be used and the actual amount of waterthat would be made available. Water availabil-ity after 1990 will depend both on how these un-certainties are resolved and on the expected con-tinuing growth in other uses of water.

Reducing Stationary Uses of Oil

Stationary uses of oil include space heating andcooling of buildings, electricity generation, pro-duction of industrial process heat, and use as achemical feedstock. These currently account fornearly half of the oil used in the United States—about 8.1 MMB/D out of a total of 16.8 MMB/Din 1980. Of this, about 4.4 MMB/D are fuel oils—middle distillates and residual oil. The remainderinclude liquefied petroleum gas, asphalt, petro-leum coke, refinery-still gas, and petrochemicalfeedstocks.

Only reductions in the fuel oil portion of thestationary oil uses are likely to lead to actual re-ductions in imports. The other oil products aredifficult to upgrade to premium fuels or use di-rectly in transportation applications and, conse-quently, a reduction in their use probably wouldhave little effect on the supply of transportationfuels. On the other hand, the crude oil fractionsnormally used to produce residual and distillatefuel oils can instead be converted profitably intotransportation fuels by refining.

Reductions in fuel oil use can be accomplishedby fuel switching and conservation. In the build-ings sector, natural gas and electricity can replacedistillate oil, and insulation, furnace improve-ments, and other conservation measures can re-duce fuel use in general. For utilities, conserva-tion in all sectors that use electricity can reducegeneration requirements, coal and nuclear canreplace residual oil for baseload operation, andnatural gas can replace distillate oil in peakingturbines. industrial oil use can be reduced by in-creases in process efficiency and fuel switchingto coal, * natural gas, and electricity.

Projection of Oil Savings

The Energy Information Administration projectsthat the fuel oil consumed in stationary uses willdecline from today’s 4.4 MM B/D to 2.6 MM B/Din 1990, assuming a 1990 price of $41/bbl of oil(1979 dollars). This 2.6 MMB/D is the target forfurther stationary use reduction OTA has assumedfor this study.**

OTA has evaluated two approaches to eliminat-ing the remaining 2.6 MM B/D stationary fuel oiluse by 2000. One approach involves total reli-ance on fuel switching. Table 6 shows the energyneeded to displace the 2.6 MMB/D, substitutingcoal for residual oil and natural gas and/or elec-tricity for distillate oil.

*The Energy Information Administration has predicted that, by1990, most of the industrial processes that can use coal (primarilylarge boilers) will have been converted. Therefore, OTA’S calcula-tions of post-1990 fuel-switching opportunities do not include coalswitching in the industrial sector.

* “If oil prices continue to decline in real dollars or stabilize atcurrent levels, 1990 stationary oil use is likely to be greater thanprojected.

Ch. l—Executive Summary 23

Table 6.—Summary of Annual Energy Requirement To Displace2.55 MMB/D of Stationary Fuel Oil Use

Oil Coal or Electricityreplaced by (106 tons)Sector (M MB/D) plus (tfc) (10’ kWh)

Buildings . . . . . . . . . . 1.2 — 2.4 425Industry . . . . . . . . . . . 0.3 — 0.6 120Utilities. . . . . . . . . . . . 0.9 (resid.) 100 — —

0.15 (dist.) . . — 0.3 —

Total . . . . . . . . . . . . 2.55 100 3.3 545SOURCE: Office of Technology Assessment.

The technical capability to accomplish this levelof switching depends on two factors. First, if natu-ral gas is to play a major fuel-switching role, pro-duction of unconventional gas sources* will beneeded. A key to this is the future price of gas.Second, production of additional electricity andswitching to coal in utility boilers depends pri-marily on the utility industry’s ability to solve itscurrent financial problems and gain access to cap-ital. Either of these potential constraints couldseverely restrict fuel switching.

A second approach combines fuel switchingwith measures to conserve oil, natural gas, andelectricity. If conservation measures can saveenough natural gas and electricity to replace the(reduced) oil requirement, additional gas andelectricity production may not be needed. Ananalysis by the Solar Energy Research Institute(SERI) indicates that conservation measures in thebuildings sector alone could save about 1.5 timesas much natural gas and electricity as would berequired to replace all remaining stationary usesof distillate and residual fuel oil by 2000.** Thiscombined approach has fewer technical con-straints than the “fuel switching only” approach.In OTA’S judgment, a significant fraction of the2.6 MMB/D of stationary fuel oil use expectedin 1990 can be eliminated by 2000 by conser-vation and fuel switching taken together.

Despite the lack of absolute constraints, how-ever, it is unrealistic to expect total eliminationor near-elimination of stationary fuel oil uses by2000. First, average capital costs are high enoughto discourage those investors who apply a high

*These sources include tight sands formations, geopressurizedmethane, coal seam methane, and Devonian shale formations.

* *ASSuming that all such stationary uses are reduced by conser-vation as well.

discount rate to their investments. Second, thesite-specific variability and the large number ofdifferent types of measures imply that some ofthe individual measures will be far more expen-sive than the average. * Third, the record of oil-to-coal switching in industry during the past dec-ade has not been a good one despite apparentlyfavorable economic incentives. Finally, the con-tinued reduction in supplies of high-quality“light” crude may lead to excess supplies ofresidual oil in the 1990’s, driving down its priceand making conversion from residual oil to coaluneconomical in some cases. To a certain extentthe latter effect will be offset by the economicattractiveness of retrofitting oil refineries to pro-duce less low-priced residual oil and more gaso-line, jet fuel, and diesel fuel.

costs

The investment costs for the two strategies forreducing stationary oil uses are similar. OTA hascalculated the investment cost of the strategy thatrelies mainly on fuel switching to be roughly $230billion, or an average of $90,000/bbl/d of oilsaved. Using SERI’S cost analysis, the strategy thatcombines strong conservation measures and re-placement of the remaining oil use with electricityand natural gas was calculated to cost roughly$225 billion, or $88,000/bbl/d of oil saved. Thedifference between the estimated costs for thetwo strategies is too small to be meaningful.

Both the investment costs and the operatingcosts paid by individual investors will vary overan extremely wide range, This is particularly truebecause the “strategies” are actually a combina-

*By the same reasoning, many will be less expensive than theaverage.

98-281 0 - 82 - 3 : QL 3

——

24 ● /ncreased Automobile Fue/ Efficiency and Synthetic Fuels: Alternatives for Reducing oil Imports

tion of several markedly different kinds of invest-ments. For example, conversion of oil-burningutility boilers to coal has an average investmentcost of $74,000/bbl/d of oil replaced, whereasconversion of distillate-using facilities to naturalgas averages $114,()()()/bbl/d (including the costof obtaining the new gas). Also, the costs of eachtype of investment will vary from site to site.

Electric Vehicles*

Automobiles can be powered by rechargeablebatteries that drive an electric motor; indeed,some of the first automobiles used battery-electricpowertrains. Present concepts of electric passen-ger vehicles generally envision small vehicles forcommuting or other limited mileage uses, withrecharging at night when electricity demand islow.

Projections of Use

OTA does not expect electric cars to play a sig-nificant role in passenger transportation in thiscentury. Battery-electric cars are likely to be veryexpensive, costing about $3,000 more in 1990than comparable gasoline-fueled autos. And thisconsumer investment may not yield any savingsin fuel costs. If batteries must be replaced every10,000 miles, as required with current technol-ogy, total electricity plus battery costs will actu-ally be considerably higher than gasoline costsfor a comparable conventional auto, even at$2.00/gal gasoline prices (1980 dollars).

Another reason that OTA is not optimistic is thatprogress in electric vehicles remains severely lim-ited by battery performance. Currently availablebatteries and components require 6 to 12 hoursfor recharging and limit electric vehicles to arange of less than 100 miles between charges. Ac-celeration is limited to about O to 30 mph in 10seconds, which is lower than the poorest per-forming (O to 40 mph in 10 seconds) gasoline anddiesel fuel cars and may not be adequate formany traffic conditions. Although predictions ofsignificant reductions in battery size and weight

2See, Synthetic Fuels for Transportation: The Future Potential ofElectric and Hybrid Vehicles–Background Paper No. 1,OTA-BP-E-13 (Washington, D. C.: U.S. Congress Office OfTechnology Assessment, April 1982).

continue, in OTA’S judgment current understand-ing of battery performance does not permit accu-rate predictions of future improvements.

Even if sufficient progress is made in battery de-velopment to encourage extensive usage of elec-tric cars, electrification of automobile travel doesnot offer the same potential for oil savings asthe other options. Under the best of circum-stances, most electric passenger vehicles are like-ly to be small, limited-performance vehicles thatwill substitute for small, fuel-efficient conven-tional autos. Consequently, a 20-percent electri-fication of the auto fleet is not likely to savemore than about 0.2 MMB/D. *


If technical developments, severe liquid fuelshortages, and/or Government promotion wereto result in significant sales of electric vehicles,the major environmental impacts probably wouldbe the air quality and other effects associated withreducing auto emissions and increasing electricitygeneration. The overall effects of widespread useof electric vehicles on urban air quality shouldbe strongly positive, because the electric carswould tend to be clustered in urban areas, manyof which have chronic automobile-related airpollution problems that would be eased by thedisplacement of conventional automobiles.There would be, however, a small net increasein regional and national emissions of SO2, be-cause conventional autos have few or no SO2emissions to offset the SO2 emissions from fossil-fueled electric power generation for battery re-charging. Additionally, when coal is the fuelsource for recharge electricity, the amount of coalmined per unit of oil replaced is comparable tothat for synfuels production,** with similar coalmining impacts. Material requirements for bat-teries could add substantially to the demands forcertain minerals, e.g., lead, graphite, and lithium.

Electric passenger vehicles are likely to besmall and thus should share safety problems

*Assuming no oil is used for electricity production and the aver-age gasoline- or diesel-powered vehicle replaced gets 60 mpg.

* *That is, 1 ton of coal yields the same oil savings in either tech-nology.

Ch. I—Executive Summary ● 2 5—

with small conventional automobiles. Additional dent-caused spill; this latter problem is balancedsafety and health problems may be caused by the somewhat by eliminating the fuel tank with itsbatteries, which contain toxic chemicals that may highly flammable contents. Finally, extensive out-pose occupational problems in manufacturing door charging of vehicle batteries may pose pub-and recycling and may be hazardous in an acci- Iic safety problems from the electrocution hazard.

FIGURE

Figure No. Page

3. U.S. Petroleum Consumption, Domestic Production, and Imports . . . . . . . . 29

Chapter 2

Introduction

Petroleum has been at the heart of the Nation’sefforts to secure its energy future since the 1973-74 oil embargo. Petroleum products currentlysupply about 40 percent of total U.S. energyneeds, and imports account for approximately 30percent of total petroleum demand. The increasesin the price of imported crude that have occurredsince 1972—a barrel of imported crude that wasselling for $4.80 in 1972 was selling for $31.40(both expressed in constant 1980 dollars) as ofFebruary 1982–have severely strained the na-tional economy by contributing to domestic infla-tion and trade deficits and by altering demandpatterns.

Instead of increasing gradually, petroleumprices have gone up in two large, rapid jumps.These price “shocks” have exacerbated strainson the economy by preventing an orderly adjust-ment to higher fuel prices. Future price behavioris also very uncertain and complicates economicplanning. Subsequent to decontrol of domesticcrude oil prices, a continued increase in the priceof oil was generally expected. The sharp drop indemand, coupled with reemergence of Iran onthe world oil market, however, has createddownward pressure on oil prices. A substantialdrop in the real price of oil over the next fewyears is quite possible.

This oil glut will not last indefinitely, however.Indeed, importing nations will eventually againface increasing competition for oil arising froma combination of dwindling world supplies* andincreases in demand from both producing anddeveloping nations. This, in turn, will force oilprice increases with the possibility that they mayagain come in the form of price shocks ratherthan an orderly rise. Thus, the U.S. petroleumproblem is not simply that we import oil–indeed,in a stable, economically rational world import-

*OTA estimates that non-Communist world oil production couldrange from 45 to 60 MMB/D in 1990 and 40 to 60 MMB/D in 2000,compared to 46 MM B/D in 1980. While an increase in world pro-duction is possible, it is more likely that production will remainfairly stable or slightly decline. See World Petroleum Availability19802000-A Technical Memorandum (Washington, D. C.: U.S.Congress, Office of Technology Assessment, October 1980),OTA-TM-E-5.

ing could be economically efficient-but that thesupply and price is subject to so much uncertaintyand dramatic change.

For 1981, the Nation’s imports of crude oil andpetroleum products averaged about 5.4 millionbarrels per day (MMB/D), accounting for about34 percent of total petroleum demand. These fig-ures compare with 7.5 MMB/D of imports, about41 percent of total petroleum demand, for 1980.In fact, imports and demand have been on asteady downward trend since their peak in 1978,when imports of 8.2 MMB/D represented 44 per-cent of total demand (see fig. 3). The principalcause of this downward trend in imports and theirshare of total petroleum demand was the nearly120-percent increase in the real price of crudeoil since 1978. If the trend were to continue, oilimports could be eliminated by the end of thecentury.

There is considerable disagreement, however,on whether this trend can be maintained. Thecurrent downward trend of prices as a result ofa soft market has already been noted. The princi-pal focus of disagreement, however, is the ade-quacy of the Nation’s future domestic supply. Thecurrent domestic production of crude oil and nat-ural gas liquids is 10.2 MMB/D. Production ofthese two liquids has been declining steadily since

Figure 3.—U.S. Petroleum Consumption,Domestic Production, and Imports

c 1 8>1 6

Consumption

Production

1973 1974 1975 1976 1977 1978 1979 1980 1981

Year

SOURCE: Energy Information Administration, U.S. Department of Energy.

29

30 . Increased Automobilr Fuel Efficiency and Synthetic Fuels; Alternatives for Reducing Oil Imports

Photo credit: U.S. Department of Energy

Oil tankers deliver most of the crude oil importedto the United States

the peak year of 1970, when it reached 11.3MMB/D, although the decline has slowed notice-ably the last 3 years (see fig. 3). This, coupled withthe remarkable upsurge in drilling activity since1979, has caused some forecasters to predict anincrease in production before the end of the cen-tury. In its most recent forecast, under a highworld oil price scenario, for example, the EnergyInformation Administration of the Department ofEnergy forecasts a production rate of 11.2MMB/D of crude oil and natural gas liquids by1995.1

Despite this increase in drilling and explorationactivity, as figure 3 shows, the effect of higher oilprices to this point has been almost exclusivelyin reducing demand, not increasing supply. Inaddition, the Office of Technology Assessmentin a study, World Petroleum Availability:79802000, did not find any evidence that a sig-nificant upturn in domestic production wouldoccur.2 In fact, OTA estimated a production rateof 5 to 8 MMB/D by 1990 and 4 to 7 MMB/D by2000. Exxon’s most recent energy outlook fore-cast domestic production of 7.1 MMB/D by 1990and 7.8 MMB/D by 2000.3 Thus, production rate

1 Annual Report to Congress, 1980, vol. 3, DOE/EIA-01 73(80)/3,Energy Information Administration, U.S. Department of Energy,Washington, D. C., March 1981, p. 85.

2 World Petroleum Availability: 1980-2000-A TechnicalMemorandum (Washington, D. C.: U.S. Congress, Office ofTechnology Assessment, October 1980), OTA-TM-E-5.

3Energy Outlook: 1980-2000, Exxon Co., U. S. A., Houston, Tex.,December 1980, p. 7.

estimates for the middle 1990’s differ by as muchas 7 MMB/D, an amount greater than current U.S.imports.

In OTA’S judgment, the lower rates estimatedin its study are still valid and it would be impru-dent to assume otherwise in planning for the1990’s. The principal justification for these lowerrate estimates is not so much an actual declinein total oil reserves as the increasing difficulty inextracting the oil that is found. The rate at whichoil can be produced is declining because oil isno longer being found in the very large oilfieldsnecessary to sustain or increase total production.

If domestic production does decline to 5 or 8MMB/D by the mid-1990’s, demand will have todecrease even faster than it has during the lastfew years just to keep imports at their current lev-el. It is possible to greatly reduce petroleum prod-uct demand by both fuel switching and increasedefficiency of use. This will require a substantialinvestment, however, and it is not clear yetwhether it will be made. The current demandresponse is a combination of shortrun elasticityto the most recent price rise and the longer runelasticity—involving turnover of capital stock—to the 1973-74 price rise. Current forecasts ofenergy demand all show a continued, but slower,decline in petroleum demand for the rest of the1980’s but a steadying during the 1990’s. In allcases, substantial imports will be necessary in the1990’s if the decline in domestic production as-sumed above takes place.

Faced with similar prospects after the 1973-74oil embargo, Congress enacted a wide range oflegislation to encourage a reduction of the Na-tion’s dependence on oil imports. First, legisla-tion was enacted to reduce petroleum demand.Foremost among these initiatives was the estab-lishment in the Energy Policy and ConservationAct of the 1985 Corporate Average Fuel Efficien-cy (CAFE) standards for automobiles. More thanhalf of total U.S. petroleum demand is in thetransportation sector, which, in turn, uses abouthalf of its petroleum in passenger vehicles. Otherlegislation to reduce petroleum demand in-cluded: 1 ) the Powerplant and Industrial Fuel UseAct, whose provisions require large combustorsto convert from oil by 199o; and 2) systems oftax credits and financial programs to encourage

Ch. 2—/ntroduction . 31

capital investments for energy conservation in in-dustry and buildings. Finally, legislation waspassed to augment domestic supplies of petro-leum substitutes through the production of syn-thetic fuels. The Energy Security Act of 1980 es-tablished the Synthetic Fuel Corp., with synfuelproduction goals of 0.5 MMB/D in 1987 and 2.0MMB/D by 1992.

The efficacy of this approach is now being re-evaluated, not only because of the time limita-tions of the legislation, but also in light of the priceincreases of 1979, which have caused the Nationto reduce imports more quickly than originallyanticipated. The debate centers on whether theGovernment should continue its approach of the1975-80 period—and if so which path or pathswould be more effective-or whether the Nationcan depend primarily on the current high oil cost—not entirely anticipated when the original legis-lation was passed–to reach an acceptably lowlevel of oil imports. Complicating current debateis uncertainty about where oil prices will go inthe immediate future. A steady decline in realprices over the next few years could substantial-ly reduce market pressure toward increased con-servation and development of alternative fuels,with the result that the Nation will be that muchmore economically vulnerable should there bedramatic upsurges in prices later in the decadeor throughout the 1990’s. Further, no matterwhich approach is preferable, general concernexists about the side effects of the Nation’s move-ment to free itself from import dependence.

Of particular interest in the ongoing evaluationof policy are the approaches directed at produc-ing synthetic liquid fuels and at reducing petro-leum consumption by automobiles. These are byfar the largest programs enacted by Congress,both in terms of their costs and of their effectson the economy.

In 1979, the Senate Committee on Commerce,Science, and Transportation requested OTA toassess and compare these two approaches to re-ducing oil imports. Because mandated increasesin CAFE standards were to end in 1985, the com-mittee was interested in whether further stand-ards would be more or less effective than the syn-fuels program in reducing oil imports. The large

increase in oil prices since 1979 and the responseto this increase have changed the environmentin which that request was made. As a result OTAhas broadened its study to consider how far andfast automobile efficiency and synthetic fuels pro-duction can develop during the next 20 years,and to evaluate the effects on and risks to the in-dustries involved. Such an evaluation is particu-larly important for the automobile industry be-cause of its current precarious state. No matterwhat course the Nation chooses—continued reg-ulation or more reliance on market mechanisms—the industry may be in for continued difficultiesif the need for large capital investments continueswhile demand for automobiles remains relative-ly low.

In addition to assessing the import reducing po-tential of synfuels and increases in automobilefuel efficiency, this study also examines, althoughin considerably less detail, the oil savings thatcould be gained through fuel switching and con-servation by stationary oil users. By combiningthe three approaches, plausible development sce-narios for reductions in oil consumption are de-rived, leading to estimates for oil imports over thenext 20 years.

The body of this report starts with an analysisof policy options that: 1 ) could influence the rateat which oil imports can be reduced or 2) affectthe consequences of changes needed to reduceconventional oil consumption. The next chapterpresents the major issues and findings of the re-port, including a discussion of the cost of oil im-ports, comparisons between increased automo-bile fuel efficiency and synfuels, and analyses ofissues related solely to one or the other of theseoptions.

The remaining chapters of the report containthe background analyses. Separate chapters onincreased automobile fuel efficiency, synfuels,and stationary uses of petroleum present the tech-nical and cost analyses for each. The final chap-ters analyze the economic and social effects andenvironmental, health, and safety impacts associ-

ated with both increased automobile fuel efficien-cy and synfuels, as well as the availability of waterfor synfuels development.

Chapter 3

Policy

Contents

Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

The Nation’s Ability To Displace Oil Imports. . . . . . . . . . . . . . . . . . . .

Rationale for a Direct Federal Role. . . . . . . . . . . . . . . . . . . . . . . . . . . .

Distinguishing Features of Increasing Automobile Fuel Efficiency andSynthetic Fuels Production . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Factors Affecting the Rate of Automobile Fuel-Efficiency IncreasesAutomobile Fuel Efficiency-Policy Background. . . . . . . . . . . . . . . .Factors Affecting the Rate of Synthetic Fuels Production . . . . . . . . .Synthetic Fuels production–Poiicy Background . . . . . . . . . . . . . . . .

Policy Options. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .EconomyWide Taxation–Oil and Transportation Fuels . . . . . . . . . .Economywide Taxation—Special Provisions . . . . . . . . . . . . . . . . . . .Research and Development . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Trade Protection . . . . . . . . . . . . . . . . . + . . ? . . . . . . . . . . . . . . . . . . . .Sector-Specific Demand Stimuli-Purchase Pricing Mechanisms . . .Sector-Specific Supply Stimuli-Subsidies and Guarantees.. . . . . . .Regulations on Output . . . . . . . . . . . . . .Other Effects . . . . . . . . . . . . . . . . . . . . . .

Conclusions . . . . . . . . . . . . . . . . . . . . . . . . .

Appendix 3A.–Policy Options To Reduce

Appendix3B. –Additional CRS References

. . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . .

Stationary Oil Use . . . . . .

. . . . . . . . . . . . . . . . . . . . . .

TABLES

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

Page

35

35

37

3840414242

444648484950525556

60

62

63

Table No. Page

7. Distinguishing Features oflncreasing Automobile Fuel Efficiencyand Synfuels Production . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39

8. Potential Average New-Car Fuel Efficiencyin 1995 . . . . . . . . . . . . . . . . . . . . . 56

Chapter 3

Policy

INTRODUCTIONSuccess of any efforts to reduce oil imports will

depend on many complex, unpredictable factors,including world oil prices, the success of tech-nological developments, consumer behavior,general economic conditions, and–significantly–Government policies and programs.

Government policy is vitally important, be-cause energy inevitably affects, whether direct-ly or indirectly, all production and consumptiondecisions in an industrial society. How quicklyand to what level the Nation displaces oil importshave direct implications for who benefits from,and who pays the costs of, energy independence.Such distributional questions arise regardless ofpolicy choices. Thus, the policy choices made byCongress transcend a simple choice between in-tervention and nonintervention.

detailed discussion of policy options related tofuel switching and conservation in stationary usesof petroleum, and to biomass, the reader is re-ferred to other OTA publications.1) The chapteraddresses the circumstances which might justifydirect Government intervention to displace oilimports. The well-established auto industry andthe newly developing synfuels industry are thendescribed; and those economywide and sector-specific characteristics which shape, direct, andpace each industry’s ability to displace oil importsare identified. A brief, recent history of Govern-ment policy towards each industry is also pro-vided. Finally, the major policy options availableto Congress are discussed and evaluated basedon the characteristics of the industries.

This chapter describes the policy issues and op-tions for increasing automobile fuel efficiency andaccelerating synthetic fuels development. (For a

‘Energy From Biological Processes, OTA-E- 124, July 1980; Resi-dential Energy Conservation, OTA-E-92, July 1979; Dispersed Elec-tric Energy Generation Systems, OTA, forthcoming; Energy Efficiencyof Buildings in Cities, OTA-E-168, March 1982.

THE NATION’S ABILITY TO DISPLACE OIL IMPORTS

The three principal means for displacing oilimports-increased automobile fuel efficiency,synfuels production, and fuel switching and con-servation in stationary uses—can all make impor-tant contributions to the Nation’s energy future.Legislation has recently been enacted in all threeareas to reduce conventional oil use, includingthe Energy Policy and Conservation Act of 1975,the Energy Security Act of 1980, the Fuel Use Act,and various taxes and credits to encourage capitalinvestment for energy conservation in industriesand buildings. z Some progress in displacing im-ports can be expected as a result of these Govern-ment programs working in concert with marketforces.

OTA’S technical analysis, presented in this re-port, concludes that if Congress wishes to elimi-

2For details of recent legislation, see for example congressionalQuarterly, Inc., Energy Po/icy, 2d cd., March 1981.

nate net oil imports, significant accomplishmentin all three areas may in fact be necessary toachieve this goal by 2000 if domestic productionfalls from 10 million to 7 million barrels per day(MMB/D) or less by 2000, as OTA expects.3 Ingeneral, if there are no additional policies andprograms, if technology developments are onlypartially successful, and if strong market forcesfor import displacement do not materialize, theUnited States can expect to import 4 to 5 MMB/Dor more by 2000 (see issue on “How Quickly CanOil Imports Be Reduced?” in ch. 4).

In the near future, Congress will face a numberof decisions about whether to increase efforts todisplace oil imports, and if so, at what speed im-ports should be displaced, Major decisions will

3 World Petroleum Availability: 1980-2000—A TechnicalMemorandum, OTA-TM-E-5 (Washington, D. C.: U.S. Congress,Office of Technology Assessment, October 1980).

35

36 “ Increased Automobile Fuel Efficiency and Synthetic Fuels: Alternatives for Reducing Oil Imports

concern the two major programs enacted byCongress: the setting of fuel efficiency (CorporateAverage Fuel Economy (CAFE)) standards beyondthe 1985 mandate for new cars sold in the UnitedStates, and the provision of large subsidies to pro-mote rapid development of the synfuels industry.Such decisions will shape the future of both theauto and synfuels industries. Whether new policyinitiatives are feasible, practical, and appropriatecannot be determined until Congress specifies thedesired level and rate of import displacement. IS

the goal to “eliminate” or to “reduce” imports?*IS this oil import displacement goal so vital to na-tional interests that an emergency effort is re-quired, regardless of any accompanying disrup-tions and dislocations?

Given the uncertainties, risks, and unpredicta-bility associated with both the automobile fuel-efficiency and synfuels options, it is difficult todetermine how far and in what direction presentpolicies and market forces will take the Nation.OTA has not attempted to predict the detailedoutcomes of alternative policy futures, but ratherto demonstrate that the ability to displace oil de-pends on complex, interrelated factors, and todemonstrate that the Government’s policychoices—whether to implement additional poli-cies or to “do nothing” —will make a differencein the ability to achieve oil displacement goals.Policies are also identified that could be effec-tive if future Government action is necessary.**

OTA’S low estimate is that the average fleet fuelefficiency for new cars could reach at least 40 toso miles per gallon (mpg) by the early to mid-1990’s and 45 to 60 mpg by 2000,*** based on

*“Elimination of oil imports” herein is assumed to mean the re-duction of net oil imports to a level of about O to 1 MMB/D by 2000.At this level, the “security premium” for oil—the difference betweenthe market price and full economic cost to the Nation of oil im-ports-would approach zero. This level of supplies could be pro-cured primarily from the United Kingdom, Canada, and Mexico;and foreign producers generally would be forced to compete formarkets. Because of the “security premium, ” policy decisions aboutthe value of displacing imports should not be based solely on theinternational price of oil.

**An examination of Government policy related to the use ofpetroleum in the stationary sector is beyond the scope of this report.A summary of the major policy options for the stationary sector,however, is found in app. 3A to this chapter.

***Earlier trends showing relatively strong demand for fuel econ-omy have encouraged some domestic manufacturers to predictnew-car fuel economy averages of over 30 mpg by 1985 (the 1985CAFE standard requires a fleet average of 27,5 mpg), while individualvehicles already on the market exceed 45 mpg.

relatively pessimistic expectations about howquickly improved automotive technology is de-ployed and purchased. Fleet fuel consumptionfor passenger cars would be about 2.1 MMB/Din 2000, for a cumulative savings of over 1 billionbarrels of oil between 1985 and 2000 (assumingthat the same proportion of large, medium, andsmall cars are sold in 2000 as are expected to besold in 1985). The “high estimate” assumes thattechnology development is both successful andrapidly introduced into volume production. Aver-age mpg ratings would be 55 to 65 mpg by 1995and 60 to 80 mpg by 2000; and fleet fuel con-sumption for passenger cars could be as little as1.3 MMB/D in 2000 for a cumulative savings ofover 4 billion barrels (relative to a 30-mpg fleetand assuming a rapid shift to small cars).

However, the actual level of fuel consumptionwill depend on market demand for fuel-efficientcars and/or additional Government policies de-signed to facilitate either the manufacture or pur-chase of these cars. Although the low estimatesare believed to be achievable in the absence ofadditional Government policies, they would becontingent on consumer expectations that thereal price of gasoline will continue to increase.The high estimates are unlikely to be achievedin the absence of supporting Government policiesunless a strong and continuing consumer demandfor fuel efficiency is coupled with favorable tech-nological progress.

OTA’S estimates for a low- and a high-develop-ment scenario for synthetic fuels production de-pend principally on the price of conventional oiland the ease and rate with which synfuels proc-esses are proven. A rapid buildup of the industrycould begin as early as the late 1980’s or as lateas the mid-l990’s, resulting in technically plausi-ble production levels of fossil-synthetic transpor-tation fuels of 0,3 to 0.7 MM B/D by 1990, 0.7 to1.9 MMB/D by 1995, and 1 to 5 MMB/D by2000. * In the absence of additional Governmentpolicies, the lower estimates are probably attain-able but are contingent on a Government-sup-ported commercialization program that reducesthe high technical and associated financial risksto private investors of first generation plants.

Without a successful commercialization pro-gram, even the low estimates are probably unat-

*These estimates exclude contributions from biomass.

Ch. 3—Pol icy ● 3 7

tainable. If there is a commercialization program,synfuels production theoretically could reach thehigh estimates without additional Governmentprograms if: 1 ) early commercial-scale demonstra-tion units are built and work successfully, and2) synfuels production becomes unambiguouslyprofitable. The maximum displacement of oil im-ports, however, would occur only if synfuels pro-duction concentrates on transportation fuels. itis OTA’S judgment that, even with a commercial-ization program, the high estimates are likely tobe delayed by as much as a decade unless poli-cies tantamount to energy “war mobilization” areenacted.

Fuel switching and conservation of oil in sta-tionary uses will also be extremely important fordisplacing oil imports and would complementboth fuel-efficiency and synfuels efforts. Althoughmuch of the potential for displacement couldprobably be achieved by market forces by 2000(under the high oil price scenario of the EnergyInformation Administration (EIA)),4 additional pol-icies to encourage fuel switching and conserva-tion will likely be required to accelerate the

4Department of Energy, Energy Information Adminstration, An-nual Report to Congress, vol. II 1, May 1982.

changes or completely eliminate stationary fueloil use. The level of displacement that can be ob-tained depends not only on future oil prices, buton financing, regulation, and technical factors.Efficiency increases in the various nonautomobiletransportation uses could also be significant.

Displacing oil imports is a necessary but nota sufficient condition for achieving nationalenergy security. Such security translates into anessential self-reliance, availability, affordability,and sustainability of energy resources. Alternativeenergy sources may present their own set of sup-ply and/or distribution problems. Furthermore,the relationship between the level of imports andthe level of insecurity is not proportional in anobvious way. Even if the Nation could eliminateall of its oil imports, U.S. energy security couldstill be seriously affected if interruptions in worldoil supplies threatened international commit-ments with allies, imbalances in the world mone-tary system, and pressures on foreign exchangemarkets. Thus, efforts to displace the Nation’smost insecure oil resources—its imports-shouldnot divert attention away from ensuring the resil-ience of the alternatives chosen and thus the sta-bility of both domestic and international energysystems.

RATIONALE FOR A DIRECT FEDERAL ROLEThe basic rationale for direct Federal involve-

ment in a market economy is that—in limited butimportant areas— market prices and costs usedto evaluate returns on private investments do notreflect the fuII value and cost of the investmentsto society as a whole. National security and envi-ronmental protection are classic examples of val-ues and costs that are not reflected in profit andloss statements. Private calculation of profits alsocauses market mechanisms to be most respon-sive to short-term economic forces as opposedto long-term social and economic goals.

The three principal reasons for such market“failures” are that: 1 ) some of the social benefitsare public and not private goods, 2) some of thecosts are not paid by the private sector, and3) costs and benefits are not fully known. All three

situations arise in the context of displacing oil im-ports in general and of both increasing automo-bile fuel efficiency and producing synfuels in par-ticular. The inability of the conventional market-place to ensure the effective and rapid displace-ment of oil imports has major implications for theFederal role, depending on the goals chosen andthe resources made available.

National security is a public good that has tradi-tionally received Government support. Nationalenergy security, promoted by the displacementof oil imports, is an important component of over-all security. Direct Government involvementwouId thus be justified if market forces alonewere not believed capable of achieving the quan-tity and rate of oil displacement required by na-tional security goals. The value to the Nation of


accelerating automobile fuel efficiency increasesand synfuels production would be in addition toany private returns to investment.

Both increased automotive fuel efficiency andsynfuels production give rise to side effects andtradeoffs; those who benefit from the investmentsare not necessarily the ones who bear the fullcosts. Side effects can fall on different sectors ofthe economy, regions, or consumer groups de-pending on the investments chosen. In the caseof increased automobile fuel efficiency, the ra-tionale for Government policy is that the activitiesstimulated by market forces alone do not provide,for example, adequate safety and employmentsafeguards. There are other possible tradeoffs,on the one hand, between improving the com-petitive position of the U.S. auto industry byencouraging investments in increased auto fuelefficiency, and, on the other hand, possible de-clines in auto-related employment levels (becauseof increased automation, contraction of the do-mestic industry), increased consumer costs, and

decreased safety. With respect to synfuels, Gov-ernment intervention could be similarly war-ranted if market decisions do not reflect environ-mental, health, safety, and other social concerns.

Both increased automobile fuel efficiency andsynfuels production are characterized by finan-cial risks and uncertainties. If market forces alonedetermine outputs, investments associated withthese alternatives might be delayed or canceled.In such cases, the Government could chooseeither to assume some of the risk or to help re-duce components of uncertainty. The auto indus-try’s uncertainty focuses on unpredictable con-sumer demand for fuel-efficient cars, the longIeadtimes for investments, and, to a lesser degree,on the rate of technological development. Syn-fuels production is subject to significant techno-logical uncertainties and, in turn, financial risks.Both the auto and synfuels industries are also af-fected by uncertain and as yet undetermined fu-ture Government policies.

DISTINGUISHING FEATURES OF INCREASING AUTOMOBILE FUELEFFICIENCY AND SYNTHETIC FUELS PRODUCTION

The major forces that will shape, direct, andpace increases in automobile fuel efficiency andsynfuels production are summarized in table 7.Identifying these forces may indicate both thepotential opportunities for and limitations ofGovernment policies in achieving a desired leveland rate of import displacement, and the appro-priateness, practicality, and desirability of specificpolicies or combinations of policies.

Although increased automobile fuel efficiencyand synfuels production share several attributes,essential differences between them suggest thatthere is no single role for Government policiesand programs. These two options should beviewed as complementary measures for reduc-ing oil imports. Each option has different implica-tions for the rate of oil import displacement andwill give rise to different types of economic andnoneconomic impacts on the Nation. In addition,within the uncertainty about investment costs(per barrel per day oil equivalent (B/DOE) pro-

duced or saved), neither increased automobilefuel efficiency nor synfuels production appearsto have an overall unambiguous enconomic ad-vantage over the other. For this reason, the non-monetary and often nonquantifiable differencesbetween these options will be the principalmeans for distinguishing between them for pol-icymaking purposes.

The factors that determine the rate of fuelswitching and conservation in stationary applica-tions will share some common elements withautomobile fuel efficiency increases and synfuelsproduction. The success of fuel switching will de-pend critically on the efficiency of stationaryenergy uses, the technologies for producingnatural gas from unconventional sources, the sup-ply and future price of conventional natural gas,and the ability of the utility industry to solve itscurrent financial problems. in the absence ofmandated conservation or performance stand-ards, conservation measures will depend primar-

Ch. 3—Poicy Ž 39

Table 7.—Distinguishing Features of Increasing Automobile Fuel Efficiency and Synfuels Production

Increasing automobile fuel efficiency Synfuels production1. Both near- and long-term restructuring of an existing

industry2. Dominated by a few large, mature companies

3. Automobiles as consumer durables; differentiable;deferrable

4. New technology involved, but can proceed incremen-tally; associated risks are an ongoing feature ofindustry

5. Industry must produce competitive products eachyear, including fuel-efficient cars

6. Precariousness of industry’s current financial posi-tion; need to ease readjustment of an industry indistress

7. Large demand uncertainty

8. Dispersion of industrial activities, domestically and,increasingly, internationally; some concentration ofactivities in the North-Central region of the UnitedStates

9. Capital intensity and associated risks10. Declining profit margins in domestic industry11. Significance of international competition (i.e., auto im-

ports); importance of domestic market to financialviability

12. Large amounts of capital continually required forredesign, retooling, etc.; final costs for improved fuelefficiency uncertain; calculation of capital costs forfuel economy dependent on methods for costallocation

13. Can make significant contributions to reducing U.S.oil imports; contributions have a long Ieadtime butcan have significance incrementally

14. Caters to a saturated market; focus on productreplacement rather than growth markets

15. Consumer costs are investment to reduce future fuelpurchases

16. Reduces consumption of fuel17. Fuel savings in automobiles limited to about 3.5

MMB/D with about 1.5 MMB/D savings coming fromachieving a 30-mpg fleet

18. Principal health impact may be increased auto deathsdue to smaller cars

1. Growth and promotion of a new industry

2. Likely to be dominated by a few large, maturecompanies

3. Synfuels as uniform, consumable commodities

4. Large technical risks; possibilities for “whiteelephants, ” major risk occurs with first commercial-scale demonstration plants

5.

6.

7.

8.

9.10.11.

12.

13.

14.

15.

16.17.

18.

Sponsoring industry is involved in a breadth of activi-ties that provides alternative investment and businessopportunities, of which synfuels is oneSoundness of sponsoring industry’s current financialposition; need to facilitate growth

No unusual demand risk except insofar as synfuelsdiffer from conventional fuelsDispersion of activities among coal regions; currentoil shale activity concentrated in a small area of theWest

Capital intensity and associated risksPotential for profit still highly uncertainLong-term export potential; importance of interna-tional competition (i.e., oil imports) in terms ofestablishing the marginal priceLarge amounts of capital required primarily in the ini-tial construction phase; final costs for synfuels pro-duction uncertain

Can make significant contributions to reducing U.S.oil imports; contributions have a long Ieadtime andwill not be significant until commercializationCaters to a slowly growing or possibly decliningmarketNo investment needed by consumer; consumer paysincrementally for each increment of consumptionSubstitutes one fuel for anotherFuel-replacement potential ultimately limited by de-mand for synfuels, environmental impacts of synfuelsplants, and coal and oil shale reservesEnvironmental and health impacts from: large-scalemining of coal and oil shale; possible escape of toxicsubstances from synfuels reactors (major risks aredirect worker exposures, contamination of groundwater); visibility degradation; development pressureson fragile, arid ecosystems


ily on investor behavior. Although there are few investments. Both fuel switching and conserva-technical constraints, there are technical uncer- tion are difficult to put into practice because thetainties about the conservation potential of build- measures are varied, site-specific, and in someings and the success of particular conservation cases costly.

98-2 B 1 3 - 82 - 4 : QL, 3

40 ● /ncreased Automobile Fue/ Efficiency and Synthetic Fuels: Alternatives for Reducing oil Imports

Factors Affecting the Rate ofAutomobile Fuel= Efficiency Increases

In order to be internationally competitive, thedomestic automobile industry is currently under-going major structural adjustments. 5 This readjust-ment is the consequence of two interrelatedforces. First, the domestic industry is undergoinga long-term restructuring that is being experi-enced by auto manufacturers worldwide. Re-source pressures and a trend towards small, fuel-efficient, and standardized “world cars” have re-sulted in a period of corporate consolidation, withfirms being more closely tied by joint design and/or production ventures, and a geographic disper-sion of product assembly. Secondly, U.S. automanufacturers are uniquely faced with a seriesof short-term problems that arise because theyhave historically served a market that demandedlarge, relatively fuel-inefficient cars. U.S. manu-facturers have been the principal producers (andpromoters) of large cars and have historicallyearned their greatest profit margins on these cars.

The strains placed on the domestic industry,as it redesigns its products and retools its facilitiesfor fuel efficiency in the near and midterm,6 arethe forces that could most appropriately be tar-geted and eased by Government policies. In addi-tion, because of the size and dispersion of theU.S. auto industry throughout the national econ-omy, maintaining the health of the industry andminimizing the side effects on both upstream ac-tivities (e.g., dealers, suppliers), and downstreamactivities (e.g., consumers) are of potentially greatGovernment concern.

Some aspects of the domestic industry’s short-term readjustment problems are caused by econ-omywide factors such as rising energy prices, tightcredit, and high interest rates. These factors haveaffected both manufacturers—by making capitalscarce and expensive—and consumers, who(with approximately two-thirds of all purchaseshistorically being on credit) are deferring pur-chases.

The market changes associated with high gaso-line prices and the threat of gasoline shortagesexperienced in the 1970’s have shown that con-sumer demand is the most powerful influence onthe rate and manner of fuel-efficiency increases.However, the prices consumers will pay, and thetradeoffs consumers will accept in vehicle attrib-utes—of which fuel efficiency is only one—arehighly uncertain and ambiguous. For example,in the mid-1 970’s, and again in 1980-81, the pro-portion of relatively small cars purchased to largecars purchased decreased. * Furthermore, con-sumers did not consistently buy the most fuel-efficient car in a given size class. The ability ofthe industry to sell cars is made additionally diffi-cult because there has been a steady slowing inthe total demand for automobiles due to stagnantper capita disposable income and a general agingof the population, implying that the industry ismainly serving a domestic replacement ratherthan growth market. And at the same time, im-ports have captured an increasing share of thedomestic market.

The need to make large investments under con-ditions of uncertain demand for fuel efficiencyand slowing overall demand for automobiles, ag-gravated by economywide stresses, is the mostsignificant contributor to the financially precari-ous position now facing the domestic auto indus-try. Losses to U.S. auto manufacturers exceeded$4 billion in 1980. As sales have decreased, prof-its have declined, and the industry’s longstandingability to reinvest with internally generated fundshas decreased. Because the industry is capital-in-tensive, any underutilization of capacity also im-plies large costs. Large amounts of outside capi-tal will be required to retool for increasing fueleconomy. If companies are forced to cut backon their capital investment programs in the nearterm (as some are doing), they will not only fore-stall fuel-efficiency improvements but may alsobecome increasingly vulnerable to foreign com-petition. 7

In adjusting to this, U.S. manufacturers face aseries of complex decisions. Domestic manufac-

s~ee ~lso us. /nduStrja/ Competitiveness-A Comparison ofstee(Electronics, and Automobiks, OTA-ISC-135 (Washington, D. C.: U.S.Congress, Office of Technology Assessment, July 1981).

K3ee General Accounting Office, “Producing More Fuel-EfficientAutomobiles: A Costly Proposition,” CED-82-14, Jan. 19, 1982.

*These data include both domestically produced and importedcars.

7U. S. Industrial Competitiveness, op. cit.

Ch. 3—Policy ● 4 1

turers’ weak competitive position is primarily dueto high production costs relative to foreign pro-ducers and consumers’ perceptions of value indomestic v. imported cars, that are difficult toquantify. If a strong demand for fuel efficiencydevelops, rapidly increasing the fuel efficiency ofdomestic automobiles could help to sell them.Demand for fuel efficiency, however, is usuallyaccompanied by a shift in demand to smallercars, the market area where U.S. manufacturershave been least competitive with imports. Thisshift would therefore also require that domesticmanufacturers devote some of their investmentsto changes unrelated or only partially related tofuel efficiency. Corporate strategy, the way com-plex investment decisions are handled, overalldemand for new cars, and the demand for fuelefficiency vis-a-vis other attributes of automobileswill all interact in a complex way to affect the ac-tual rate at which fuel efficiency increases.

Technological uncertainties will also figure indetermining fuel efficiencies actually achieved.These uncertainties relate to the behavior of vari-ous elements of the vehicle system, the way inwhich these elements are integrated and possi-ble performance tradeoffs among elements, andthe cost of specific manufacturing techniques.The rate of product and process development,and particularly the success of the developmentefforts (by no means assured) will influence theextent of fuel-efficiency increases. Basic researchcould lead to additional fuel economy gains byproviding a better understanding of some of thecomplex processes related to fuel consumption(e.g., nonsteady-state combustion).

The single most important factor limiting thedevelopment of electric vehicles (EVS) is batterytechnology. Even if EVS were to become practi-cal, however, they would not have the potentialto displace significant amounts of imported oil,primarily because they would be substitutes forthe most fuel-efficient gasoline or diesel-poweredcars. The Government could justify acceleratingthe development and introduction of EVS if the

goal is to reduce automobile pollution in theinner cities or to promote a transportation modethat does not use petroleum. EVS are petroleumindependent except insofar as electricity is gen-erated from oil.

Automobile Fuel Efficiency—Policy Background

The industry has been regulated by Govern-ment policies and programs primarily since the1960’s. Worker and public health and safety as-pects are regulated by the Environmental Protec-tion Agency (EPA) and the Occupational Safetyand Health Administration; product safety andemissions by the National Highway Traffic Safe-ty Administration (N HTSA) and EPA. Auto salesare affected by all policies that influence consum-er demand. In the aftermath of the 1973-74 oilembargo, the Government became actively inter-ested in promoting automobile fuel efficiency,and legislation was subsequently enacted to re-duce U.S. dependence on oil imports.

Policy for increasing automobile fuel efficien-cy is embodied principally in two programs. * Theprincipal policy instrument for increasing fuel effi-ciency was established by the 1975 Energy Policyand Conservation Act (EPCA) and specifies CAFE(i.e., fleet) standards for new cars and light trucksbetween the model years 1978 and 1985. Provi-sions of the CAFE program have generally triedto recognize the financial difficulties of the autoindustry. CAFE standards mandate that new-carfuel efficiency will double, incrementally, be-tween the early 1970’s and the mid-1980’s.(American-made cars had averaged about 14 mpgover the period 1965-75; the 1985 CAFE stand-ards require fleet averages of 27.5 mpg and areto remain in force after 1985.**) Subsequent pro-visions in the Fuel Efficiency Act of 1980 easedthe compliance requirements of the CAFE pro-gram, but the basic efficiency standards remainin force. Possible alteration of the standards setby the program for post-1 985 could be a majorpolicy issue coming before Congress.

a The sources ancJ magnitude of these cost advantages are not wellunderstood. Understanding the nature of any cost advantages en-joyed by competitors will be critical for determining how, when,and if U.S. manufacturers can get their cost structure into line, SeeU.S. Industrial Competitiveness, op. cit., pp. 96-99.

*A third program was enactment of the 55-mph speed limit.**The Energy Policy and Conservation Act provided guidelines

for the Department of Transportation to set standards for the early1980’s; Congress set standards for 1978 and 1985.

42 . /ncreased Automobjile Fuel Efficiency and Synthetic Fuels: Alternatives for Reducing oil Imports

The second major program, part of the 1978National Energy Act, establishes excise taxes forpurchases of automobiles with low fuel-economyratings beginning in the model year 1980. Cur-rent “gas-guzzler” taxes range from $200 for carsrated at 14 to 15 mpg in model year 1980 up to$3,850 for specialty cars rated under 12.5 mpgbeginning in model year 1986. Such taxes raisedonly $1.7 million in fiscal 1980.9

OTA’S analysis indicates large uncertainties inboth economic and noneconomic costs of fuel-economy increases. OTA has also identified un-certainties in demand for fuel-efficient cars ascritical to increasing fuel efficiency. These uncer-tainties, together with the desire of domestic man-ufacturers to serve a wide variety of consumertastes with a limited level of capital investment,are mainly responsible for the industry’s reluc-tance to accelerate the development and intro-duction of fuel-efficiency increases. Until fuel ef-ficiency is a–or perhaps the–major selling pointfor new car buyers, this reluctance, under-standably, is likely to continue.

Factors Affecting the Rate ofSynthetic Fuels Production

Unlike increasing automobile fuel efficiency,which may entail restructuring a major existingU.S. industry, production of synthetic fuels in-volves the emergence of a major new industry. *The costs and therefore the profitability of pro-ducing synfuels are influenced by major uncer-tainties that both characterize the economy asa whole and are specific to the synfuels industry.At the economywide level, factors such as theprice of oil, the cost of capital, inflation in generaland hyperinflation in the construction industries,

‘Congress rejected 60 proposals to tax purchases of inefficientnew cars and 19 proposals to raise gasoline taxes between 1973and 1977. In 1977, President Carter proposed a stringent “gasguzzler” tax keyed to CAFE standards which included a rebate pro-gram for purchases of especially efficient cars (which Congress de-cided would violate the General Agreement on Tariffs and Trade—orGAIT). When the current excise taxes were enacted, critics arguedthat they would save only 10,000 bbl/d of imported oil, while theCarter proposal, with higher taxes applied to more vehicles, wasestimated to be able to save 170,000 bbl/d (New York Times, Dec.10, 1978).

*The liquid and gaseous fuel industry may also undergo a restruc-turing, however. Synfuels development will tie up capital in consid-erably larger blocks and for longer periods than historically experi-enced by the industry. A concentration of ownership is also likely.

and the availability of appropriate labor and mate-rials will determine the financial risks that mustbe assumed by investors. Like the auto industry,the synfuels industry is capital-intensive. A mod-erately sized (50,000 B/DOE) fossil synfuels plantcould require an investment of $2 billion to $5billion; the industry’s growth and ability to attractcapital will thus be highly sensitive to the invest-ment climate.

The major constraint on the development ofa synfuels industry is the technical uncertaintyassociated with synfuels processes. There is essen-tially no domestic commercial experience withsynfuels processes, and processes and designconcepts have not yet been adequately demon-strated at a commercial scale. It is thus con-ceivable that design errors and unexpected oper-ational problems could delay construction orcause a completed facility to be inefficient, evena “white elephant” operating at only a fractionof its capacity or at greatly higher cost than antic-ipated. As with any capital-intensive industry,there are high costs associated with the underutili-zation of capacity.

Synfuels production will bean attractive invest-ment if investors view the technical risks as beinglow (i.e., commercial-scale demonstration unitsare successful) and they expect oil prices to risesharply in the future, or if they want to securean early market share in case synfuels do becomecompetitive with the oil market. Unless oil pricesrise more rapidly than synfuel construction costs,however, synfuels plants may not be economical-ly attractive even after the processes are provenand the technical risk is small.

Synthetic Fuels Production—Policy Background

Congress created the National Synfuels Produc-tion Program (NSPP) under the Energy SecurityAct of 1980 (ESA) to promote the rapid develop-ment of a major synfuels production capability.Specific goals are set for 500,000 B/DOE by 1987and 2 MMB/DOE by 1992.10

0IOsynthetic substitutes for petroleum and natural gas are the focusof the NSPP; other programs support development of synthetic fuelsfrom biomass. Biomass subsidy options have been reviewed in de-tail in the Ot%ce of Technology Assessment’s Energy From Bio/ogica/Processes. For the NSPP definition of synfuels see Public Law 96294,sec. 112 (17) (A).


In the first of three phases, the Department ofEnergy (DOE) was authorized to offer financialincentives for the production of alternative or syn-thetic fuels. The original authorizing legislation(Public Law 96-1 26) made about $2.2 billion avail-able, mainly for purchase commitments or priceguarantees pursuant to the provisions of the Fed-eral Nonnuclear Research and Development Actof 1974; total funds were subsequently increasedto approximately $5.5 billion. ’ 11

ESA created the Synthetic Fuels Corp. (SFC) inthe second phase as a quasi-investment bank, toprovide incentives to promote private ownershipand operation of synfuels projects. SFC is backedby funds deposited in a special Energy SecurityReserve in the U.S. Treasury and to be used forfinancial assistance in the form of: 1 ) price guar-antees, purchase agreements, and loan guaran-tees; 2) direct loans; and 3) support to joint ven-tures. ” The governing board of SFC can decide

11This figure does not include an additional $1.27 billion that hasbeen made available for biomass energy, including alcohol fuelsand energy from municipal waste. For further details on this legisla-tion, see Public Law 96-294, Public Law 96-304, and the CRS issuebrief (No. MB70245) “Synthetic Fuels Corporation, Policy and Tech-nology,” by Paul Rothberg.

During the interim program, DOE awarded, in a first “round,”$200 million of these funds, half each for feasibility studies and forcooperative agreements. Of the first $200 million, approximatelytwo-fifths, or $80 million were for biomass projects, while the re-mainder went to synfuels activities. DOE has in past years also pro-vided support for a variety of research and demonstration activitiesto support synfuels development.

DOE originally planned a second “round” of awards for feasibilitystudies and cooperative agreements, but at the request of theReagan administration the $300 million authorized for these awardswas rescinded as an economy measure.

*ln its guidelines to investors, SFC indicates that it strongly favorsprice guarantees, purchase agreements, and loan guarantees, whichemphasize “contingent liabilities. ” The cost to the Governmentof such aid varies with the success of the assisted projects; it is min-imized when projects produce synfuels that can be priced competi-tively with other fuels. To prevent overconcentrating funds, SFCcan give no project or person more than 15 percent of its authorizedfunds, which is about $3 billion during its early years (1981-84).In the case of loan guarantees, SFC cannot assume a financial liabilityfor more than 75 percent of the initial estimated cost of the proj-ect, requiring the assisted company or companies to risk a sizableamount of their own funds. Although there are broad guidelines,the terms of each award will be negotiated separately with projectsponsors.

None of the contingent liability incentives available to SFC canexceed the amounts held for SFC in the U.S. Treasury; that is, SFCcannot “leverage” its funds by guaranteeing loans in excess of itsactual reserves. In the period since the passage of the synfuels legis-lation, estimates of the cost of commercial-scale synfuels plants havecontinued to increase; therefore, unless investors are willing tonegotiate guarantees for smaller percentages of project costs thanallowed by legislation, the amount of synfuels produced by the sub-sidy program may be much smaller than originally anticipated.

which DOE projects it will take over once theboard becomes fully operational. Total fundsauthorized for SFC are approximately $17 billionthrough June 30, 1984.12

The third phase of the NSPP is to begin in mid-1984, at which time Congress may appropriatean additional $68 billion on the basis of a com-prehensive synfuels development strategy to besubmitted by the SFC board. SFC is scheduledto lose the authority to make awards in 1992 andto be terminated in 1997. Some revision of NSPPdates, goals, and/or financing may have to bemade if the production of synfuels falls short ofthe original NSPP goals—as is expected. In OTA’Sjudgment, Congress is unlikely to have sufficientinformation by 1984 about the technical aspectsof synfuels processes to be able to make long-term synfuels decisions.

SFC has received continued political support,and the administration is committed to ensuringthe development of a commercial synfuels indus-try, as announced in its A Plan for EconomicRecovery (February 1981 ). In addition DOE hascommitted about 50 percent of its approximate-ly $5.5 billion, and provisional commitments havebeen made to two projects. *

Reflecting a recent major policy change how-ever, Government support is now to emphasizelong-range, high-risk research and development(R&D) activities that are unlikely to receive privatesponsorship. This shift is likely to have differenteffects on the two main types of synfuels projects:1) projects designed to test and demonstrate aker-native design concepts, learn more about the de-tails of the processes involved, and gain operatingexperience (demonstration plants); and 2) proj-ects designed to demonstrate commercial-scaleprocess units (CSPUs).

Demonstration plants are generally smallerthan commercial-scale plants and are not in-tended to earn a profit. Under the new policies,DOE programs to support demonstration plants

ISynthetic Fuels corp,, “Assisting the Development of SyntheticFuels,” p. 1. The $17-bilIion figure is an approximation. SFC isauthorized to spend up to $2o billion, but the money obligatedin the interim program is to be subtracted from the larger figure.

*These are a loan guarantee of up to $1.5 billion to the GreatPlains Coal Gasification plant in North Dakota and an as yet un-negotiated assistance agreement with the Union Oil Co, oil shaleproject (product purchase guarantee).

. .


are being terminated, but these projects presum-ably can apply to SFC for support.

If CSPUs can be made to operate properly, sev-eral such units might be built and operated inparallel in a commercial synfuels plant. Becausethe process unit is intended to be part of a com-mercial synfuels plant, support for CSPU demon-strations continues to be available through SFC,under the new administration policies.

Termination of DOE support may lead to can-cellation of several demonstration plants, sincethey must now compete against more developedtechnologies for SFC support. * This would resultin a poorer understanding of various synfuelsprocesses and a narrower range of technologyoptions available to potential investors. It couldalso reduce the prospects for commercializingplants capable of producing fuels from a varietyof coals found in different regions of the coun-

*Apparently as a result of reduced Government interest in directlypromoting synfuels, three projects previously supported by DOEhave been canceled (SRC 11 and two high-Btu gasification projects,the Illinois Coal Gasification Project and the CONOCO Project inNoble County, Ohio). Four additional demonstration projects arecontinuing with reduced levels of DOE support and their futuresare in doubt: H-Coal, EDS, Memphis Medium-Btu, and SRC 1. Atleast one upcoming project, not yet at the demonstration stage,has also been canceled in light of recent developments (a Iow-BtuCombustion Engineering project).

try.13 Finally, processes with the greatest immedi-ate (i.e., not necessarily long-term) commercialpromise are likely to be favored by SFC in orderto meet production targets. *

Although every commercial-scale process willhave gone through a demonstration plant stage,the design of the CSPU will also be based on nu-merous other sources of relevant information.Terminating demonstration plant projects will re-duce this pool of information, thereby increas-ing the risks that CSPUS will not function properlyand reducing the design options for correctingmalfunctions. Development of promising longerterm synfuels processes may also be delayed oroverlooked entirely. For these reasons, it is OTA’Sjudgment that DOE’s termination of support fordemonstration plants is likely to reduce the rateat which a synfuels industry is built.

1 JSee paul Roth berg, ‘‘Coal Gasification and Liquefaction, ” CRS

issue brief No. IB77105, Aug. 12, 1981.*Legislation calls for SFC to consider a wide range of alternative

synfuels technologies in order to broaden industry’s experience withthe technical and economic characteristics of many processes. Thisrequirement may conflict with the mandate to meet productiontargets—targets that already appear unrealistic.

POLICY OPTIONSThis section evaluates the major policy options

available for displacing oil imports generally andspecifically for stimulating auto fuel-efficiency in-creases and synthetic fuels production. The evalu-ation is based on the industry characteristics sofar discussed and on the technical analysis whichappears later in this report. In particular, the im-pacts of several policy options that have recent-ly received congressional attention are estimated.Note, however, that policies are not discussedin the context of emergency oil shortfalls.

The policy choices available to Congress dif-fer along several key dimensions: 1) the rate anddegree of oil import displacement; 2) the degreeand specificity of Government intervention andbudgetary effects; 3) the types, magnitude, anddistribution of benefits and costs; 4) implicationsfor the long-term, sustainable, and competitivehealth of the affected industries; 5) the relation-ship of the choices to other Government pro-grams; and 6) the feasibility of future actions. Theselection of policy instruments and resulting

Ch. 3—Policy 45

tradeoffs will reflect the priority ascribed to eachdimension. Policies can generally be designed ei-ther as incentives or penalties, incentives moreclosely approximating the conventional market-place. Policies can be directed at either economy-wide or sector-specific measures.

Economywide Level.–’’Economywide” policychoices are concerned with overall economicand business conditions—as measured by suchindicators as inflation, unemployment, and inter-est rates—that determine the financial health, in-vestment climate, and productive capabilities ofU.S. industries. Fiscal and monetary policies arethe primary instruments in this category; othermeasures could promote innovation, regulatoryreform, technology development, and human re-sources development. Such Government policiesgenerally seek either to remove or to reduce im-pediments to a strong and stable economy, aswell as to raise business and consumer confi-dence in the face of changing economic condi-tions. The advantage of such policies is that theycan be directed at many industries, although theywill have different impacts on the various affectedindustries. They are most commonly preferred asa complement to market forces, because theirscope enables them to enlist the broadest baseof support, and they are best equipped for inte-grating a wide range of economic and social ob-jectives. General economic policy, however, hasonly limited ability to promote the displacementof oil imports and to stimulate specific actions,and may indirectly distort capital flows amongoil displacement alternatives.

Automobile fuel efficiency and synfuels pro-duction (as well as fuel switching and conserva-tion in stationary applications) are influenced bysuch economywide factors as high interest rates,tight credit, increasing resource costs, andchanges in real disposable income. As for theauto industry, general economic conditions in-fluence the ability of consumers to buy cars andthe ability of manufacturers and suppliers to in-vest in needed changes. General economic policycould help to stimulate demand for automobilesby lowering the costs of consumer credit and bymaking credit available. Strong demand for newcars, together with stimuli that reduce the effec-

tive costs of capital and retooling can, in turn,stimulate the supply of fuel-efficient automobiles.

Economywide measures, however, would notinduce consumers to buy domestically manufac-tured vehicles rather than imports; and they couldhave a mixed effect on local automobile produc-tion and employment. Economywide measuresmay facilitate investments by foreign firms in U.S.facilities, but they also assist investments by localproducers in labor-saving equipment and invest-ments by domestic manufacturers in low-costproduction facilities abroad. These investmentsmay ensure the financial health of individual,American-owned firms, but attendant reductionsin domestic employment may aggravate regionaleconomic problems.

Deployment of synfuels production capacitywill also be sensitive to general economic condi-tions: interest rates not only influence the availa-bility of capital for building plants; the capitalcosts also help determine whether products canbe priced competitively. Once established, how-ever, the synfuels industry is expected to be rela-tively insensitive to general economic conditionsto the extent that synfuels are indistinguishablefrom conventional fuels and are competitivelypriced, and the plants do not require frequentretooling. Based on the analysis provided in thisreport, it is OTA’S judgment that favorable econ-omywide conditions, by themselves, are still un-likely to provide sufficient incentive for privatefirms and investors to accelerate the commerciali-zation of a synfuels industry because of the largetechnical risks associated with as yet commercial-ly unproven synfuels processes.

Sector-Specific Level .–Policies can be aimedat specific industries to stimulate industrial com-petitiveness, ease the adjustment of firms to neweconomic conditions (rapid growth, short-termdistress, or long-term decline), or to promote theachievement of national or regional objectives(e.g., national security, regional development).To formulate policies at this level, analyses of indi-vidual sectors and linkages among sectors areessential. The major disadvantages of such poli-cies are that they do not always address the un-derlying causes of market distortions and theydiscriminate against other industries which are

46 INreased Automobile Fuel Efficiency and Synthetic Fuels: Alternatives for Reducing Oil Imports

not similarly assisted. in terms of the auto indus-try, sector-specific policies would be most effec-tive if they addressed the market risk, which isa major factor determining the rate of fuel-economy improvements. The major constraint onrapid deployment of a synthetic fuels industry istechnical uncertainty with respect to unprovenprocesses and, currently, the cost of conventionaloil products.

Economywide Taxation—Oil andTransportation Fuels

General taxation measures are one vehicle forstimulating capital investment across the econ-omy. Economywide taxation measures that spe-cifically relate to displacing oil imports are taxeson oil imports, on oil in general, and on transpor-tation fuels (e.g., gasoline and diesel fuel) in par-ticular. To the extent that the Nation’s energy“problem” is defined as dependence on insecureforeign sources, an oil or transportation fuel taxwould promote security by reducing oil demand.However, an oil or gasoline tax could be counter-productive to the degree that the energy “prob-lem” is defined as a lack of relatively low-cost,high-quality fuels. Consumers may oppose an oilimport tax, even though its impact would beminor compared with that of large OPEC priceincreases, as was the case when an oil import taxof $0.33/bbl was in effect briefly during the Fordadminstration. Its impact, if any, was minor incomparison with that of OPEC’s hikes.

Oil taxes can be imposed either on oil generallyor on oil imports in particular. The advantagesof an oil tax arise because of three features. First,the tax would make all uses of oil more expen-sive without prejudging which kinds of adjust-ments would be most desirable. A general tax onoil would thus reduce consumption and, in turn,imports. Second, the tax could be designed toisolate consumer oil prices from reductions in in-ternational oil prices. For example, if OPEC pricesremain steady through 1984 and if inflation con-tinues at current rates, the real price of oil coulddecline by as much as 20 to 30 percent duringthis period. While perhaps beneficial to con-sumers in the short term, declining real prices forpetroleum products would probably lead to in-

creased petroleum demand. Consistent price sig-nals would also provide assurance both to theauto industry that demand for fuel-efficient carswould be at least sustained if not increased, andto synfuels developers that they would receiveat least a constant real price for their products.Finally, tax revenues could be used, for exam-ple, to support import displacement investments,or to offset some of the potential adverse effectsof the tax (e. g., to fund income support pro-grams).

Taxing only crude oil, however, and not itsproducts could reduce the international compet-itiveness of industries heavily dependent on oil—such as refineries and petrochemical companies. *Furthermore, because oil taxes do not differen-tiate among industries that use oil, they are noteffective means of altering the competitive posi-tion of either automobile fuel economy or syn-fuels production relative to any other method fordisplacing imports (if such alteration is desired).Such taxes could also contribute to inflation gen-erally and would be paid for disproportionatelyby consumers with low incomes. Compensatoryprograms and payments could deal with suchside effects, but at additional implementation andadministrative expense.

Taxes targeted at only oil imports could discrim-inate against companies, and regions of the coun-try, that are heavily dependent on imported oil.It is more likely, however, that import taxeswould cause the general price of oil to increaseto a level close to the price of taxed imports. AnygeneraI price increase, in turn, would create addi-tional revenues for domestic petroleum produc-

*lt is possible that refining activities would relocate overseas unlessadditional import restrictions were also imposed. With respect tosynfuels production, refiners might be able to cut profit marginsand continue to process and sell oil at prices below synfuels pricesin order to maintain refining volumes (which are already rather low).Because many old refineries do not have capital charges, refiningcosts would be dominated by the variable costs of about 15 to20¢/GALproduct plus the oil acquisition costs. Consequently, taxesmay have to raise the cost of imported oil to within about $6/bblof synfuel product (150gal of gasoline equivalent) to ensure thatthe refiners cannot economically use oil imports as copetitors forsynfuels. This would directly harm some companies ened in oilrefining, but this may be a necessary tradeoff to ensure that syn-fuels actually displace imported oil, rather than act to reduce domes-tic petroleum product prices and thereby discourage a reductionin domestic oil consumption.


ers. The total revenues generated by an importtax would thus be only partially received by theGovernment. Compared with a general oil tax,an import tax is thus likely to result in a smallerfraction of revenues being available to the Gov-ernment for additional import displacementmeasures or to offset any adverse impacts ofhigher oil prices. The windfall profits tax capturessome additional revenue, but is more complexto administer than a general tax on all oil.

Another disadvantage of an import tax, fre-quently discussed, is the possibility that oil export-ing nations might see the acceptance of addedcost by U.S. users as an indication that their crudeprices could be further increased without reprisalor economic hardship. This objection probablyis not valid, however, during times of crude oilsurplus in the producing nations.

With respect to transportation fuel consump-tion, a tax either on oil or on transportation fuelsreduces demand for all uses of transportation fuel,including automobile travel, as well as increas-ing the relative demand for fuel efficiency. Itcould reduce new-car sales, however, and couldalso reduce the profitability of truck transports,agriculture, airlines, tourism, and other fuel-dependent industries. Taxes on only gasolinewould avoid some of these problems, but theycould encourage the purchase of diesel-fueledautomobiles.

Gasoline taxes in this country have increasedonly slightly during the past two decades.l A TheFederal tax has been $0.04/gal since 1960, whilethe average State tax has increased from $0.065to $0.08/gal. A gasoline tax that increased theprice of gasoline by, say $0.05/gal (i.e., a 3-per-cent increase over a $1.50 price) would raiseabout $5 billion per year at current consumptionrates, as would a $1 .00/bbl crude oil tax. I n orderto offset inflation since 1960, the current gasolinetax would have to increase by about $0,1 5/gal.Taxes on gasoline are significantly lower in theUnited States than abroad. *

~4See Hans H. Landsberg, Energy Po/;cy Tasks for the 1980s, RFF

reprint 174 (Washington, D. C.: Resources for the Future, 1980).*Taxes on gasoline (per gallon) in 1979 were, as examples, $1.59

in France, $0.88 in the United Kingdom, $1.14 in West Germany,and $1.58 in Italy.

The ultimate effect an oil, gasoline, or dieselfuel tax would have in displacing oil imports de-pends on at least three factors. First, the effective-ness of the tax in the long run depends on theactual purchase and use of fuel-efficient vehicles.Estimates of the responsiveness of demand (its“elasticity”) to changes in gasoline, auto, andother prices vary widely from study to study, butthey suggest that a tax on crude oil or transpor-tation fuels would have to be relatively large tomotivate consumers to trade in their relatively in-efficient cars for more efficient ones. ’ 5 Note,however, that tax provisions per se would not dif-ferentiate between domestic and foreign manu-facturers except insofar as one produces morefuel-efficient vehicles.

Secondly, tax impacts will depend on final oilor fuel prices. The entire tax amount need notbe passed onto consumers if producers are ableto maximize profits by lowering the price of gas-oline, absorbing part of the tax, and increasingsales. As long as demand for oil is slack relativeto supply, at least part of the tax will be absorbedby producers.

Finally, the effect of taxes will depend on thedegree to which driving is reduced. While OTA’Sanalysis of oil savings attributable to fuel economyimprovement assumes a steady increase in vehi-cle miles traveled (VMT) (but a drop in VMT percapita), lower total VMT induced by high gasolineprices would increase actual oil savings. * How-ever, this could also reduce car sales.

While gasoline stations and refineries would beaffected by reduced demand, industry analystsalready expect that the number of service stationsand refineries will decline in the 1980’s. Remain-

IsDemand response is difficult to quantify because there is onlylimited past experience with periods of gasoline price increases(“preenergy crisis” conditions appear to be of limited value for pre-dicting “postcrisis” consumer behavior); crude oil and transporta-tion fuel prices affect consumers in dynamic, multiple ways to alterreal income and demands; and it is difficult to understand demandresponse when vehicles have many different attributes, of whichfuel efficiency is only one. (See Motor Vehic/e Demand Mode/s:Assessment of the State of the Art and Directions for Future Re-search, prepared by Charles River Associates, Inc., for the U.S.Department of Transportation, April 1980.)

*For example, OTA estimates that about half of the 0.5 MMB/Dreduction in gasoline consumed by autos in 1978-80 was due toreduced driving, while about half was due to increased efficiencyof vehicles in use.

48 . Icreased Automobileel Efficiency and Synthetic Fuels: Alternatives for Reducing Oil Imports

ing stations and refineries should be financiallystronger and better able to adapt to gasoline priceincreases. Any reduction in highway trust fundrevenues could presumably be balanced by pro-ceeds from the gasoline tax or other taxes.

Economywide Taxation—Special Provisions

Special taxation provisions are often applied atthe economywide level to promote investment(e.g., by encouraging capital formation and re-structuring cashflow positions). Examples of suchprovisions are investment tax credits, deprecia-tion allowances, R&D tax credits, and capitalgains. As with other taxes, special taxation provi-sions could have differential impacts on industriesand distort private returns to capital. Both theauto and synfuels industries (as well as the elec-tric utility industry), being capital-intensive, couldbenefit from special taxing provisions. The scopefor additional special taxing provisions, however,is believed to be limited because of the many ex-isting provisions,

Although a firm would generally have to beprofitable to take advantage of special taxationprovisions, tax credit sales rules have been ex-panded and liberalized to give unprofitable firmsthe chance to sell their investment tax credits anddepreciation rights. The auto industry has alreadytaken advantage of liberalized rules for selling taxcredits.lb This type of sale can help to strengthenthe financial position of the auto industry, al-though it does not directly encourage increasedfuel economy.

It is speculative to analyze how special taxingprovisions would stimulate investment in syn-fuels. Special taxing provisions have historicallybeen applied at the sector-specific level for do-mestic oil producers in the form of special depre-ciation allowances, and currently for expensingdrilling costs and for foreign tax credits.

lcFOr example, FOrd Motor Co., with losses in excess of $700 mil-lion in 1981, sold its tax credits on 1981 equipment purchases forsomewhere between $100 million and $200 million to IBM (kVash-ington ~05t, Nov. 1 1 , 1 9 8 1 ) .

Research and Development

Government policies and programs could stim-ulate technical R&D at either the economywideor sector-specific levels to help displace oil im-ports. The primary rationale for Government sup-port of R&D is that there are social benefits fromR&D which surpass private gains, in large part be-cause of high front-end learning costs. in addi-tion, the Government tends to support researchthat is too risky for private funding, and whichdoes not, for a variety of reasons, attract privateinvestment in the short term. A major advantageof Government support for R&D is that programscan assist the economy, and specific industries,without direct intervention. However, the typesof basic research that Government has tradition-ally supported often have benefits only in the longterm, so a nearer term oil import savings impliesGovernment involvement in shorter term R&Dareas. Applied R&D also offers the opportunityfor the Government to acquire equity in projectsor royalties from the results of the R&D.

EconomyWide R&D support could be designedto stimulate opportunities for displacing oil im-ports generally and for both increasing automo-bile fuel efficiency and accelerating synfuels pro-duction. Such measures could, as examples,sponsor basic research, promote the climate fortechnical innovation (e.g., increasing the rewardsto innovators through patent laws and/or specialtax incentives), or establish mechanisms for as-sembling and disseminating technical informa-tion. Such nonspecific support, however, is un-likely to have much impact on resolving the spe-cific technological uncertainties that impede bothauto fuel-efficiency increases and synfuels devel-opment.

Although the Government has supported sec-tor-specific R&D in the past, policies have seldomsupported product development with direct com-mercial application except in agriculture andnuclear power. Research to increase automobilefuel economy and to develop synfuels, as wellas other technologies for displacing oil imports,would have direct commercial application.

C h . 3 — P o l i c y 9

There has not yet been any substantial Govern-ment support for R&D to assist the automobileindustry. Several R&D and technology demon-stration programs have been Government-spon-sored, and a joint industry-Government-universityR&D program (the Cooperative Automotive Re-search Program) was attempted unsuccessfully in1979-80.17 Some of the basic research areas thatcould result in substantial long-term fuel-econ-omy payoffs include:

1.

2.3.4.5.

6.

the engine (e.g., advanced alcohol-fueledengines, nonsteady-state combustion, micro-processor controlled fuel injection, high-tem-perature materials);vehicle structure (e.g., crashworthiness);aerodynamics;friction, lubrication, and wear;innovative production technologies for light-weight materials; andexhaust emissions.

The Government might also continue to providesome support for the advanced development ofelectric and/or hybrid vehicles, alternativeengines, and alternative automobile fuels.

The technical uncertainties associated with syn-fuels development are substantial. It is OTA’sjudgment that, even in the presence of favorableeconomywide conditions, investors would nothave sufficient incentive to accelerate synfuelsdevelopment because of the magnitude of thetechnical risks associated with process technol-ogies. For example, one of the major componentsof technical uncertainty is concerned with theflow and abrasive properties of soIid/liquid proc-ess streams. Gaining a basic understanding of theproperties of these streams so that equipment willfunction properly is both a theoretical and an em-pirical engineering challenge. At present, engi-neers must proceed to full-scale commercialplants without adequate analytical descriptionsof how well designs will work. OTA believes theremay be considerable benefit in continuing theoriginal concept of a demonstration program toprovide technical information. The results of both

‘The Justice Department also recently agreed to ease restrictionswhich had barred the four major manufacturers from working to-gether on development, as well as sale and installation, of pollu-tion control devices. See The Washington Post, Nov. 10, 1981.

basic and applied research could lead to impor-tant near- and long-term advances in synfuelstechnology, as well as in other technologies con-cerned with solids handling.le

Trade Protection

Trade protection–tariffs and duties, quotas, lo-cal content requirements—has economywide im-plications but has traditionally been used to tem-porarily insulate specific industries and productsfrom foreign competition. The case for importprotection for the domestic auto industry is basedon the claim that the industry requires only tem-porary protection in order to increase sales andthus to improve its revenue position, to generatecapital for reinvestment, and to position itself formanufacturing fuel-efficient cars. On the otherhand, it is argued that temporary trade protec-tion would neither ameliorate the shot-t-termcompetitive problems of the industry nor pro-mote long-term restructuring for fuel economy.It is seen as inefficient and indirect adjustmentassistance that can lead to higher consumer pricesdue to reduced competition, to higher produc-tion costs for those industries that must compete,unsubsidized, against autos for resources, and toless innovation in general. Trade protection couldalso lead to retaliation on the part of trading part-ners, and some measures are restricted by theGeneral Agreement on Tariffs and Trade(GAIT).19

Import quotas are generally considered less ef-ficient than tariffs in reducing imports and stim-uIating domestic industries. This inefficiencyarises because quotas directly distort both pro-duction and consumption (whereas tariffs changerelative prices), and quotas can be bypassed withproduct differentiation. Duties have not generallyfigured in the policy debate, * but U.S. auto man-ufacturers have been granted temporary tradeprotection in the form of a 3-year Japanese auto-mobile quota agreement. The ultimate effects of

}Bsee also “Repo~ to the American Physical Society by the Study

Group on Research Planning for Coal Utilization and Synthetic FuelProduction,” Rev. Mod Phys, 53, 4 (pt. 2), October 1981.

19’’ General Agreement on Tariffs and Trade, ” 55 U. N.T.5. 194,T.1.A.S. No. 1700 (1947).

*Duties on car imports into the United States are 6 percent; thiscompares with 14 percent into Canada, and 11 percent into France,Italy, Germany, and the United Kingdom.

50 Ž Increased Automobile fuel Efficiency and Synthetic Fuels: Alternatives for Reducing Oil Imports

these quotas on the domestic industry are un-known, but thus far the impacts of import re-straints appear to be small due to low new-carsales. However, if new-car sales recover, theprices of all small cars could increase to the ex-tent that shortages are artificially induced by thetrade restrictions.

Local content provisions are another form oftrade protection that has been discussed in thecontext of the domestic automobile industry(H.R. 51 33). These measures would not displaceoil imports directly but could help to protectdomestic automobile manufacturing jobs. Suchprovisions are generally viewed as being econom-ically inefficient, although they could serve othersocial/equity objectives.

Trade protection is not likely to address any ofthe major issues on which the future of the syn-fuels industry depends. Trade concerns mayeventually arise if large quantities of materials andequipment are imported to construct synfuelsplants or if the United States is in a position toexport synfuels products or production experi-ence.

Trade protection could be used to limit oil im-ports directly. Such a quota, however, could leadto domestic shortages and price increases in theabsence of replacements. The Carter administra-tion placed a quota on oil imports (and exploredalternatives for allocations within the UnitedStates should demand exceed the quota), but itwas set at a level which did not influence imports.import quotas were also in effect from 19591971.20

Sector-Specific Demand Stimuli—Purchase Pricing Mechanisms

to

Demand for increased automobile fuel econ-omy is an extremely important factor influenc-ing the rate of increases in new-car fuel efficien-cy. Autos are large, long-term, and deferrableinvestments for consumers. Furthermore, thedecision to buy a particular car depends on manyattributes, of which fuel economy is only one.Imported oil will not be displaced by the man-

‘“’’Energy Policy,” 2d cd., Congressional Quarterly, Inc., Wash-ington, D. C., March 1981, p. 30.

ufacture of more fuel-efficient cars unless thesecars are actually bought. Demand uncertainty canbe reduced, and the demand for fuel-efficient carscan be stimulated, by raising the costs to con-sumers of buying and operating inefficient carsand/or by lowering the costs of owning relative-ly efficient ones. The risks to manufacturers ofproducing fuel-efficient cars could thus be re-duced. Car ownership costs can be altered by tax-ing gasoline, as discussed, or by taxing/subsidiz-ing automobiles directly.

Synfuels per se should not be directly influ-enced by consumer behavior except insofar asweak demand for liquid fuels limits the profitabil-ity of synfuels production. Some synfuels, how-ever, may not fully conform to end-use fuel speci-fications without more extensive processing orend-use equipment modifications. The extent ofthis potential demand problem cannot be deter-mined in the absence of end-use testing, but islikely to be minor except for alternative fuels suchas methanol.

Purchase Taxes and Subsidies

Automobile purchase taxes or price subsidiescan directly change the costs of owning cars ofdiffering fuel efficiencies. Purchase pricing mech-anisms can be linked either implicitly or explicitlyto fuel-efficiency performance criteria. Currenttaxes are now only loosely related to CAFE stand-ards. The extent to which additional measureswould discourage the purchase of inefficient cars,or encourage the purchase of efficient cars, de-pends on many factors, including the level of theeffective tax (or subsidy), the range of vehiclesaffected, the extent that auto manufacturers’ pric-ing policies counteract the effect of the taxes (orsubsidies), and the responsiveness of consumerbehavior to changes in car prices. * There is also

*The difficulty in quantifying elasticities (i.e., the percentagechange in demand for a l-percent change in price) is discussedin “Economywide Taxation—Oil Imports and Gasoline. ” The inter-national Trade Commission analysis of the Carter gas-guzzler taxproposal implied an elasticity of demand for subcompacts of –0.79(i.e., sales of subcompacts increase less than proportionately withdecreases in their prices) and an elasticity of demand for full-sizecars of – 1.12 (i.e., sales of full-size cars decrease more than propor-tionately with increases in their prices). Assuming that these figuresare accurate, to reduce full-size car sales by 50 percent, for exam-ple, their prices should be raised by about 45 percent, or at least$3,500.


some risk that price subsidies targeting only do-mestic vehicles could violate the GATT provisionswhich prohibit most-favored-nation trading part-ners from taking actions that discriminate againstimports from one another.

Low-interest loans for consumers could stim-ulate the purchase of all new cars, which aregenerally more efficient than the average car onthe road. By tying the interest rates to the newcar’s fuel efficiency, sales of the more fuel-efficient new cars could be stimulated.

Gas-guzzler taxes, as another type of pricingmechanism, would reduce the demand for rel-atively fuel-inefficient cars. However, becausesuch taxes do not discriminate among differenttypes of users, a disproportionate share of thetaxes could be paid by those who are most con-strained to using large vehicles. An equity argu-ment can be made for excepting certain classesof drivers in a tax program (e.g., taxis, hearses),Income support programs could aid in the casesof financial hardship; and tax proceeds could beused to fund these relief measures.

Both purchase subsidies and taxes may addi-tionally and at least temporarily strain the revenueposition of U.S. automakers because they arethe principal suppliers of large cars, but subsidiescould strengthen their long-term position due toincreased car sales.

If Congress wishes to avoid discriminationagainst large cars (which are inherently less fuelefficient than small cars), purchase taxes or sub-sidies could be based on the fuel efficiency of agiven model relative to other models within thesame size or market class. This type of approachwould lead to numerous cases where less fuel-efficient cars are taxed at lower rates or subsi-dized at higher rates than the more fuel-efficientones, but it would create a demand for cars withless powerful engines and technologically im-proved cars (as opposed to simply smaller ones)and it would not favor imports in most cases.

Bounties

Another way to use purchase pricing mecha-nisms to stimulate rapid fleet turnover to higherfuel economy is by offering a gas-guzzler boun-

ty. Bounties could be designed, as examples, asfull payment for a trade-in, or as a payment uponproof of scrappage of a fuel-inefficient car. Be-cause consumers are relatively unresponsive tochanges in prices,21 the bounty would have tobe large to induce significant increases in salesof more fuel-efficient cars. For example, if a valueof –0.3 is assumed for the price elasticity of de-mand for new cars, then for total new car salesto rise by 10 percent, net prices would have tofall by one-third. Since the average new car costsabout $8,000 in 1980-81, a bounty of about$2,700 would be necessary on average to raisenew-car sales by 10 percent. Since many usedcars have market values under $2,700, thisscheme would be profitable for the owners ofused cars. However, it would be costly both tothe Government and to potential buyers of usedcars.

The bounty price would become the effectiveminimum used-car market price and all used-carprices would be proportionately increased. Be-cause bounties distort existing relationships andoperations of both the new and used car markets,bounties would be difficult to design and imple-ment efficiently. Unless bounties were tied tohigh-fuel-efficiency car purchases, they mightneither help manufacturers nor lead to significantfuel savings.

Registration Taxes

Car registration taxes represent another de-mand-side stimulus. These taxes would affect theowners of all automobiles, and they could be ex-plicitly tied to fuel efficiency or some surrogatemeasure* to encourage replacement of fuel-ineffi-cient cars. However, they would make auto own-ership more expensive regardless of the amountand nature of the travel, and they would worktowards reducing demand for autos in the longrun. By effectively lowering consumer income,registration taxes would also disproportionately

21 “Motor Vehicle Demand Models: Assessment of the State ofthe Art and Directions for Future Research, ” prepared by CharlesRivers Associates, Inc., for U.S. Department of Transportation, April1980.

*One possible measure would be ton miles/gallon (e.g., howmuch a vehicle weighs per rate of fuel use), Several foreign coun-tries already have registration taxes that depend on automobileweight and/or engine size.


affect low-income groups. * In addition, a registra-tion policy implies State action, and consistent,concerted implementation may be difficult toachieve.

An important possible side effect of all demand-side stimuli which have the effect of reducinglarge-car demand is that only those domesticmanufacturers with a clear competitive advan-tage in producing large cars will continue to servethis shrinking market. This reorientation of do-mestic production would be consistent with long-term international trends towards corporate con-solidation and a standardized “world car. ”

Methanol

Promoting the use of methanol as an automo-bile fuel is likely to require coordination of supplyand demand stimuli. A limited supply of metha-nol, however, is currently available from thechemical industry. * *

Automotive uses of fuel methanol are principal-ly in a blend (with cosolvents) in gasoline or inengines designed or converted to use straight(neat) methanol. Because many automobiles nowon the road cannot accept methanol-gasolineblends with more than 1 to 3 percent methanol,the blend market for methanol is currently quitelimited (less than 50,000 B/DOE); but the poten-tial market could be expanded if incentives wereprovided to make new cars compatible with high-er percentage blends. This would also add flexi-bility with respect to matching supply and de-mand, which would help to avoid methanol fuelshortages and gluts. The use of blends could beencouraged through direct subsidies and throughapproval of methanol by EPA as a blending agentin gasoline.

Demand for fuel methanol can also be stimu-lated with incentives to convert captive fleets***(current fuel consumption by larger fleets is about

0.6 MMB/DOE22) to methanol. Captive fleets arecurrently more attractive for neat methanol usethan privately owned cars because fleets oftenhave their own fuel storage and pumping facil-ities, which can be converted to methanol at thesame time as the fleet conversion.

Introduction of vehicles for general use whichare fueled with neat methanol probably will re-quire coordinated planning to ensure that neatmethanol is available at service station pumps atabout the same time or before the vehicles ap-pear for sale. However, if this fuel supply prob-lem can be solved (see supply stimuli below) andmethanol is available at prices (per Btu) compar-able to gasoline, it is likely that some auto manu-facturers will supply alcohol-fueled vehicles with-out Government incentives.

Sector-Specific Supply Stimuli—Subsidies and Guarantees

Supply-oriented stimuli–in the form of directsubsidies, grants, and loan, price, and purchaseguarantees–are methods for quickly providingvisible and directed sector-specific support to in-dustries and firms.23 These stimuli, by shifting aportion of the costs and risks to the Government,can provide a temporary inducement to firms toaccelerate investments (i.e., to the auto industryto increase fuel economy and to the synfuels in-dustry to accelerate production). Supply-orientedstimuli can also be structured so as to minimizeor alleviate costly side effects associated with theinvestment or stimulus. The rationale is that mar-ket-driven business practices would not provide,at the time required, nationally desirable outputlevels.

Sector-specific, supply-oriented policy meas-ures share the disadvantages described earlierthat are associated generally with any sector-specific policy approach. In addition, they couldput direct pressure on the Federal budget. De-

*other fees that could discourage fuel use by all drivers includecommuter taxes, car-pooling incentives, and parking fees.

**Total U.S. methanol production, which comes from naturalgas and residual fuel oil, corresponds to about one 50,000 B/DOEsynfuels plant or about 1.5 billion gal of methanol per year.

***A captive fleet is a fleet of cars or trucks owned and operatedby a single business or Government entity and often used primar-ily in a localized area with central refueling facilities also ownedand operated by the fleet owner.

ZZThe Department of Energy (“Assessment of Methane-RelatedFuels for Automotive Fleet Vehicles,” DOE/CE/501 79-1, vol. 2, pp.5-23, February 1982) has estimated that automobile fleets of 10 ormore, truck fleets of 6 or more, and bus fleets consume about 0.6MMB/D of gasoline and about 0.4 MMB/D of diesel. Replacing allof the gasoline would require about 20 billion gal of methanol peryear.

~BSee An Assessment of Oil Shale Technologies, OTA-M-l 18(Washington, D. C.: U.S. Congress, Office of Technology Assess-ment, June, 1980).

Ch. 3—Policy Ž 53

tailed analysis is required to ensure that policiesand programs do not perpetuate inefficient opera-tions, that production changes and other efficientinnovations are not being discouraged, and thattargeted manufacturers are not benefiting inequi-tably. A major implementation problem is in link-ing of payment with performance: to ensure thatsupply-oriented mechanisms promote oil dis-placement, they would need to be contingent onsavings or production performance. Ideally, a de-tailed study of the many factors that determinefuel use could illuminate how much stimulationwould be required to reduce oil imports and howsuch measures would affect the Nation’s energybill. In practice, however, any such study is like-ly to have numerous shortcomings and inaccura-cies.

Loan Guarantees and Grants

The principal sector-specific supply-orientedpolicy mechanisms are loan guarantees, priceand purchase commitments, and direct grants.Of these, grants are the most advantageous forinvestors, since they are a form of direct assist-ance. Grants for improving auto fuel efficiencyand producing synfuels may be unpopular be-cause both alternatives are sponsored by the pri-vate sector and have profit-generating potential.Objections to direct grants could be offset some-what if the Government purchased equity in thecompanies with the money.

Loan guarantees are also advantageous to in-vestors because they allow investors to reducetheir financial exposure in case of default. Unlikegrants which require that the Government appro-priate funds immediately, loan guarantees requireGovernment payment only in the event of a com-pany’s default. Loan guarantees have been ap-plied to both the auto and synfuels industries. Inthe case of autos, loan guarantees were admin-istered by the Government to the Chrysler Corp.when it judged that the costs of not interveningwould be unacceptable from a national view-point. These loan guarantees represent a breakwith historic policy. No direct aid had previous-ly been given because the industry as a wholewas profitable and there was a reluctance bothon the part of the Government to subsidize theprivate sector (except under unusual circum-

stances) and on the part of the private sector toaccept Government support and related condi-tions.

Three policy complications also would arisewhen considering subsidizing domestic automanufacturers:

1. determining the eligibility of foreign firmsthat establish production subsidiaries in theU.S. (e.g., Volkswagen of America, Honda);

2. compliance with GATT provisions; and3. the treatment of auto suppliers. *

Loan guarantees are administered by SFC underESA for the synfuels industry. These guaranteeshave been justified on the basis that the costs andtechnical risks of synfuels production are so greatthat, in the absence of loan guarantees and othersupply-oriented stimulation, private investmentwould be slow in coming. With the large (75 per-cent) guaranteed loans that are possible underESA, investments in synfuels appear to be attrac-tive. * * industry has generally favored Govern-ment support in the form of loan guarantees tostimulate investment, and OTA believes that thisis an effective way of making synfuels investmentsfinancially attractive.24 Because of general infla-tion and steady increases in the estimated costsof synfuels projects, however, the funds currentlyavailable to SFC and the limitation of about $3billion in aid per project may not be adequateto support the number of projects originally en-visioned or allow a full 75-percent loan guaranteefor the larger projects.

*Although many suppliers will have to invest to accommodateautomotive change, it may be most efficient to subsidize only manu-facturers, who would in turn, fund suppliers as appropriate, for tworeasons: first, the amount of U.S. supplier investment (and to a lesserdegree U.S. manufacturer investment) depends on the amount ofoutsourcing and the degree to which foreign supplies are used; andsecond, it is easier to deal with the handful of manufacturers thanthe thousands of suppliers they may use.

**The 61 proposals received by SFC in its first general solicita-tion are a preliminary confirmation of this. These proposals reflectthe variety of approaches considered viable by private industry:14 oil shale projects, eight tar sands (including heavy oil) projects,one coal-oil mixture project, one solid-fuel additive from coal proj-ect, and one hydrogen-from-water project. Of course, general eco-nomic conditions, as well as the price of imported oil, will also havea major impact on the decisions of private investors. These condi-tions will, in turn, heavily influence the terms that SFC is able tonegotiate as it seeks to employ the funds available to it.

24lbid.


Purchase and Price Guarantees

Purchase and price guarantees protect investorsby ensuring that products can be sold at a priceequal to or greater than the minimum guaran-teed, regardless of market conditions. But unlessthe price is set at extremely high levels, these in-centives do not ensure against losses that occurif initial estimates are wrong with respect to cost,price, production volume, or product quality.These guarantees are most appropriate whenmarket demand and price are the major uncer-tainties (and are expected to be “too low”),where commodities are homogeneous, andwhen commodities have a value to the Govern-ment in use or resale. They could, however, dis-tort relationships among producers and consum-ers; they can be administratively complex, andthey do not reduce investors’ financial exposurein the case of poor performance.

Purchase and price guarantees have generallynot been considered viable for the auto industrybecause of the differentiation of its products andthe complexity of manufacturer-dealer-consumerrelationships.

Although purchase and price guarantees do notaddress the central technical uncertainties of syn-fuels production, they may nevertheless be usefulin conjunction with other incentives. Provisionsfor price guarantees and purchase commitmentsare included in the 1980 synfuels legislation.

Subsidies and guarantees can lead to large an-nual investments by the Government. For exam-ple, given that 6 to 8 million cars are produceddomestically each year, subsidies of several hun-dred dollars per car for fuel-economy improve-ments (which corresponds to the investmentsneeded to make the necessary changes) couldrequire annual expenditures of several billiondollars.

To illustrate the magnitude of subsidy thatcould be necessary through a price guarantee toaccelerate synfuels production, assume thatcrude oil costs $40/bbl, that synfuel from a new-ly opened 50,000 B/DOE plant requires a $10subsidy for each barrel of oil replaced, and thatsynfuels production costs follow general inflation.if the real price of oil were to escalate by 2 per-cent per year, the synfuel would have to be sub-

sidized for 11 years at a total cost of about $1billion. If the real price of oil escalates at 4 per-cent per year, the period of subsidization and thetotal cost would be half as large. Similarly, a1-percent real inflation rate for oil would doublethe duration and magnitude of the subsidy. Thus,price guarantee subsidies can reach levels thatare a significant fraction of the investment initiallyneeded to build the plant.

The Government could, however, require re-payment of a subsidy if the manufacture of fuel-efficient cars or the production of synfuels be-came profitable without subsidies.

Methanol

The supply incentives mentioned above andthose described under “demand stimuli” areprobably adequate to encourage production ofmethanol from coal for use by the chemical mar-ket and some captive fleets of automobiles, and,possibly, as blends in gasoline. However, addi-tional supply incentives may be necessary to en-courage the use of methanol in automobileswhich are not part of a captive fleet.

Once significant quantities (probably more than0.1 to 0.2 MMB/DOE) of methanol are being usedin captive fleets and, possibly, in gasoline blends,it may be possible to offer methanol for sale tothe public in enough places to make ownershipof a methanol-fueled vehicle practical for individ-uals. Incentives can be offered to owners of meth-anol-fueled captive fleets, who have their ownmethanol storage and pumping facilities, to sellmethanol to the public. Incentives can also begiven to service station owners who sell methanolblends to install methanol storage tanks and blendthe methanol with gasoline at the pump. Theycould then sell straight methanol, as well.

Many owners of captive fleets probably can-not be easily induced to offer methanol for sale,because it would not be related to their otherbusiness activities and would be tantamount toentering the service station business. Similarly,very large economic incentives may initially benecessary to induce service station owners to in-stall methanol facilities, because the investmentwould not lead to a near-term increase in sales.


On the other hand, it could be mandated thatany supplies of methanol used for Government-owned captive fleets be made available for publicsale. And some captive fleet and service stationowners would be willing to offer methanol to gainan early market share or for the financial incen-tives offered by the Government. If these mone-tary and nonmonetary incentives were adequate,methanol could compete directly with gasolineand diesel fuel as an automobile fuel.

Regulations on Output

One of the most direct policy mechanisms forpromoting alternatives that can displace oil im-ports is regulation. Regulations are a common,if controversial, form of Government interventionin the economy. Although their effects can be felteconomywide, regulations are typically directedat specific industries or products. In general, theywould target the supply aspects of oil import al-ternatives. Measures could also be designed totarget consumers (e.g., the 55-mph speed limit,end-use fuel restrictions in the stationary sector),but the Government has traditionally been reluc-tant to mandate changes in consumer behaviorand habits.

Regulations can be designed for two major pur-poses. First, they can serve to protect the publicfrom the side effects caused by the conduct ofindustrial activities. These effects include impactson the environment, health, and safety which arediscussed in the next section. Regulations canalso be used to determine outputs directly—thelevel of consumption or production of fuels–ifthe market is unable to ensure desirable levels.

the investment risks since, although regulationscan affect the supply of fuel-efficient cars, theydo not directly influence purchases. Through the1970’s consumers failed to demonstrate a con-sistent demand for fuel economy, * and the CAFEstandards probably increased fuel efficiencyabove what the market would have achieved.And recent data (fall 1981 ) show that, in fact, theproportion of relatively large cars sold has onceagain increased compared with the number ofsmaller cars sold.

The arguments for extending CAFE standardsbeyond 1985 are inconclusive. To the extent thatCAFE standards are met through sales of smallercars, as opposed to purely technological changes,U.S. manufacturers must increasingly competewith imports for the small-car market. Increasinglystringent fuel economy standards could, there-fore, result in higher import levels if domesticmanufacturers are unable to increase their com-petitiveness in this market, despite the productchanges they have made. Post-1 985 standards arealso likely to require additional capital for morerounds of redesigning and retooling. But post-1985 standards could result in important fuel sav-ings to the Nation, especially if the demand forfuel-efficient cars remains sluggish. Additional de-mand stimuli may also be necessary, dependingon national and international conditions, to en-sure that fuel-efficient cars are bought.

In considering the effects of CAFE standards itis important to recognize that CAFE standards donot distinguish among average efficiency in-creases that result from: 1) technological improve-

The auto industry has been regulated in theUnited States in the areas of emissions, safety, andmore recently fuel economy. The major Govern-ment program mandating fuel-efficiency increasesis the CAFE standards. Whether or not to increasethese standards beyond levels set by current legis-lation for 1985 and beyond may be a major up-coming decision before Congress.

Effectiveness of the CAFE standards in spurringfuel economy improvements is controversial, Animportant feature of these standards is that theyare effective only if they force manufacturers todo more than consumers demand. This increases

*The drop in demand for fuel efficiency and the resurgence inlarge-car demand in the mid to late 1970’s led manufacturers topetition (unsuccessfully) NHTSA to lower CAFE standards for theearly 1980’s because sluggish sales of fuel-efficient cars madenecessary investments appear especially costly and risky. Markettrends in the 1970’s also led manufacturers to concentrate initiallyon improving fuel economy for relatively large cars rather than ondeveloping new small-car designs. Manufacturers attributed theirexpectation to exceed voluntarily the 1985 CAFE standard of 27.5mpg to renewed, strong demand for fuel economy arising from the1979 oil crisis and increases in gasoline prices. This increase in de-mand and current industry efforts to raise fuel economy recentlyled NHTSA, which administers the CAFE program, to terminate

rulemaking with respect to post-1 985 average fuel-economy im-provements (Fed. Reg. 22243, Apr. 16, 1981 ). A petition from theCenter for Auto Safety that requested NHTSA to continue rule-making was also subsequently denied (Fed. Reg. 48383, Oct. 1,1981 ).

98-281 IC - 82 - 5 : ~11, 3


ments, 2) consumers’ purchasing the more fuel-efficient cars in each size class, and 3) consumerspurchasing smaller cars. Depending on marketdemand, success of technical developments andauto manufacturers’ financial positions and capi-tal stock, CAFE standards could be met throughvarious mixes of the three (see table 8). Conse-quently, without special provisions it probablyis impossible to establish conventional CAFEstandards which simultaneously: 1 ) are effective(i.e., increase new-car fuel efficiency above whatmarket forces would dictate), and 2) do not pro-mote the sales of small imported cars. Separatefuel-efficiency standards for each automobile sizeor market class could significantly reduce the in-direct promotion of small-car sales; however, thiswould greatly reduce automobile companies’flexibility in responding to the regulations.

The NSPP sets targets for synfuels productionbut the mandating of synfuels output has notbeen of central congressional interest. The majordifficulty associated with developing synfuelsstems from technical uncertainties which, in turn,affect the likely cost at which synfuels initially willbe produced. in addition, contributions to oil im-port savings from synfuels would not be madeincrementally (as with increasing automobile fuelefficiency) but rather depend on the proper func-tioning of large-scale facilities. As experience andknowledge is gained, it may become possible toestablish realistically achievable production levelsif the Government desires an assured level of syn-fuels supply.

Table 8.—Potential Average New-CarFuel Efficiency in 1995

Fuel efficiency ofCar size class average model (mpg)a

Large . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ....30-45Medium . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ....45-60Small . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ....60-75

Average new-car fuelSize mix of cars sold efficiency (mpg)a

1961 size mixb. . . . . . . . . . . . . . . . . . . . . . . . ....40-50Moderate shift to small carsc. . . . . . . . . . . ....45-60Large shift to small carsd . . . . . . . . . . . . . . ....50-65a All mpg figures rounded to nearest 5 mpg. Mpg refers to the composite con-

slstlno of 55 percent EPA city cycle and 45 percent EPA highway cycle,b 1~1 gales: 47 percent large cars, da percent medium-sized cars, and 5 per-

cent small cars.C l= gales: m percent large Cars, 55 percent medium, 25 percent small.d l% gales: s percerlt large cars, 45 percent medium, 50 percent smali.


One possible form of regulation would be tostipulate that a certain percentage of the outputfrom domestic oil producers be synfuels. How-ever, this provision would be unworkable forsmall oil producers, so it would have to be tar-geted at the larger oil companies. Similar prob-lems arise with regulations aimed at refiners orretailers. Furthermore, because refining and re-tailing are considerably less profitable than gasand oil production, regulations aimed at the form-er might induce some of the companies that arevertically integrated to abandon refining ratherthan to incur the added costs and risks. For thesereasons, it would probably be very difficult to ad-minister mandates on synfuel content.

Other Effects


Both increased automobile fuel efficiency andsynthetic fuels production have the potential forcreating large-scale environmental, health, andsafety hazards. A principal rationale for policy in-tervention is the general past failure of privatemarkets to internalize these other effects in invest-ment decisions and operating practices. Policiesto protect the public have tended to take the formof regulations that govern known or anticipatedimpacts through performance standards or con-trol specifications.

Apart from fuel efficiency, the auto industry isregulated in the areas of emissions and safety.Emissions standards require that each vehicle–and automobile safety standards require that eachof certain vehicle parts-meet minimum perform-ance standards. (By contrast, fuel-economy stand-ards are for fleet averages.) There are proposalsbefore Congress to delay, modify, or eliminateover 30 automotive-related environmental andsafety regulations.25

A potential threat to the public from size andweight reduction of vehicles used to increase fuelefficiency is decreased automotive safety. Thebasic policy issue is whether the Governmentshould act to help prevent future highway fatal-ities if consumer demand for safety does not result

25 See Gwenell Bass, “The U.S. Auto Industry: The Situation inthe Eighties,” CRS issue brief No. IB81054, Sept. 30, 1981.


in adequate safeguards. * Regulatory policy could,as examples, reconsider the passive restraint pro-gram (rulemaking for that program was ter-minated by NHTSA although a recent U.S. Courtof Appeals ruling has reinstated the program, atleast for the moment), mandate the use of safetybelts, strengthen crashworthiness design stand-ards, or maintain or tighten speed limits. Othertypes of programs could provide for stricter driverlicensing standards and improved road mainte-nance and traffic control, support R&D, and pro-vide for driver safety education. Another poten-tial adverse effect is the air quality impact of alarge increase in diesel-powered autos. Policy al-ternatives include more stringent particulate andNOX emission regulations for diesel engines andGovernment assistance in diesel health effectsresearch and emissions control development.

Potential environmental and worker-relatedproblems associated with synthetic fuels develop-ment (e. g., contamination of drinking water, re-lease of cancer-causing agents and other hazard-ous pollutants, highly visible plant upsets, obnox-ious odors, and localized water availability con-flicts) are substantial and have considerablepotential for arousing strong public opposition.There are also elements of the present synfuelsdevelopment strategy that appear to increase thepotential for adverse impacts. These elements in-clude the proposed siting of some synfuels dem-onstration plants close to heavily populated areas,research budget cuts at EPA and the proposeddismantling of DOE, the current policy to shiftenvironmental management responsibilities toState and local agencies without a concomitantshifting of resources, and an industry environ-mental control program that appears reluctant tocommit resources to currently unregulated pollut-ants and that may be overconfident about theperformance of integrated control systems.**

“Manufacturers may not pursue safety for fear that the addedcost will dampen market demand. In most cases safety has not beena strong selling point in automobiles in the past.

*“In apparent response to their confidence that adequate environ-mental control of synfuels plants will involve only “fine tuning”of existing control technologies, developers have passed up someopportunities to test out control systems on demonstration plants.For example, Exxon feeds the wastes from its Baytown, Tex., EDSplant into a neighboring refinery rather than developing and testingspecific controls for the plant. In OTA’s opinion this increases therisk of unforeseen problems at the first large-scale plants. Such prob-lems appear quite possible given the differences between the con-ditions under which proposed control systems have been usedpreviously (in chemical plants, refineries, etc.) and the expectedconditions in synfuels plants.

Finally, the multiplicity of pollutants associatedwith synfuels production and the difficulty of de-tecting and evaluating some of the potential im-pacts (e.g., long-term cancer impacts from low-Ievel exposures), coupled with the above factors,leads to a strong concern about the adequacy offuture regulation of a synfuels industry.

Government actions targeted at the potentialrisks of synthetic fuels development may be animportant factor in assuring that the risks areproperly measured and in causing the private sec-tor to account for these risks. A problem the Gov-ernment faces, however, is that premature adop-tion of rigid standards could ultimately act to sti-fle innovation or force suboptimal environmentaldecisions. Also, the capital-intensive nature of theindustry leaves it vulnerable to delays caused byshifts in environmental requirements or standardsthat ultimately prove unattainable.

The existence of these problems places a pre-mium on an intensive research program and around of demonstration plants, that include fullenvironmental control systems, to avoid surprisesand provide timely information for intelligent reg-ulation. Also, the impact of an environmental sur-prise might be minimized by choosing isolatedsites and requiring particularly strict controls forthe first round of plants, thus minimizing the ac-tual impact suffered from excessive discharges orother problems.

The vulnerability of synfuels plants to schedul-ing delays has also generated pressures on Con-gress and State legislatures to streamline environ-mental permitting for energy facilities. Althoughit is too early to assess recent streamlining effortsat both the Federal and State levels, considerableimprovement appears possible without a full-scale Energy Mobilization Board.26 In most cases,regulatory delays are important only to the ex-tent they delay construction starts or requirechanges in plant design; otherwise, all necessarypermits are likely to be obtained before signifi-cant construction investment has been made. De-lays after construction has started could, how-ever, be costly. A 3-year delay, for example,

ZbCongressional Research Service, “Synfuels From Coal and theNational Synfuels Production Program: Technical, Environmental,and Economic Aspects. ”

58 ● Increased Automobile Fuel Efficiency and Synthetic Fuel: Alternatives for Reducing Oil Imports

could increase synfuels product costs by 20 per-cent or more. In addition, the risk that retrofit-ting may be required will remain until synfuelprocesses have been proven and both emissionsand products extensively tested. Formulating reg-ulatory policy entails an assessment of the fullrange of regulatory costs to industry versus thepossible costs arising in the absence of policy.At present these complex tradeoffs are often de-termined in a lengthy, case-by-case process basedon judicial interpretations.

policy alternatives to regulation include effluentcharges and pollution vouchers. Although suchmechanisms have a strong theoretical basis, thereis a general lack of practical experience in usingthem. Also, the toxic pollutants of most concernin synfuels production cannot safely be tradedoff among sources the way pollutants such as SO2

and NOX can be.

There are two additional levels for policy in-volvement. First, Congress could decide to in-crease the environmental capabilities of respon-sible regulatory agencies. One specific option isto target resources for specific State and local en-vironmental agencies, as was done under theClean Air Act in the early 1970’s. As a part of thisoption, Congress may also wish to investigate theeffects of the programmatic changes and budgetreductions for synfuels environmental researchand control system development at EPA andDOE.

Secondly, environmental concerns could be in-tegrated directly into financial support decisions—i.e., of SFC. Although some would claim thatthis latter option is redundant given current envi-ronmental legislation, there are neverthelessmany concerns (e.g., siting) which are notwell-addressed by existing laws. In addition, theprotection of SFC investments would be well-served by an ability to influence environmentalplanning. SFC has not yet moved aggressively tobuild a technical capability for the environmen-tal assessment of projects it will support.

The availability of water resources may posespecial problems for policy because of the pres-ent controversies surrounding the allocation and

use of increasingly scarce supplies.27 How con-flicts are resolved in areas where users presentlyor could potentially compete for water will haveimportant implications for the distribution of costsand benefits to all water users, especially sincethe costs of procuring water are likely to be small(in comparison with other costs) for the synfuelsindustry. Present water policies and planningmechanisms are fragmented and generally inade-quate to assess water availability and plan forfuture water needs on a consistent, compre-hensive, and continuous basis. Because of themagnitude, diversity, and nationwide distributionof water resource problems and because the out-come of water-resource allocation conflicts willhave local, State, regional, and National impacts,the Federal Government has an important roleto play in improving water resource managementpractices in cooperation with the States. Majorpolicy issues include the resolution of uncertain-ties surrounding water rights and future waterneeds and the definition of responsibilities, objec-tives, and priorities for water planning and alloca-tion. Legislation pending before Congress (e.g.,S.1095 and H.R. 3432, which both call for thedismantling of the U.S. Water Resources Coun-cil) seeks to redefine the respective responsibil-ities of Federal and State Governments and toclarify the role of regional and local interests inmanaging water resources.

Social Adjustment Assistance

Increasing automobile fuel efficiency and de-veloping a synfuels industry will result in socialcosts and benefits which are side effects, i.e., ef-fects that are external to the transactions madebetween consumers and producers. The move-ment of capital and labor as a result of industrialchange (i.e., restructuring, contraction, orgrowth) will have implications not only for thecharacter of the labor market but also, conse-quently, for lifestyles and standards of living.

ZzFOr further discussion of these issues see An Assessment of oilShale Technologies, OTA-M-118 (Washington, D. C,: U.S. Congress,OtYice of Technology Assessment, June 1980); and A TechnologyAssessment of Coal Slurry Pipelines, OTA-E-60 (Washington, D. C.,U.S. Congress, Office of Technology Assessment, March 1978).

Ch. 3—Policy • 59

in the auto industry, the major social effects arerelated to job losses resulting from structural ad-justment. In the synfuels industry, the major socialexternalities are related to new employment andresult from large, rapid and fluctuating popula-tion growth in some areas where the industry maylocate. Government policy may be importantboth for easing those social adjustments that themarket does not address and for ensuring thatassociated costs do not fall disproportionately onparticular groups. Social-adjustment assistance inthis country has generally been limited in the pastto sectors affected by international trade and toseveral programs focusing on regional adjust-ment.

Labor market dislocations are of primary impor-tance to the Nation because of the penetration,numbers, and dispersion of auto-related jobsthroughout the economy as well as the geograph-ic concentration of auto production jobs. Restruc-turing of the industry for improved internationalcompetitiveness, productivity, and fuel efficien-cy is resulting in what is likely to be a long-termdecline in auto-related employment.

The problem of unemployment in the auto in-dustry could be addressed by policy measuresthat seek to ease the adjustment of firms, workers,and communities to changing economic condi-tions. Policy options include, as examples: reloca-tion assistance, support of retraining programsand training institutions, local content provisions,manpower training vouchers for targeted individ-uals, plant-closing restrictions, tax incentives toother industries (or regions) to attract displacedworkers, and community aid programs (e. g., todiversify local economies).

Some assistance has been available under theTrade Act of 1974 provisions and through Hous-ing and Urban Development, the Economic De-velopment Administration, and other Govern-ment agency programs. These programs havegenerally been limited in scope and funding andhave generally required evidence of economicdistress (i.e., they are not preemptive). They arealso candidates for curtailment under proposedFederal budget cuts. Note that because employ-ment displacement depends in part on laborcosts, automobile-related employment levels will

also vary with the degree to which autoworkersaccept changes in compensation and work rules,behavior which is not generally subject to directFederal policy initiatives.

Major social side effects arise from synfuels de-velopment because the communities which ab-sorb the large, rapid population increases (a por-tion of which is only temporary) are vulnerableto institutional and social disruptions. These ex-ternalities could constrain synfuels developmentby generating public opposition to synfuels andby adversely affecting worker productivity. Theprincipal policy issues relate to who will bear thecosts of managing and mitigating these disrup-tions and how up-front capital can be made avail-able to finance necessary public facilities andservices. Those who view social impacts as theprice of regional development emphasize the re-sponsibilities of State and local governmentsworking with private developers. Those who as-sociate local impacts primarily with the pursuitof national energy objectives call for a continuedand expanded Federal role.

There are also many questions of equity thatarise in allocating resources among differentareas, because of the large variations in themagnitude and character of adverse impacts andthe resources available to cope with these im-pacts. An acceptable assistance program mustdeal with the problem that some of the shortagesof impact-mitigation resources are caused by limi-tations on planning and borrowing powers im-posed by local and State governments them-selves.

Current policies to deal with the social impactsof energy development are unable to addressconsistently and comprehensively the cumulativeimpacts arising from the large-scale, rapid-growthsituations that characterize synfuels development.Government policies could be directed at eitherenergy development generally or synfuels pro-duction specifically, and could provide, as ex-amples: financial aid, technical assistance, growthmanagement planning assistance, regulation(e.g., with respect to siting, phasing, pacing, mon-itoring), lending and borrowing assistance, or tax-ing provisions. The various forms of technical andfinancial assistance for growth management are

60 Ž Increased Automobile Fuel Efficiency and Synthetic Fuels: Alternatives for Reducing Oil Imports

examined in detail in previous OTA stud-ies.28 29 30 31 All relevant Federal programs havebeen targeted for substantial budget reductions,or elimination, in fiscal year 1982 under proposalssubmitted to Congress by the present administra-tion.32

The development of a synfuels plant will leadto the creation of new jobs in construction and

ZBAn Assessment of Oil Shale Technologies, Op. cit.zgThe Dir~t L/se of~a~ OTA-E-86 (Washington, D. C.: U.S. Con-

gress, Office of Technology Assessment, April 1979).30&fanagement of Fue/ and Nonfuel Minerals in Federal Land,

OTA-M-88 (Washington, D. C.: U.S. Congress, Office of TechnologyAssessment, April 1979).

31An ~sessment of ~ve~pment and production Potential of Federa/ Coa/ Leases, OTA-M-150 (Washington, D. C., U.S. Congress,Office of Technology Assessment, December 1981).

3zCongressional Research Service, “Energy Impact Assistance Leg-islation, “ issue brief No. 1679022, Sept. 9, 1981.

engineering. Technically qualified personnelshould be available for most of these jobs. How-ever, a shortage of experienced chemical processengineers and project managers could arise, caus-ing costly mistakes and production delays. Theoverall number of chemical process engineers,for example, would have to increase by aboutone-third by the mid to late 1980’s to accom-modate an optimistic level of synfuels plant con-struction.

The Federal Government could encourage theeducation of engineers by providing financial sup-port for facilities, equipment, retraining programs,scholarships, and the hiring and retraining of fac-ulty. Training in skills needed for complex proj-ect management could similarly be stimulated.The auto industry would also benefit from pro-grams to train engineers if the industry pursuedextensive development efforts domestically.

CONCLUSIONSBoth increasing automobile fuel efficiency and

synfuels production have economic and noneco-nomic risks and external costs. The decision topursue either, or both, alternatives–as well asto pursue the third major technical alternative offuel switching and conservation in stationary usesof petroleum—depends on the desired rate andlevel of oil import displacement and what the Na-tion is prepared to spend to achieve its oil-dis-placement goals.

The availability and cost of capital are especiallyimportant for the automobile and synfuels indus-tries, since they are both capital-intensive. Gen-eral economic conditions affect consumer confi-dence and purchasing power. Among the policiesmentioned in this chapter are general tax policiesand special taxing provisions which would en-courage capital formation and stimulate industrialinnovation economywide.

The rate at which automobile fuel efficiencycan be increased and a synfuels industry devel-oped are also affected by factors that are specificto each alternative. Contributions to oil-importdisplacement from increased automobile fuel effi-ciency depend critically on consumer demandfor fuel-efficient cars. Government actions to stim-

ulate demand are a direct way to help ensure thatfuel-efficient cars are bought and, in turn, thatthey will be produced. Demand-oriented meas-ures that appear promising and that deserve fur-ther analysis include registration, purchase, andfuel taxes and purchase subsidies. Supply incen-tives, depending on their nature, could help man-ufacturers pay for the investments necessary toincrease fuel efficiency, especially if there is anabsence of strong demand for either cars in gen-eral or fuel-efficient cars in particular. In the caseof weak demand for efficiency, increasing CAFEstandards beyond the 1985 level may help to en-sure continued oil import displacement. How-ever, the increased cost of the efficiency increasescould reduce new-car sales and thereby reducethe potential savings. In general, a combinationof demand and supply incentives would be themost effective means of promoting more efficientfuel use in automobiles. This would contrast withpast policy, which has been aimed largely at pro-ducers.

The success of synfuels development in displac-ing oil imports hinges on the resolution of majortechnical uncertainties associated with as yet un-


proven processes. private investments are likelyto be accelerated once processes are demon-strated in commercial-scale units—provided theprocesses are economically competitive sourcesof fuels. The high costs and other risks associatedwith demonstration projects are likely to necessi-tate Government support if synfuels productionis to become a significant fuel source by the endof the century.

Other policy considerations for displacing oilimports are applicable generally to planning ina world of uncertainty. First, flexible and nonspe-cific policy interventions provide both public andcorporate decision makers with the maximum op-portunities to adjust internally to changing eco-nomic and technical circumstances. Secondly,periodic reviews and adjustments can help pre-vent prematurely locking the Nation into techni-cal choices that discourage a continuing searchfor better methods, although too much flexibil-ity can lead to ad hoc programs.

A long-term, stable policy commitment to oilimport displacement, and to alternatives for dis-placing imports, is essential in order to send clearsignals about Government intentions and pro-mote mutual confidence in any public-privaterelationship. In the past, the Government hassometimes sent conflicting signals. For example,concurrent Government programs were in place,on the one hand, to encourage automobile fueleconomy with CAFE standards and, on the otherhand, to discourage fuel conservation with pricecontrols on oil which helped to keep the priceof gasoline low. *

Increased automobile fuel economy and syn-fuels production contribute in different ways tothe Nation’s energy security. The advantages ofautomobile fuel efficiency include the following:1) through conservation, it directly eliminates theneed for oil imports in the Nation’s highest petro-leum-consuming sector; 2) after large numbersof fuel-efficient vehicles have been sold, the fuel

*U.S. policies in the 1970’s also implicitly encouraged oil usefor stationary purposes (e.g., Federal curtailment policy for naturalgas).

savings does not depend on the operation of afew large plants, and there will continue to befuel savings even if particular vehicles performbelow standards; 3) it does not result in a netreduction of natural energy resources and thuspreserves options for future generations; and4) although there are long Ieadtimes for commer-cializing new products in the auto industry, sav-ings are already occurring as technologies are dif-fusing into the consumer market. However, ifmarket and/or Government pressures for in-creased automobile fuel efficiency damage theU.S. auto industry, there will be repercussionsthroughout the economy.

The principal national security advantage ofsynfuels production is that it may provide long-term strategic insurance against sustained short-falls. Rapid and successful deployment could con-ceivably serve to reduce the rate at which oil im-port prices increase and thus help to reduce infla-tion. The vulnerability of the synfuels alternativeis related to the complexity of technical controls,the high risks and costs of failure, potentially haz-ardous environmental side effects, institutionalbarriers to deployment, and, in some cases, thegeographic concentration of facilities.

Because increasing auto fuel efficiency and syn-fuels development are both capital-intensive,each will incur major economic penalties if facili-ties function below capacity. However, becausethe “normal” rate of capital turnover is likely tobe lower in the synfuels than the auto industry,synfuels production will be more limited in adapt-ing to changing demands.

Developing a long-term, coordinated, andcomprehensive energy policy will be an incre-mental process. A prime objective is to choosea least cost mix of options for reducing oil im-ports. Because investment costs (per barrel perday of oil saved or replaced) for the various op-tions considered in this report for the 1990’s arehighly uncertain, yet appear to be comparablein magnitude, the judgment of relative costs willdepend largely on value assessments of the var-ious externalties of pursuing each option.


APPENDIX 3A.–POLICY OPTIONS TOREDUCE STATIONARY OIL USE

Conservation

1. Tax credits for investments in conservation:Ž Current 15-percent credit for investments by

homeowners–single-family and 4-unit or lessmuItifamiIy.

● 10-percent tax credit for energy efficiency or re-newable resource investments by industry.

2. Residential Conservation Service:● UtiIity audit service for homeowners.● Proposed extension to include apartment and

commercial buildings.3. Subsidized loans to homeowners to finance conser-

vation investments:● Currently the purpose of the Solar and Conserva-

tion Bank.● Private savings and loan institutions also offer be-

low-market loans for conservation in some cases.4. Targeted tax credits (20 percent) for investments in

energy efficiency by industry:● Currently proposed in legislation now before

Congress.● Tax credit in addition to current credits.

5. State public utility commission actions to encourageconservation efforts by utilities:

Allowance of conservation investments in the ratebase of utilities (proposal).Permission to sell saved energy to private in-vestors (proposals).Allowing utilities to set up separate companies toprovide conservation services—now occurs insome cases,

6. Legislating standards and/or information:●

●

●

●

Appliance efficiency standards–labeling of appli-ances.Building standards–currently prescriptive stand-ards through the minimum property standards ofthe Department of Housing and Urban Develop-ment.Building energy performance standards are legis-lated but currently not enforced.Efficiency standards for industrial electric motorswere proposed but never enacted.

Fuel Conversion

1. Prohibition on oil use by utility boilers and largeindustrial boilers:

● Principal focus of the Fuel Use Act.● Goal to eliminate use of oil by 1990.

2. Financial assistance for utilities to convert from oilto coal:Ž State commissions have allowed New England

Electric Co. to secure a “loan” from their custom-ers to pay for an oil-to-coal conversion.

● Federal legislation proposed to provide loan guar-antees for these conversions was never passedand is not likely to be pursued now.

3. Legislation to remove regulatory restrictions on useof natural gas by industry:

● Currently part of several proposals to encourageconversion to natural gas.

• Current regulations (Federal and State) either pro-hibit or discourage natural-gas use for many appli-cations that now use fuel oil.

4. Environmental regulations affecting coal use in in-dustry:

● Lowering of emission standards for applicationsbelow a certain size.

• Financial assistance to help install control tech-nologies.

General

1. R&D to increase efficiency of end-use technologies:● Promotes general conservation.• Can also be directed at developing efficient elec-

tric energy using technologies to make the eco-nomics of switching to electricity attractive.

2. Tax on oil–either on imports or on specific prod-ucts such as fuel oil for boilers or space heating:

3. Economic incentives for development of unconven-tional natural gas:

● Currently unconventional natural gas is complete-ly deregulated.

• Tax credits to encourage development of uncon-ventional gas (this is currently not available).


APPENDIX 3B.–ADDITIONAL CRS REFERENCESThe Congressional Research Service has recently 3.

published many reports on various aspects of energypolicy to which the reader is referred. These reportsinclude the following:

1.

2.

Rothberg, Paul, ‘Synthetic Fuels Corporation and 4.National Synfuels Policy, ” CRS issue brief No.1681139, Oct. 13, 1981.Chahill, Kenneth, “Low-Income Energy Assist- 5.ance Reauthorization: Proposals and Issues, ” CRSmini brief No. MB81227, Aug. 26, 1981.

Abbasi, Susan R., “Energy Policy: New DirectionsIndicated by the Reagan Administration’s BudgetProposals, ” CRS mini brief No. MB81222, June29, 1981.Parker, Larry, Bamberger, Robert L., and Behrens,Carl, “Energy and the 97th Congress: Overview,”CRS issue brief No. 1681112, Sept. 15, 1981.Rothberg, Paul, “Synthetic Fuels Corporation,Policy and Technology,” CRS mini-brief No.MB79245, July 13, 1979, updated Apr. 20, 1981.

Page

introduction . . . . . . . . . . . . . . . . . . . . . . . . 67

What Do Oil Imports Cost? . . . . . . . . . . . 67Dependence of Price on Quantity

Imported . . . . . . . . . . . . . . . . . . . . . . . 68Loss of U.S. Jobs and GNP Caused by

Rising Oil payments and byPotential Supply Interruptions . . . . . . 68

Military Outlays and Foreign PolicyDirections Forced by Oil ImportDependence . . . . . . . . . . . . . . . . . . . . 69

Conclusion . . . . . . . . . . . . . . . . . . . . . . . 69

How Quickly Can Oil Imports BeReduced? . . . . . . . . . . . . . . . . . . . . . . . 69

Chapter 4

Issues and Findings

What are the Investment Costs forReducing U.S. Oil Consumption? . . . 74

Introduction . . . . . . . . . . . . . . . . . . . . . . 74Conventional Oil and Gas Production . 74Automobile Fuel Efficiency . . . . . . . . . . 75Electric Vehicles . . . . . . . . . . . . . . . . . . . 77Synfuels . . . . . . . . . . . . . . . . . . . . . . . . . . 77Stationary Uses of Petroleum . . . . . . . . 78Conclusion . . . . . . . . . . . . . . . . . . . . . . . 78

Page

How Do the Environmental Impactsof Increased Automobile FuelEfficiency and Synfuels Compare? . . . 83

Public Health and Safety . . . . . . . . . . . . 84Worker Effects . . . . . . . . . . . . . . . . . . . . 84Ecosystem Effects . . . . . . . . . . . . . . . . . . 85Summary . . . . . . . . . . . . . . . . . . . . . . . . . 85

How Will the Social Impacts ofSynfuels and Increased AutomobileFuel Efficiency Compare? . . . . . . . . . . 86

Employment . . . . . . . . . . . . . . . . . . . . . . 86Community Impacts . . . . . . . . . . . . . . . . 86

How Do the Regional and NationalEconomic Impacts of Synfuels andIncreased Auto Efficiency Compare? . 87

Inflation and Economic Stability . . . . . . 87Employment and International

Competition . . . . . . . . . . . . . . . . . . . . . 87Income Distributing Among Regions . . . 88Capital Intensity and Ownership

Concentration . . . . . . . . . . . . . . . . . . . 88

PageWhat Do Incentives for Increased Fuel

Economy Imply for the Evolutionand Health of the U.S. AutoIndustry ?. . . . . . . . . . . . . . . . . . . . . . . . 89

Can We Have a 75-Mpg Car?. . . . . . . . . . 90

Are Small Cars Less Safe than LargeCars?. . . . . . . . . . . . . . . . . . . . . . . . . . . 92

How Strong Is Current Demand forFuel Efficiency in curs?... . . . . . . . . . 93

What Are the Prospects for ElectricVehicles? . . . . . . . . . . . . . . . . . . . . . . . 94

lf a Large-Scale Synfuels Industry lsBui l t . . . Will Public and WorkerHealth and Safety as Well as theEnvironment be AdequatelyProtected? . . . . . . . . . . . . . . . . . . . . . . 95

Will Water Supply Constrain SynfuelsDevelopment? . . . . . . . . . . . . . . . . . . . 98

Are Synthetic Fuels Compatible WithExisting End Uses? . . . . . . . . . . . . . . . . 99

Alcohols . . . . . . . . . . . . . . . . . . . . . . . . . 99Shale Oil . . . . . . . . . . . . . . . . . . . . . . . . . 99Direct Coal Liquids . . . . . . . . . . . . . . . . . 99Indirect Coal Liquids . . . . . . . . . . . . . . . 100Caveat .. . . . . . . . . . . . . . . . . . . . . . . . 100

TABLES

Table P/o, Page

9. Minimum Oil lmports for Base Case . 7010. Contributions to the Reduction of

Oil imports Beyond the Base Case . . 7011. Estimated Investment Costs for

Conventional Oil and NaturalExploration and Development . . . . . . 75

12. Capital Investment Allocated to FuelEfficiency Plus AssociatedDevelopment Costs . . . . . . . . . . . . . . . 76

Table No.13.

14.

15.

16.

17.

18.

19.

Investment Cost for VariousPage

Transportation Synfuels and ElectricVehicles . . . . . . . . . . . . . . . . . . . . . . . . 77Estimated investment Cost of FuelSwitching and Conservation inStationary Petroleum Uses Duringthe 1990’s. . . . . . . . . . . . . . . . . . . . . . . 78Plausible Consumer Costs forincreased Automobile Fuel EfficiencyUsing Alternative Assumptions AboutVariable Cost Increase. . . . . . . . . . . . . 80Estimated Consumer Cost ofVarious Fossil Synfuels UsingTwo Financing Schemes . . . . . . . . . . . 81Comparison of Average and HighestFuel Efficiency for 1981 ModelGasoline-Fueled Cars in EachSize Class . . . . . . . . . . . . . . . . . . . . . . . 93properties of Selected Jet FuelsDerived From Shale Oil and SRC-11 . . 100Properties of Selected Diesel FuelsDerived From Shale Oil and SRC-ll. . 100

FIGURES

Figure No. Page

4. Comparison of Base Case Oil Imports

5.

and Potential Reductions in TheseImports . . . . . . . . . . . . . . . . . . . . . . . . . 73Consumer Cost of Selected SynfuelsWith Various Aftertax Rates of Returnon lnvestment . . . . . . . . . . . . . . . . . . . . 81

A.

B.c.

BOXESPage

Consumer Cost of lncreasedAutomobile Fuel Efficiency . . . . . . . . . . 79Consumer Cost of Synfuels. . . . . . . . . . 81Total Cost and lnvestment Cost . . . . . . 82

Chapter 4

Issues and Findings

INTRODUCTION

This chapter is a summary and comparison ofmajor results from the analyses discussed later inthe report. It also contains additional analyseswhere needed to put the results in perspective.

It begins with a discussion of the true cost ofimported oil. Increased automobile fuel efficien-cy, synfuels, and conservation and fuel switchingin stationary petroleum uses are then comparedaccording to the speed with which they can actto reduce oil imports and their respective invest-ment costs. Increased auto fuel efficiency andsynfuels are compared according to their environ-mental, social, and economic impacts. Estimatedconsumer costs for increased automobile fuel effi-ciency and synfuels are also given in separateboxes, but the uncertainties are too large for anymeaningfuI comparison. In addition, there is abox discussing the uncertainties in total consumercosts for each of the oil displacement options.

Following the comparisons, several issues re-lated specifically to increased automobile fuel effi-ciency or to synfuels are covered. For automo-biles, the issues include the effects of incentivesfor increased fuel efficiency on the evolution andhealth of the U.S. auto industry, the possibilitiesfor a highly fuel-efficient car, the safety of smallcars, current demand for fuel efficiency i n cars,and the prospects for electric vehicles. For syn-fuels, probable environmental dangers, waterconstraints, and compatibility of synfuels with ex-isting end uses are considered.

Each separate entry in this chapter is designedto stand alone and generally does not build onother material in the chapter. The chapter is notdesigned to be read from beginning to end; rath-er, each reader can turn directly to those compar-isons and issues of interest without loss of con-text or regard for the way the entries are ordered.

WHAT DO OIL IMPORTS COST?The private U.S. consumer pays the going mar-

ket price for imported oil, but that is not its onlyeconomic cost. In the last decade, the Nation hasbeen forced to pay a substantial additional “pre-mium” because of its strategic dependence ona small number of foreign oil producers. Duringespecially unstable periods, such as the 1973-74Middle East War and the 1978-79 Iranian Revolu-tion, this import premium payment is highly visi-ble and, when measured in terms of the incre-mental cost for that segment of demand whichclearly exceeds available supplies, it can greatlyexceed the actual market price.

It is reasonable to attribute an exceptional pre-mium payment to oil, and not to other importedgoods and services, because uninterrupted oilsupplies are critical to economic stability (i.e., fewsubstitutes exist at least in the short run) andbecause the United States has become the prom-inent importer on the world scene and has as-

sumed major responsibility for protecting worldoil trade. No other import constitutes such a vitaleconomic resource that must flow in such a largecontinuous stream around the world. Althoughthe third quarter of 1981 has witnessed falling oilprices and a modest supply surplus, future short-age risks remain plausible because of the ex-pected longrun depletion of world oil reservesand because of unresolved and potential inter-national conflicts.

The existence of a national premium paymentfor oil imports can be explained in terms of threeeconomic relationships:

1.

2.

the dependence of international price on thequantity of U.S. imports;the loss of U.S. jobs and gross national prod-uct (GNP) caused by oil payments abroadand the associated depreciation of the dollar;and

67

68 ● Increased Autornobile Fuel Efficiency and Synthetic Fuels: Alternatives for Reducing Oil Imports

3. the budgetary cost of military outlays and for-eign military assistance related to assuringthe security of oil imports. These are de-scribed below.

Dependence of Price onQuantity Imported

Market price is a good measure of real or totalcost when markets are competitive and in a stateof stable equilibrium. Neither situation is charac-teristic of international oil markets, which aredominated by a small number of sellers and buy-ers in an unstable marriage of short-term conven-ience. Despite the complexity and unpredictabil-ity of this relationship, it seems reasonably clearthat raising U.S. oil imports drives price upwardand vice versa, simply because any movementby such a prominent importer appears to the restof the world as a shift in the world demand curve.This positive relationship between quantity im-ported and price means that the cost of incremen-tal U.S. consumption exceeds current pricebecause the increment makes all future consump-tion more expensive. Conversely, decrements inU.S. consumption save more money than themarginal reduction in purchases.

Eventually, oil markets may anticipate thisprice/quantity relationship, but market adjust-ments may not be smooth. Shocks can be ex-pected, leading to domestic inflation and reces-sion, because international relationships betweenexporters and importers have become politicizedand because significant reductions in oil con-sumption are difficult to achieve over periods ofup to several years due to the long lifetimes ofenergy-related capital stock and the long lead-time for alternative domestic fuels.

Loss of U.S. Jobs and GNP Causedby Rising Oil Payments and byPotential Supply Interruptions

Oil imports accounted for 26 percent of U.S.payments for imports in 1979, which is abouttwice the level of the second largest item. Con-sequently, compared with equivalent rates ofgrowth or decline for other imports, changes overtime in oil payments have a relatively large im-

pact on the U.S. balance of trade, and, hence,a relatively large impact on the exchange valueof the dollar.

In periods when the dollar is relatively strong,as it has been recently (second half of 1981), itis due in part to declining oil payments. In periodswhen the dollar is weak, as it was during mostof the 1970’s and especially after 1975 becauseof large deficits in merchandise trade, growingoil payments increase selling pressure on the dol-lar, lowering its foreign exchange value. Whilethis makes U.S. exports more attractive to foreignbuyers, export sales may not increase elasticallybecause of stagnant world economy or failure ofU.S. goods to meet quality standards. Therefore,market adjustments, including both higher pricesand undoubtedly reduced purchases, are forcedon U.S. importers.

Furthermore, the declining value of the dolIarhas relatively little effect on oil imports, again dueto the long lifetimes of capital related to oil con-sumption. Barring economic recession, oil con-sumption significantly declines only with the slowreplacement of capital. Thus, even though risingoil imports or sharply rising oil import prices maybe clearly responsible for dollar depreciation, oilconsumption may not bear the brunt of the re-sulting short-term adjustment.

Overall, adjustments in the U.S. balance of pay-ments also affect domestic economic activity. Asharply rising oil import price directly increasesdomestic inflation while at the same time largerforeign payments can lower total demand for do-mestic goods and services if, as is likely in theshort run, oil exporters do not spend their largerreceipts in the United States. This combinationof rising inflation and declining total demand putsthe Federal Government in a difficult position be-cause corrective policies are contradictory. If con-trol of inflation is the primary objective, sharp oilprice increases may force the Government tobrake the growth momentum of the nationaleconomy or exaggerate downward cycles in or-der to limit propagation of inflationary pressures.

In addition to oil price shocks, potential sup-ply interruptions of oil imports present the clear-est, most direct threat to national economic activ-ity. As discussed above, few good substitutes exist

Ch. 4—issues and Findings ● 6 9

for oil in the short run, so that reduced flow re-sults in lost production and unemployment assoon as stockpiles can no longer make up for thedeficit.

The potential premium payment, implied byboth unstable oil import prices and supply inter-ruptions, can be illustrated in terms of the 1973-74 shock. In 1974 and again in 1975, real GNPdeclined by more than a percentage point afterhaving grown at a rate of 5 percent in 1973 and4 percent in 1972. Although cause and effect inmacroeconomics is highly speculative, the lossesin 1974 and 1975 are widely believed to havebeen due in part to the disruption of oil suppliesand the associated quadrupling of imported oilprices. If, in fact, real GNP growth had been re-duced by just one percentage point by oil-relatedevents, it would have meant a loss of about $15billion in U.S. production ($1.5 trillion GNP in1975), which amounts to $6.80 per barrel (bbl)for the 2.2 billion bbl imported that year. Theprice of oil at that time was about $11.

Military Outlays and Foreign PolicyDirections Forced by Oil

Import Dependence

Military and foreign policy are predicated onmany national objectives, but apparently onevery important consideration for the United Statesis protection of oil supply lines. The cost of suchprotection cannot be ascertained directly, butcurrent debate over defense budget prioritiesindicates that the United States intends to developweapons systems and train personnel in order tobe able to fight a war in the Middle East, if nec-essary.

If 10 percent of estimated 1982 defense outlayswere justified to meet military threats to MiddleEast oil supplies, it amounts to about $18 billionor about $9/bbl for the 2 billion bbl of oil im-ported (net of exports) in 1981.

Conclusion

The complexity and unpredictability of worldoil markets and world oil politics make it difficultto predict the oil import premium over time. * InOTA’s judgment, the possible future import pre-mium could range up to $50/bbl. It could be neg-ligible if world demand continues its sharp down-ward trend and if major new discoveries aremade outside the Middle East, but could be muchlarger than the current price of oil if hostilitiesbreak out which cut off most supplies from theMiddle East.

A technical analysis must stop short of greatercertainty except to indicate that a significant re-duction of imports would drive the premiumdown by reducing the visibility of the UnitedStates in world oil markets and by reducing U.S.dependence on supplies from politically unstablecountries. In other words, the premium paymentfor the last barrel of imports is much higher thanfor the first, and it is the last barrel which wouldbe displaced by domestic synfuels or by higherfuel efficiency in automobiles.

*A number of estimates for both components of the oil importpremium are available. For the most detailed discussion of relatedeconomic issues and documentation of results from current eco-nomic models, see VVor/cf Oil, Energy Modeling Forum, StanfordtJniversity, Stanford, Calif., ch. 5 (forthcoming).

HOW QUICKLY CAN OIL IMPORTS BE REDUCED?Options for reducing U.S. oil consumption are Table 9 shows the estimated level of imports

considered in detail later in the report. Here, the in the absence of synthetic fuels and automobileresults of those analyses are summarized and the fuel-efficiency increases beyond a 1985 level ofrelative contributions that the various options can 30 mpg. This base case also assumes: 1) the En-make to reducing imports over the next two dec- ergy Information Administration’s (EIA) high oilades are considered. price future to 1990 for the consumption of oil


Table 9.-Minimum Oil Importsfor Base Case (MM B/DOE)

1980 1985 1990 1995 2000

Stationary demanda (noadditional measurespast 1990) . . . . . . . . . . . . 8.1 7.3 6.4 6.4 6.4

Transportation demand(other thanautomobiles . . . . . . . . . . 4.5 4.7 5..0 5.4 5.7

Automobiles (with1985 new-car averageof 30 mpg, no changethereafter) . . . . . . . . . . . . 4.3 3.6 3.0 2.7 2.7

Sum of demand . . . . . . . . . 16.9 15.6 14.4 14.5 14.8Domestic production . . . . 10.2 8.6 7.6 7.1 7.0Imports. . . . . . . . . . . . . . . . . 6.7 7.0 6.8 7.4 7.8%cludes all nonfuei oil uses such as asphalt, petrochemical feedstock, liquefiedpetroleum gas, etc., which are projected by the Energy Information Administra-tion to total 3.8 MMB/D by 1990,

boil PIUS natural gas liquids.

SOURCE: Office of Technology Assessment,

for stationary uses;1 2) the transportation petro-leum demand (other than for passenger cars) ex-plained in chapter 5; 3) fuel oil demand by sta-tionary uses is held constant after 1990; and 4)the maximum domestic oil production projectedby OTA.2 For 1995 and 2000, the trends of the1980’s for stationary uses of petroleum other thanfuel oil have been extrapolated, while holdingfuel oil consumption constant at the projected1990 level. The assumption of constant fuel oildemand for the 1990’s was chosen as the basecase to help illustrate the importance of eliminat-ing this demand relative to other options for re-ducing oil imports in the 1990’s.

It should be emphasized that a considerablereduction in oil consumption through increasedefficiency and fuel switching in the 1980’s is al-ready built into the base case. in particular,achieving an average new-car fuel efficiency of30 mpg by 1985 saves about 0.8 million barrelsper day oil equivalent (MMB/DOE) by 1990, rela-tive to 1980 demand;* and conservation and fuelswitching in the EIA high oil price scenario reduceoil consumption by 1.7 million barrels per day

1 Energy Information Administration, U .S. Depatiment of Errergy.z wOr/d petfo/eurn Availability: 19802~ Technical Memoran-

dum, OTA-TM-E-5 (Washington, D. C.: U.S. Congress, Office ofTechnology Assessment, October 1980).

*The fuel saved in cars is 0.9 million barrels per day (MMB/D),but the assumed increase in transportation needs raises consump-tion in other types of transportation by 0.1 MMB/D.

(MMB/D) in stationary uses by 1990. However,domestic oil production is likely to drop by atleast 2.6 MMB/D during this same time period,3

thereby nullifying any reduction in oil importsfrom these measures alone.

Table 10 shows the various reductions in oilconsumption that may be achieved beyond thebase case. These include contributions from fur-ther conservation and fuel switching in stationaryuses, increased automobile fuel efficiency beyonda 1985 level of 30 mpg, electric vehicles (EVs),and synfuels. Each of the areas where additionaloil savings are possible is discussed below.

By 1990, stationary demand for residual anddistillate fuel oil is 2.6 MMB/D in the base case. *As explained in chapter 7, a combination of cost-effective conservation measures and switching tonatural gas and electricity can eliminate this sta-tionary fuel oil demand without a need to in-crease gas production or electric generating ca-pacity. How much of this potential actually isreached will depend on such things as individualdecisions about conservation investments and

3 World Petroleum Availability: 19802-Technical Memoran-

dum, op. cit.*The remainder of the 6.4 MM B/D of stationary oil use includes

asphalt, petrochemical feedstocks, liquefied petroleum gas, and re-finery still gas.

Table 10.—Contributions to the Reduction of OilImports Beyond the Base Case (MMB/DOE)

1980 1985 1990 1995 2000Conservation and

switching instationaryapplications . . . . . 0 0 0 0.8a-1.3 l.5a-2.6

Increased automobilefuel efficiencybeyond 1985average of 30mpg . . . . . . . . . . . . 0 0-0.1 0.1-0.5 0.3-1.0 0.6-1.3

(Average new-carefficiency,mpg b). . . . . . . . . (23) (30-37) (36-49) (40-63) (45-79)

Electric vehicles . . . 0 0 0 0 0-0.1Synthetic transpor-

tation fuels:Fossil . . . . . . . . 0 0-0.1 0.3-0.7 0.7-1,9 1.3-4.5Biomass . . . . . . 0 (c) 0-0.3 0-0,6 0,1-1.0. — — — —

Total . . . . . . . . . . . 0 0-0.20 .4-1.5 1.8-4.8 3.5-9.5aEnergY Information Administration forecast.bss percerlt EPA citylds percent EPA highway test wcles.CLeSS tharl 0.05 MMBID.SOURCE: Office of Technology Assessment.


availability of transmission and distribution sys-tems. The most recent projection of EIA providesa reasonable lower bound on the reduction infuel oil that can be achieved during the 1990’s.

The range of potential savings from increasedautomobile fuel efficiency corresponds to the lowand high estimates derived in chapter 5 and dif-ferent assumptions about relative future demandfor small-, medium-, and large-sized cars, i.e.,1) no shift to smaller cars and pessimistic assump-tions about efficiency increases from automotivetechnologies, and 2) a substantial shift to smallercars and optimistic assumptions about the tech-nologies. By 2010, automobiles containing theaverage technology of 2000 would have replacedmost cars on the road and the savings, relativeto the base case, would be 0.8 to 1.7 MMB/D (2.3to 3.2 MMB/D relative to 1980 demand).

It should be emphasized that average new-carfuel efficiencies shown in table 10 do not repre-sent a technical limit to what can be achieved.In any given year, cars with higher (and lower)mileages than those shown would be producedand sold. * Rather, the mileage ranges correspondto what OTA considers to be feasible through avariety of technological improvements.

If a very strong demand for fuel-efficient carsdevelops, e.g., as the result of continued largeoil price increases, consumers may be willing toaccept poorer performance or pay the added costin order to achieve higher fuel efficiency. In thiscase, the estimated average fuel efficiency shownfor 2000 in table 10 could be achieved by themid- 1990’s.

The contribution from EVs was calculated byassuming that O to 5 percent of passenger auto-mobiles would be electric by 2000, growinglinearly from O percent at 1985, The savings fromEVs is relatively small, however, because of therelatively low consumption of petroleum by auto-mobiles (1.3 to 2.1 MMB/DOE in 2000) and the

*For example, u p until January 1981, the 1981 model new-carfuel efficiency of cars sold averaged slightly less than 25 mpg, butif the most fuel-efficient cars in each size class had been bought,the average would have been about 33 mpg.4

4Derived from data in J. A. Foster, J. D. Murrell, and S. L. Loos,“Light Duty Automotive Fuel Economy . . . Trends Through 1981 ,“U.S. Environmental Protection Agency, SAE paper No. 810386, Feb-ruary 1981,

fact that EVs would be a substitute for the mostfuel-efficient cars.

The final category in table 10, synthetic fuels,must be considered carefully to ensure that onlythe synthetic fuels production that displaces oilis included and technical difficulties are ac-counted for. To derive the low estimate of syn-fuels contributions, it is assumed that by the timesynthetic fuels become available, the only re-maining stationary uses of petroleum are forchemical feedstocks, asphalt, petroleum coke,still gas, and liquefied petroleum gas (LPG). Sincethese products cannot now be economically con-verted to transportation fuels, the low estimatein table 10 assumes that their replacement by syn-thetic fuels (synthetic gas) would not result in ad-ditional transportation fuels. * In addition, poorperformance of the first round of synfuel pIantsis assumed, limiting production until the early tomid-1 990’s. As a consequence of this, the lowsynfuels production scenario from chapter 6 isused in the table 10 low estimate.

A more optimistic scenario is possible if it is as-sumed that market or other forces strongly favorthe production of transportation fuels over syn-thetic fuel gases and that half of the syntheticgas* * plants projected in chapter 6 actually arebuilt to produce synthetic transportation fuels.With these assumptions and the high scenariospresented in chapter 6, one arrives at the upperestimate for oil displacement by synfuels shownin table 10. The high estimate, however, repre-sents a vigorous dedication to synfuels produc-tion and what might be termed near “war mobil-ization” development of the industry.

The range of oil savings from each of thesesources is shown in figure 4, alongside the im-port levels calculated in the base case. As canbe seen, under the most favorable circumstancesit is technically possible to eliminate oil importsby 2000. However, if domestic oil productions

is below that shown in table 9 and if only the lowestimates of table 10—or even only the low esti-

*LPG can, however, be used directly in appropriately modifiedautomobiles.

**Excluding biogas from manure, which would be used principal-ly on the farms where it is produced.

5 World Petroleum Availability: 1980-2000- Technical Memoran-dum, op. cit.

98-281 0 - 82 - 6 : QL 3


Photo credit: Paraho Development Corp.

The Paraho Semiworks Oil Shale Unit at Anvil Points, Colo.


Figure 4.—Comparison of Base Case Oil Imports and Potential Reductions in These Imports

10 —

9 —

8 -

7 -

6 —

5 —

4 “

3 -

2 —

1

0

1980 1985 1990 1995

B

2000Year

O i l i m p o r t . i n b a s e c a s e

Increased efficiency and fuel switching in stationary uses Fossil and biomass synthetic fuels

= Electric vehicles

Increased automobile fuel efficiency (relative to 1985 A = Low scenarios as outlined in textaverage of 30 mpg)

SOURCE Office of Technology Assessment.

mate for synfuels—are reached, then it is quiteunlikely that oil imports can be eliminated beforesometime well into the first decade of the 21stcentury.

Although the large number of noteworthy un-certainties make an exact determination of thecourse of oil imports impossible, several conclu-sions can be drawn.

First, increased efficiency and fuel switching inbuildings and industry are extremely importantfor the reduction of oil consumption. Althoughmuch of the potential in this area will be achievedthrough market forces by 2000 under the highoil price scenario of EIA, implementing the nec-essary changes at an earlier date could significant-ly reduce oil imports before 2000. For example,

B = High scenarios as outlined in text

fully implementing the potential for reducing sta-tionary uses of fuel oil by 1990 would save about15 billion bbl of oil imports or $600 billion (atan average of $40/bbl) for the imports during theperiod 1981-2000.

Second, synthetic fuels development has ap-proximately the same importance as the conser-vation and fuel switching options but its contribu-tion to reduced imports will not be as large untilat least the late 1990’s. Further, if a large part ofthe synfuels is used as a substitute for increasedefficiency and for conventional fuel switching instationary uses or as a substitute for petroleumproducts not readily converted to transportationfuels, elimination of oil imports is likely to bedelayed.


Third, increases in automobile fuel efficiencybeyond a 1985 average of 30 mpg could reduceautomobile fuel consumption 20 to sO percent(0.6 to 1.3 MMB/D) by 2000 below the fuel con-sumption of a 30-mpg fleet. in addition, becausefuel efficiency increases in automobiles (to andbeyond 30 mpg) could reduce the automobile’sshare of transportation fuel needs from so per-cent (in 1980) to 20 to 25 percent (in 2000), itis likely that efficiency increases in various non-automobile transportation uses beyond those as-sumed in the base case could also make signifi-cant contributions to reducing transportation fuel

needs. This option has not been analyzed byOTA.

In summary, it probably will be necessary toimplement fully all of the options for reducingoil consumption if one wants to eliminate net oilimports before the first decade of the next cen-tury. This will require full implementation ofcharges needed for increased efficiency in all usesof oil and fuel switching in stationary uses, as wellas directing synfuels production to transportationfuels.

WHAT ARE THE INVESTMENT COSTS FORREDUCING U.S. OIL CONSUMPTION?

Introduction

investment costs are an important considera-tion when comparing alternatives for reducingU.S. oil consumption. OTA’s analysis indicatesthat synfuels production, increased fuel efficiencyin automobiles, and conservation and fuel switch-ing in stationary uses of oil all will require invest-ments of the same order of magnitude for com-parable reductions in oil. consumption in the1990’s; whereas, synfuels production appears torequire larger investments than the other alterna-tives for the 1980’s. Uncertainties in the cost esti-mates as well as the fundamental differences inthe nature of the investments are too large, how-ever, to allow a choice between approaches onthis basis alone.

In order to compare investment costs, theyhave been expressed as the investment neededto either produce or save 1 barrel per day oilequivalent* of petroleum products. This methodwas chosen in order to avoid problems that arisewhen comparing investments in projects with dif-ferent lifetimes and for which future oil savingsmay be discounted at different rates.** In addi-tion, from a national perspective the per unit in-

*One barrel of oil equivalent = 5.9 MMBtu.**It does not, however, avoid the problem that the different par-

ties making the investments will have fundamentally different con-straints on and perspectives about these investments and thus willreact quite differently in the face of investments of the same size.

vestment cost is important in that it is the param-eter used in the aggregate to make choicesamong competing investments. Conventional oiland gas exploration are considered first to pro-vide a reference point. Following this, OTA’s esti-mates for the investment costs for increased auto-mobile fuel efficiency, EVs, synfuels, and in-creased efficiency and fuel switching in stationaryuses are discussed briefly.

Conventional Oil and Gas Production

Two estimates of recent investment costs forconventional oil and gas exploration anddevelopment in the United States are shown intable 11. The data in this table were developedfrom estimates of the annual investments in oil,gas, and natural gas liquids exploration and devel-opment per barrel of increased proven reservesof these fuels (corrected for depletion). These lat-ter estimates were then converted to investmentsfor an increase of 1 barrel per day (bbl/d) of pro-duction (corrected for depletion) using the 1980ratio of crude oil reserves to crude-oil produc-tion and assuming an 8 percent refining loss. Theratio of reserves to production for natural gas wasnot used because price controls on natural gastend to inflate this ratio and thus the estimatedcosts; and investments for oil exploration and de-velopment were not separated from those for nat-ural gas because there is no practical way to doso.

Ch. 4—issues and Findings Ž 75

Table 11 .—Estimated Investment Costsfor Conventional Oil and Natural

Exploration and Development

Estimated investment cost(thousand 1980 dollars perbarrel per day of petroleum

production )

Year Estimate Ab Estimate Bc

1974 . . . . . . . . . . . . . . . . . . . . . . 13 151975 . . . . . . . . . . . . . . . . . . . . . . 17 191976 . . . . . . . . . . . . . . . . . . . . . . 20 171977 . . . . . . . . . . . . . . . . . . . . . . 22 201978. . . . . . . . . . . . . . . . . . . . . . 29 181979. . . . . . . . . . . . . . . . . . . . . . 31 57d

1980. . . . . . . . . . . . . . . . . . . . . . Not available 39”Extrapolated to 1985f . . . . . . . 53 49“ ASSUflleS&perCent refirlingloss”rld a1980ratio ofcrude Oil reServe9tOPrO-

ductionof3.07 x 10’ barrels ofreserves ~erbarrel perday production. lfEIAdata for petroleum resemes are used, the figures are increased byabout 10percent.

b lnvestrnen!cos!perbarrel of increased rese~es from A. T. Guernsey, ”Econom-Ics of Domestic Crude 011 and Natural Gas Exploration and Development1959-1976,” December 1977, and “1977 and 1976 Addendum,” June 1979,Prepared for Exploration and Production Department, Shell Oil Co., Houston,Tex.; and W. C, Hamber, Manager, Forecasting, Exploration and ProductionEconomics, Shell 011 Co., Houston, Tex., private communication, Nov. 11,1981

c Investment cost ~r barrel of ~ncreased resewes fOr the 26 major ener9Y corn.panles in the United States calculated for OTA by John Rasmussen, Economicsand Statistics, Energy Markets and End Use, EIA, October 1981, based on EIAand American Petroleum Institute data See also “Pedormance Profiles of MajorEnergy Producers 1979,” EIA, U.S. Department of Energy, DOE/EIA-0206(79),July 1981.

d This estimate iS anomalously high due to downward revision Of estimated re-serves by Texaco during the year and because Ashland Oil sold aome of itscrude oil reserves to a company not included in a sample of 26 ma)or energycompanies.

e Thia estimate may be IOW because petroleum reserve additions are overstateddue to the purchase of Texas Pacific Oil & Gas (not one of the 26 major com-panies included in the calculation) by Sun 011 Co. (one of the 26 major energycompanies Included in the calculation).

f B~ed on least squares fit of 1974-80 data, exclusive Of 1979 data in estimateB. Correlation coefficient la 0.98!5 for estimate A and 0.82 for estimate B.


There are significant uncertainties in these esti-mates due to numerous anomalies in the data,some of which are detailed in footnotes to table11, and because the ratio of reserves to produc-tion changes with market prices, production tech-niques (e. g., enhanced oil recovery), and the na-ture and quantity of reserves. Nevertheless, thesedata do indicate that it is reasonable to expectcosts of $50,000/bbl/d or more for conventionalpetroleum exploration and development by themid-1 980’s if recent cost trends continue.

Automobile Fuel Efficiency

OTA’s estimates of the investment plus associ-ated product development costs for increasedautomobile fuel efficiency are shown in table 12.

There are notable technical, accounting, andmarket uncertainties associated with this type ofcost analysis, however.

The estimates in table 12 were derived by firstestimating the efficiency gains that can reason-ably be expected over time from various changesin the automobile system. They are based on bothpublished estimates and OTA’s analysis. The ratesat which these technologies may be incorporatedinto new cars were then estimated and resultantschedules for capital turnover derived. Next, theinvestment cost calculations were based on pub-lished estimates for the cost of replacing the ap-plicable capital equipment (e.g., facilities for pro-ducing a new engine or transmission, etc.). Theactual investment cost and resultant fuel efficien-cy increases, however, will depend on a numberof factors specific to individual production plants(and their future evolution), the way various pro-duction tradeoffs are resolved, and the results offuture product development programs.

In addition to capital investment, developmentcosts have been included as part of the invest-ment necessary to produce modified vehicles.During the 1970’s, domestic auto manufacturers’R&D (mostly development) costs averaged from40 to 60 percent of their capital investments.6 Intable 12, development costs are assumed to be40 percent of the capital investment allocated tofuel efficiency (see below), but the actual costsof developing the technologies for producingmore efficient cars at minimum cost are highlyuncertain. *

Beyond the uncertainties in the investment anddevelopment costs, there is the problem of deter-mining what fraction of the investments shouldbe ascribed to fuel efficiency. This arises becausesome of the investments can be used not onlyto increase fuel efficiency, but also to make other

6G. Ku[p, D. D. Shonka, and M. C. Halcomb, “TransportationEnergy Conservation Data Book: Edition 5,” oak Ridge NationalLaboratory, ORNL-5765, November 1981.

*lt should be noted that R&D costs are not included for synfuelsbecause several essentially identical synfuels plants could be con-structed with little additional R&D costs beyond those needed forthe first plant, whereas product and process development are nec-essary for each major change in automobiles.


Table 12.—Capital Investment Allocated to Fuel Efficiency Plus Associated Development Costs

Average capital investment plusassociated development costsb

Thousand 1980 dollars perbarrel per day oil

New-car fuel efficiency at equivalent ofTime of investment Mix shift end of time perioda (mpg) fuel savedc 1980 dollars per car producedd

1985-1990 . . . . . . . . . Moderatee 38-48 20-60 g 50-1909Largef 43-53

1990 -1995 . . . . . . . . . Moderatee 43-59 60-1309 70-1809Largef 49-65

1995-2000 . . . . . . . . . Moderatee 51-70 50-1509 50-1509Largef 58-78

a EpA rated 5w45 percent city/highway fuel efficiency of avera9e new car.b Development costs assumed to be 40 percent of capital investment allocated

to fuel efficiency (see text). One barrel of oil equivalent contains 5.9 MMBtu.c Averages are calculated by dividing average investment fOr technological im-

provements by fuel savings for average car at end of time period relative toaverage car at beginning of time period. The resultant average cost per barrelper day is lower than a straight average of the investments for each car sizebecause of mathematical differences in the methodology (i.e., average of ratiosv. ratio of averages) and because extra fuel is saved due to demand shift tosmaller cars. The averaging methodology used is more appropriate for compari-sons with synfuels because it relates aggregate Investments to aggregate fuelsavings. It should be noted that the cost of adjusting to the shift in demandto smaller-sized cars is not included. Only those investments which increasethe fuel efficiency of a given-size car are included.

d Assuming lnve9tment iS used to produce cars fOr 10 Years, on the avera9e.e Moderate shift In demand to smaller cars. Percentage Of new carS sold in each

size clasa are:


changes in the car. * The cost allocation problemassociated with multipurpose investments is wellknown in accounting theory, and there is no fullysatisfactory solution to it.7

For table 12, it was assumed that 50 percentof the cost of engine and body redesign, 75 per-cent of the cost of most transmission changes,and 100 percent of the cost of advanced materialssubstitution and energy storage and automaticengine cutoff devices should be allocated to fuelefficiency. This results in between 55 and 80 per-cent of the investments being allocated to fuelefficiency, depending on the time period and sce-nario chosen. For further details on how this andother problems in estimating the cost of fuel effi-ciency were resolved, see chapter 5.

*For example, automobile designs with low aerodynamic dragmay be preferred by consumers on esthetic grounds; front wheeldrive may be introduced to improve traction and increase interiorvolume; microprocessor control of carburetion or fuel injection,spark advance, exhaust gas recirculation, and other operating condi-tions can be used to reduce exhaust emissions, improve perform-ance, and enable the use of lower octane fuels; continuously vari-able transmissions may be introduced to produce smoother acceler-ation and improve performance. These and many other changescan also be exploited to improve fuel efficiency.

7A. L. Thomas, “The Allocation Problem in Financial Account-ing Theory, ” American Accounting Association, Sarasota, Fla., 1969,pp. 41-57, and A. L. Thomas, “The Allocation Problem: Part Two, ”American Accounting Association, Sarasota, Fla., 1974.

Year/size class Large Medium Small1985 35 60 51990 25 60 151995 20 55 25

35fLarge shift in demand to smaller cars. Percentage of new cars sold in each

size class are:Year/size class Large Medium Small

1965 15 75 101990 5 65 301995 5 45 502000 5 25 70

gWithin uncertainties, the costs are the same for both mix shifts.

During the period 1985-2000, total capital in-vestments in changes associated with increasingfuel efficiency (i.e., allocating 100 percent of themultipurpose investments to fuel efficiency) couldaverage $2 billion to $5 billion per year, depend-ing on the number of new cars sold and the rateat which fuel efficiency is increased. However,if one deducts the cost of changes that wouldhave been made under “normal” circum-stances, * the added capital investment neededto achieve the lower mpg numbers in table 12would be $0.3 billion to $0.7 billion per year. Thehigher mpg numbers in table 12 would requireadded capital investments (above “normal”) of$0.6 billion to $1.5 billion per year. Adding 40percent of the capital investment for developmentcosts results in added outlays of $0.4 billion to$0.9 billion per year and $0.8 billion to $2 billionper year for the low and high scenarios, respec-tively.

A detailed examination of the scenarios pre-sented in chapter 5 shows that a 1990 new-car

*Assuming “normal” capital turnover is: engines improved after6 years, on average, redesigned after 12 years; transmissions sameas engines; body redesigned every 7.5 years; no advanced materialssubstitution.

Ch. 4—issues and Findings . 77

average fuel efficiency of 35 to 45 mpg (depend-ing on the proportion of small, medium, and largecars sold) probably can be achieved with whatis termed here “normal” rates of capital turnover.However, the validity of this conclusion and ofthe above incremental investment and develop-ment cost estimates will depend on market de-mand for fuel efficiency, and, in OTA’s judgment,there is no credible way to predict future marketdemand for fuel efficiency.

Electric Vehicles

Use of EVs more nearly approximates synfuelsthan increased automobile fuel efficiency, in thatEVs involve switching from conventional oil toanother energy source rather than reducing en-ergy consumption. Consequently, the costs (perbarrel per day of oil replaced) for EVs are in-cluded in table 13 with synfuels. As shown intable 13, the costs for EVs appear to be significant-ly higher than for the various synfuels options,due to the high purchase price of the vehicle (rel-ative to a comparable gasoline-fueled car) andthe fact that EVs would be replacements for rel-atively fuel-efficient cars (because of an EVs lim-ited size and acceleration). Furthermore, if bat-

teries must be replaced at regular intervals andthe cost of this is included as an investment cost,the total investment per barrel per day rises dra-matically.

Synfuels

The best available estimates for the investmentcosts for various liquid synthetic transportationfuels are shown in table 13. Because of uncertain-ties in the cost estimates, no meaningful inter-comparison among synfuels on the basis of costis currently possible. I n addition, as discussed inchapter 6, the final investment in synfuels is like-ly to be different from these estimates. As theprocesses approach commercial production, theywill be revised as costs to overcome problemsencountered in demonstration units are deter-mined. Construction costs will inflate at an un-known rate relative to general inflation. And de-lays during construction due to such possibilitiesas lawsuits, strikes, late delivery of constructionmaterials, or other causes can increase the invest-ment cost. In sum, current investment estimatesprovide a very tentative guide to what synfuelsplants constructed in the 1990’s will cost. In addi-

Table 13.—lnvestment Cost for Various Transportation Synfuels and Electric Vehicles

Thousand 1980 dollars per barrel per day oil equivalent to end usersMethanol Coal to methanol and

Shale oil from coal Mobil methanol to gasoline Direct liquefaction Electric vehicleMining . . . . . . . . . . . . . . . . (Included in 4-15 4-15 4-15 5-19

conversion plant)Conversion plant . . . . . . . 49-73 a 47-93 a 53-110 a 67-100 a 0-69b

Refinery. . . . . . . . . . . . . . . 0-10’ o 4-22d oDistribution system. . . . . 0 O-2e

o 0End use. . . . . . . . . . . . . . . 0 0-11f o 0 320°3909

Total . . . . . . . . . . . . . . . 49-83 51-121 57-125 75-137 325-478aflange of investments in ch. 6 plUS possible 50-percent cost overrun and assuming plants operate at m Wrcent of rat~ capacitybupper limit corresponds t. ca9e where new coal-fired eiectric generating capacity would be naeded. That Is not currently the case, however.Cupper limit corres~nds to a dedicated reflnOV.dupper Iimlt from UC)P and Systems Development corp., “Crude Oil v. Coal Oil Processing Comparison Study,” DOE/ET1031 17, TR-80/ml, November 1979. Infiated

by 12 percent to refiect 1980 cost. Assumes a refinery dedicated to conversion faciiity. Lower iimit from “SRC-ii Demonstration Project, Phase Zero, Task No. 3, MarketAssessment Transportation Fueis From SRC-ii Upgrading,” Pittsburgh and Midway Coai Mining Co., prepared for U.S. Department of Energy, Juiy 31, 1979. Assumesoniy upgrading of iiquid for use as feadstock in an existing refinery.

eupper limit ~sumes that haif of Cap=lty must use newiy constructed or expanded facilities, as foiiows: 500-miiO piPOiine at $1 miiiion Per mile, 500,000 bbild caPacitY;

tank truck (9,200 gai) costing $90,~, 10 runs per week; storage tanks and pumps costing $700/bbi/d of throughput.f UpWr limit ~aume~ new engine design costing $540 miiilon for a ~,~ car per year f~tory; 0.15 capitai recovery factor; repiacing car consuming 250 gai of gasoiine/yr.

Beyond the initial investment in new engine production facilities, the Investments are the same as for a gaaoiine engine, making the added investment zero, reiativeto gasoline vehicies.

gAssumes an eiectric vehicie costs $3,000 more than a comparably performing gasoiine-powerw car and the eiectric vehlcie repiaces 8,000 to 10,MI miles/yr thatwouid have been driven in a 60-mpg gasoline- or diesei-fueied car. if batteries must be replaced every 10,000 miles (nine timea over iife of cad at a cost of $2,000each time, the totai investment becomes $2.7 miiiion/bbi/d repiaced. These calculations assume that no oii is used in the eiectric generating faciiitles; however, ifoii is used to generate part of the electricity, the investment costs per barrei per day of oil dispiaced grow rapidly,

SOURCE: Office of Technology Assessment,


tion, they most likely represent a lower limit ofthe synfuels investment costs. *

Stationary Uses of Petroleum

OTA has also considered the costs of conser-vation and fuel switching in stationary uses of oil,although not in the same detail as for synfuelsand increased automobile fuel efficiency. Themajor candidates are the residual and distillatefuel oils still used in stationary applications after1990. Other stationary uses of petroleum—as-phalt, petrochemical feedstock, still gas, and liq-uefied petroleum gas (LPG)—were not consideredto be major potential supplies of increased trans-portation fuels. Although LPG can technically beused as a transportation fuel, and petrochemicalfeedstocks can be replaced by synthesis gas fromcoal, a preliminary analysis indicates that the fueloils are more economically attractive alternativesfor increasing supplies of transportation fuels inmost cases.

OTA’s estimates for the investment costs of fuelswitching and increased energy efficiency in sta-tionary uses during the 1990’s are shown in table14. Although only single numbers are shown, infact there will be a range of costs depending ondevelopment costs for new energy supplies, in-stallation costs for end-use equipment, the extentof changes needed at oil refineries, and variationsin conservation investments. Of the fuel switchingoptions the range is narrowest for fuel switchingto electricity because of the fairly well-defined

“Decisions about whether and how quickly to proceed with in-vestments in synfuels production, however, will be strongly influ-enced not only by estimated costs but by corporate strategy.

cost of producing electricity from coal and largestfor fuel switching to natural gas because of differ-ences in the cost of developing various uncon-ventional gas supplies.

In deriving the numbers in table 14, it was as-sumed that the lower cost opportunities for fuelswitching and conservation would already havebeen carried out by 1990. To the extent that thisdoes not occur, the per-unit investment cost esti-mates for the 1990’s wouId be lowered some-what. Also, increased end-use efficiency of elec-tricity for heat and hot water would reduce theinvestment needed for electric powerplants; andif large supplies of relatively inexpensive gas arefound, fuel switching to gas could be a very at-tractive option, in terms of capital investment. Be-cause of these uncertainties and site-specific dif-ferences in installation costs, one cannot clearlychoose among the alternatives on the basis of in-vestment costs alone. All of the options to elimi-nate stationary fuel oil use seem to require thesame order of magnitude of investments.

Conclusion

Three principal conclusions emerge fromOTA’s analysis of investment costs for the variousways of reducing oil consumption. First, there isa great deal of uncertainty about investment costsdue to technological unknowns, lack of experi-ence, and site-specific cost differences. * Second,

*The situation is further complicated by the different nature ofthe investments. Synfuel plant construction requires large invest-ments over a number of years before any product is sold. Auto in-dustries tend to make incremental changes in capital stock, withthe sum of several such investments sometimes costing more thanone abrupt changeover in capital stock. Investments in fuel switch-ing and conservation are paid back through future fuel cost sav-ings rather than product sales.

Table 14.—Estimated Investment Cost of Fuel Switching andConsecration in Stationary Petroleum Uses During the 1990’s

Thousand 1980 dollars per barrel per day of oil replaced or saved

Conversion to Conversion to Conversion of boilers from Increased efficiencyInvestment at natural gas electricity residual fuel oil to coal and fuel switching

End use equipment . . . . . . . . . . . . . . . . . . 24 32 37 (51)a 88New production of fuel . . . . . . . . . . . . . . . 90 78b 16C (22)a o

Total . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114 110 53 (74)a 88The number in parenthesis is corrected for the 72-percent efficiency of refining residual fuel oil and is the Investment per barrel per day of resultant distillate oil pro-duced from the residual 011.

btin9truction of coal-fired powerpiant ($74,000) Plus new coal mining (S4,400).CM~ification of reflnerleg t. upgr~e residual oil to distillate fuels, $14,000, and incre~~ COSI production, $2,000. The refinery modification is based on data presented

in ch. 6, assuming that 0.6 MMB/D of domestically produced residual oil is already being upgraded in 1990.


Ch. 4—issues and Findings Ž 79

even if more certain cost data were available to- been made, the annual capital investmentday, different inflation rates in different sectors needed to maintain all aspects of liquid fuel pro-of the economy or modest technical develop- duction and use will depend on the level of fuelments could change any conclusions about rela- efficiency actually achieved. In particular, hightive costs by the 1990’s. Finally, once the initial levels of efficiency in end uses will require lowerinvestments to reduce oil consumption have levels of annual capital investment.

Box A.-Consumer Cost of Increased Automobile Fuel EfficiencyFor the purposes of this section, the consumer

cost of increased fuel efficiency was defined tobe the added cost of producing a more fuel-effi-cient car (relative to an otherwise comparablebut less efficient car) per gallon of fuel saved byusing the more efficient vehicle. The added costof producing more fuel-efficient cars will dependnot only on the investments needed to changeautomobile production facilities and the produc-tion volumes, but also the resultant changes inthe variable costs* of production, such aschanges in materials and labor costs.

As discussed on page 75, the capital invest-ments that are needed to increase fuel efficien-cy also produce other changes in automobiles;and allocation of costs among fuel efficiency andthe other changes is somewhat arbitrary. In addi-tion, if market demand for fuel efficiency isstrong, many changes that increase fuel efficien-cy would be incorporated into the normal capi-tal turnover of the industry. For the purpose ofcalculating consumer costs, however, essentiallythe same approach was taken as with the invest-ment cost estimates. The fraction of investmentsallocated to fuel efficiency are the same as forthe investment costs per barrel per day of fuelsaved; and only the average costs per averagegallon saved have been calculated-relatingeach 5-year period to the previous 5-year period—rather than compounding errors by assumingsome market-driven scenario as a point ofreference.

In addition, it was assumed that productionvolumes are sufficiently large so that there areno significant diseconomies from small-scaleplants or losses from underutilized facilities.Weak demand for fuel efficiency and/or for newcars in general could of course result in addition-al costs of this sort.

A key factor in consumer costs is the changein variable costs associated with producing morefuel-efficient vehicles. Variable cost estimates,

● Variable production costs are those that vary in proportion to thenumber of units produced as opposed to fixed costs such as capitalcharges.

however, are generally proprietary and can varyconsiderably from one company to another. Fur-thermore, changes in variable costs with in-creased fuel efficiency will depend, to a largeextent, on the success of efforts to develop pro-duction technologies that can hold down pro-duction costs. Some of the uncertainties in vari-able cost changes are discussed below, followedby illustrative examples of plausible consumercosts for increased fuel efficiency.

Some changes that increase fuel efficiency willlower variable costs; some will increase somevariable costs while lowering others; and stillother changes are likely only to increase variablecosts.* However, factors which are only periph-eral to the nature of the technology incorporatedin the car often dominate the change in variablecosts, These factors include: 1) the existingnature and layout of equipment in the plantbeing modified to produce the new car, 2) vari-ous specific production decisions (e.g., whichof various processes is used in manufacturing acomponent, what equipment will be modified,what will be the production volume), and 3) thesuccess of developing new, lower cost proce-dures for producing a component and assem-bling it in the vehicle. In other words, the netchange in variable costs depends not only onthe nature of the new technology and the wayit is produced, but also on the path the manufac-turer has chosen to evolve from the current pro-duction facilities and configurations to thoseneeded to produce the more advanced tech nol-

*For exampte, reducing automobile size and weight by reducingthe quandtyofmetedals mshms variable costs. Switching to lighterwei@t matedats has the side dfsct of reducing the needed size ofaxtes, auto fra~ ate,, which rbduces costs; but the higher cost ofthe new matena“ tand i~ dtfficuky of handling that material(e@ ddiw dw X finishin& painting, heat treating) canincraase vat?abtecmsts. Srt#@dy, producing a more efficient enginemay enatdemductton }a##ne s&e, number of cylinders and com-plexity of the polhKion control w@prnent, which reduces costs; butthe need for more precise machining and possibiy added equipment(e.g., turbochargers) can raise thevariabie costs. Finally, changes suchas going from a three-speed transmission to a four-speed, five-speed,or continuously variabie transmission are iikeiy to increase variabiecosts because of increased complexity and materiais and process-ing requirements.


BOX B.-Consumer Cost of Synfuels

The consumer cost of synthetic transportation Figure 5.-Consumer Cost of Selected Synfuels fuels will depend on a number of factors. The WithRates of Returns on Investmentsmost important of these are the actual capital in-vestment needed to build the synfuels plant, theway plant construction is financed, the requiredreturn on investment, the cost of delivering thefuel to the end user, and the end-use efficiencyof the synthetic product. For the first generationof synfuels plants, the costof producing the synfuel will also depend critically on plant perform- ‘ance, specifically the amount of time the plantis operated Mow its rated capacity due to thetechnical problems. (See ch. 6 for a more de-tailed sensitivity analysis.) Depending onassumptions about these factors, one can derivea wide variety of consumer costs.

Table 16 shows two sets of consumer costs forvarious fossil synfuels. These costs are based onthe best available investment and operating costestimates and assume no cost overruns, goodplant performance (90 percent of rated capac-ity), a lo-percent real return on equity invest-ment* and two financing schemes: 100-percentequity financing and 75/25 percent debt* */equi-ty financing. Figure 5 shows how these con-

*ln other words, a return on investment that is 10 percent higherthan general inflation.

● *The debt must be.project.specific, i.e., the money is loaned forthe specific project and is not general debt capital whose paybackis guaranteed by other company assets.

0 5 10 15 20 25

Real return on investment (%)

C o a l t o m e t h a n o l. o i l s h a l e and SNG

, Coal to Methanolto gasoline

a5% real interest rate on debt.


Table 16.-Estimated Consumer Cost of VariousFossil Synfuels Using Two Financing Schemes

Cost of Synfuel delivered to end usera($/gallon gasoline equivalent)Liquid transportation fuel 100% equity financingb 75% debt, 25°A equity financingReference cost of gasoline from

$32/bbl crude oil . . . . . . . . . . . . . . . . . 1.20Shale oil ‘ :. . . . . . . . . . .. . . . . ..Methanol from coal . . . . . . . . . . . . . . . 1.30d(1.10)e 0.95d(0.80)e

1.60 f(1.30)e 1.10 (0.90)e

Coal to methanol with 1 . 2 5d

Mobil methanol to gasoline. . . . . . . . . . . . . 0.80d

. . . 1 . 6 0g

1.00 g

0 .85d

anol usually costs more per gallon oftwice as high for methanol as for

real return on equityinvestment, 5-percent real investment on debt. ~~~~-m’ “ ‘ ‘ “ ‘“ :’:- - ~, (, ~,+

%Ju~ln ~hOOOS aaeume methanof ueed In englnedealgned for methanot uaewhlch Is 20 @“~ta~~lclentthm ● amapondmgsine Ongtne.

fMettWot@Geeollne R $%s =:SOUftOE: Offtoe of Teohndogy Aeaeaement.


creases the apparent cost of fuel efficiency pergallon saved by a factor of 2.5 over the situa-tion where no discount is applied to future fuelsavings. In practice, there will be a wide varietyof discounting rates used by various consumers,and the rates will change with market conditions(including oil prices and interest rates) and con-sumers’ beliefs about future oil prices and avail-ability, among other things.

Synthetic Fuels

For synfuels processes that produce sizablequantities of different fuel products (e.g., synthet-ic natural gas and fuel oil), one encounters ac-counting problems similar to those discussedunder increased automobile fuel efficiency-i.e.,how to allocate production costs to the varioussynfuels products. However, these problems areless severe for synfuels than for automobiles be-cause all of the major products of the former areseparate consumer products with known currentprices. Furthermore, the accounting problemscan be largely avoided by considering only proc-esses that produce only fuels of similar quality(e.g., gasoline, jet, and diesel fuel), and avoidedentirely by considering processes that produceonly one major product.

Variations in operating and maintenance costsand future coal prices produce some uncertain-ty in the cost of synfuels. The more importantuncertainties, however, involve the cost of build-ing a synfuels plant, how this cost will be fi-nanced, and investors’ required rates of return

on investment. The cost of synfuels from the firstgeneration of synfuels plants will also dependcritically on plant performance, with frequentshutdowns and repairs increasing costs dra-matically. Presumably, later generations of plantswill perform reliably.

Fuel Switching and Conservation

The total cost of switching utility boilers fromfuel oil to coal is fairly welt known. There are,however, some areas of the country where coalis not readily available, and there is insufficientspace to accommodate coal handling facilitiesat some electric generating plants.

The major variability in the cost of switchingto natural gas and electricity for buildings andin industry results from widely varying requiredrates of return on investments in different indus-tries and for different building owners or renters.At some sites, though, natural gas is not availableor space limitations prevent the installation ofgas facilities. There is also uncertainty in the costof finding and processing unconventional natu-ral gas (from tight sands, etc.).

The total cost of conserving heat and hot waterin buildings is probably the least certain of themeasures for reducing stationary oil use. Notonly are there large uncertainties and variabilityin the savings that can be achieved through vari-ous conservation measures, but also consumerswill discount future fuel savings at widely differ-ing rates.

HOW DO THE ENVIRONMENTAL IMPACTS OF INCREASEDAUTOMOBILE FUEL EFFICIENCY AND SYNFUELS COMPARE?

The synfuels, auto fuel efficiency, and electricauto alternatives for displacing imported oil havesharply different potential impacts on publichealth and safety, on workers, and on ecosys-tems. In addition, probabilities of these impactsactually occurring—few of them are inevitable—are also quite different. Both the potential impactsand their risks are briefly compared below. Thenature of some of the risks, however, is obscuredby the brevity of the following discussion. For ex-ample, the actual risk associated with possible

contamination of drinking water by synfuels pro-duction is heavily dependent on the degree ofprior recognition of the risk and response to thisrecognition –for example, development ofground water monitoring systems. Also, risks thatare similar in magnitude are often valued differ-ently because of the degree of choice involved(e.g., willing exposure to the risks of auto travelv. unwilling exposure to accidental toxic spills)and the precise nature of the risks (e.g., multipleautomobile accidents involving only a few peo-

84 ● Increased Automobile Fuel Efficiency and Synthetic Fuels: Alternatives for Reducing Oil lmports

ple at a time v. a serious accident or control fail-ure at a large synfuels plant).

Public Health and Safety

Reductions in vehicle size, part of the auto fuel-efficiency measures, could have the strongest ef-fect on public health and safety through their po-tential adverse effects on vehicle safety. The ef-fect is difficult to estimate because of a lack ofcomprehensive traffic safety data that would al-low an evaluation of the relative effect of car sizeand other key safety variables on vehicle crash-’worthiness and accident avoidance, and becauseof uncertainty about the compensatory measuresthat might be taken by the vehicle manufacturersand by drivers. Although the National HighwayTraffic Safety Administration has projected vehi-cle size reductions to cause an additional 10,000annual traffic deaths by 1990 if compensatorymeasures are not taken, this and other quantita-tive estimates of changes in traffic safety are basedon limited data and relatively crude models. Nev-ertheless, an increase in traffic deaths of a fewthousand per year because of vehicle size reduc-tions does seem plausible.

Diesel use could have an adverse effect onemissions of nitrogen oxides (NOX) and particu-Iates, and conceivably could cause public healthproblems in congested urban areas. The risk ismoderated, however, because: 1) controls forNOX and particulate are under active develop-ment, although success is not assured and it ispossible that the current level of effort will notbe continued; and 2) the evidence for healthdamage from diesel particulate is equivocal.

Electric passenger vehicles are likely to besmall, and thus should share safety problems withradically downsized high-mileage conventionalautomobiles. Additional safety problems causedby the batteries, which contain toxic chemicalsthat may be hazardous in an accident-causedspill, are offset somewhat by eliminating the fueltank with its highly flammable contents. Also,electric cars shouId have a positive effect on airquality, especially in urban areas, because thereductions in automobile emissions outweigh in-creased emissions from powerplants, except forsulfur dioxide (SO2).

Synfuels plants may expose the general publicto health and safety hazards in a variety of ways:contamination of drinking water from leachingof wastes, accidental spills, or failure of effluentcontrols; accidental release of toxic vapors; ex-posure to contaminated fuels; and routine emis-sions of conventional air pollutants such as SO2

and NOX. Only the routine emissions are essen-tially inevitable, however, and health and safetyproblems from these should be minimized byFederal ambient air quality standards and by therelative magnitude of these emissions, whichshould be considerably lower than emissionsfrom projected levels of development of coal-firedelectric generation during the same time frame.The extent of risk from the other sources is notwell understood because the toxic waste streamsfrom the plants have not been fully characterized,the effects of some of the known and suspectedwaste products are not yet well understood, andthe effectiveness and reliability of some criticalenvironmental control systems have not beendemonstrated under synfuels plant conditions(see issue on p. 95). Chemical industry sourcesbelieve that few problems will arise, but, asdiscussed in the above-mentioned issue, someareas of concern remain.

Worker Effects

With the possible exception of some workerexposure to toxic materials in battery manufac-ture, the only significant occupational health andsafety problem associated with the automobilemeasures appears to be mine safety and healtheffects involved in any increased mining of coalfor electricity needed for recharging electric carbatteries, and, to a lesser extent, for aluminummanufacture. These impacts are not trivial, be-cause the amount of coal needed per barrel ofoil saved for electric cars is of the same order ofmagnitude as that needed for synfuels produc-tion (assuming coal-fired electricity and coal-based synfuels). The coal-to-oil balance for alumi-num use is somewhat less certain, although someanalyses have calculated it to be similarly high.The use of aluminum is not the major part of theefficiency measures, however, and the actualamount of coal required is not likely to be signifi-cant in comparison with the coal used for syn-thetic fuels.


As noted, synfuels production has a coal re-quirement similar to that of electric autos, andthus shares similar mineworker problems. It hasimportant additional problems. The sources ofmoderate risks to public health and safety—fugi-tive emissions, spills, plant accidents, and con-taminated fuels–pose more serious risks to work-ers because of their frequency and severity of ex-posure. For example, workers will be continuous-ly exposed to low levels of polynuclear aromaticsand other toxic substances because fugitive emis-sions cannot be reduced to zero. Another impor-tant source of possible worker exposure is themaintenance requirements of synfuels reactors;the materials that must be handled in these opera-tions are likely to have the highest concentrationsof dangerous organics.

Exposure to hazardous substances is commonin the petrochemical industry, and worker-pro-tection strategies developed in this and relatedindustries will be used extensively in synfuelsplants. These strategies clearly will reduce thehazards, but the degree of reduction is highly un-certain (see issue on p. 95).

Ecosystem Effects

The only significant sources of ecosystem ef-fects from the automobile measures are likely tobe the changes in air quality caused by the useof electric autos (which probably will be positive)and diesels, and the air, land, and water pollu-tion associated with the mining and processingof both coal for electricity (for battery recharg-ing or aluminum production) and battery materi-als such as lead and lithium. Obtaining the newbattery materials is thought unlikely to cause im-portant environmental problems, but there aremany different kinds of potential battery materi-als, and final judgment probably should be with-held at this time. Nevertheless, with the excep-tion of the electric-car coal-mining requirements,any adverse ecosystem effects of the automobilemeasures appear likely to be mild.

Synfuels production is likely to cause signifi-cantly greater adverse effects, because it will havecoal-mining damages per barrel of oil roughlysimilar to electrical autos as we//as several addi-tional and potentially important adverse impacts.

These include substantially increased mining andwaste disposal requirements if oil shale is the syn-fuels feedstock, and a variety of potential adverseimpacts stemming from the possibility that toxicmaterials generated during the conversion proc-esses will escape to the environment. The path-ways of potential damage from toxics are essen-tially identical to those threatening public healthand safety—surface and ground water contamina-tion, toxic vapors, and exposure (in this case fromspills) to contaminated fuels. Unfortunately, prob-ability of the damage actually occurring is equallydifficult to evaluate.

An additional concern is that synthetic fuelsfrom biomass sources–which in general havesimilar or less severe environmental problemsthan coal-based synfuels—may have more severeecosystem effects because of the very extensivenature of their resource base. The adverse ecosys-tem effects of large increases in grain productionto produce gasohol, for example, can be quiteserious, and, given the nature of the currentagricultural system, the probability of such effectsoccurring is high.

Summary

The environmental impacts of increased auto-mobile fuel efficiency and synthetic fuels develop-ment will be quite different and difficuIt to com-pare. The major impacts of auto efficiency im-provements are likely to be increases in crash-related injuries and fatalities from auto size reduc-tions. The severity of these impacts is heavilydependent on vehicle design and driver behavior(especially seatbelt usage). Synthetic fuels devel-opment’s major impacts will include the well-known ecosystem effects as well as public andworker health and safety effects of large-scalemining and combustion of coal. Oil shale devel-opment will have many similar effects; a mostserious environmental risk may come from inade-quate disposition of the spent shale. In addition,there are potentially serious impacts on peopleand ecosystems from the escape of toxic sub-stances from synfuels conversion processes. Theseverity of these impacts is unclear because im-portant waste streams have not been character-ized and environmental control effectiveness andreliability has not been demonstrated.


HOW WILL THE SOCIAL IMPACTS OF SYNFUELS AND INCREASEDAUTOMOBILE FUEL EFFICIENCY COMPARE?

Identifying, assessing, and comparing the socialimpacts of synfuels development and improvedautomobile fuel efficiency are difficult becausethese impacts will not be distributed evenly intime or among regions. Moreover, they cannotnecessarily be measured in equivalent (e.g., dol-Iar) terms, and they are difficult to isolate and at-tribute to specific technical choices. Both benefi-cial and adverse social consequences will arisefrom these two approaches to reducing oil im-ports.

Employment

Synthetic fuels production presents two majorconsiderations about social impacts related toemployment. First, there is the possibility of short-ages of experienced chemical engineers andskilled craftsmen. A rapid growth in synfuelswould likely put increased pressure on engineer-ing schools, which are now suffering from insuf-ficient numbers of faculty. The second concernarises from the large and rapid fluctuations inlabor requirements for construction. While noshortages of construction workers are expected,on the average, fluctuating labor requirementsduring construction and startup can have severesecondary effects on communities at the con-struction sites. A population increase of aboutthree to five people per new worker could occur,leading to possible population fluctuations of30,000 to 60,000 people for some synfuels con-struction.

The changing structure and markets of the es-tablished automobile industry are likely to leadto a long-term, permanent decline in auto-relatedindustrial employment. The nature of this declinewill depend on import sales, the growth rate ofthe U.S. auto market, the competitiveness and

labor intensity of U.S. manufacturing, the use offoreign suppliers and production facilities, andthe adoption of more capital-intensive produc-tion processes and more efficient managementpractices. The skill mix will also shift increasing-ly towards skilled labor. Scarcities of experiencedengineers and certain supplier skills could inflatethe prices of skilled manpower resources for bothsynfuels and changing automotive technology.

Community Impacts

Synfuels development will have its most imme-diate effect in relatively few small and rural oilshale communities in the West, as well as in thesmall rural communities located near many ofthe Nation’s dispersed coal resources. In the longterm, local communities should benefit from syn-fuels in terms of expanded tax bases and in-creased wages and profits. However, in the nearterm, there are risks of serious disruptions in boththe public and private sectors of these communi-ties. The nature and extent of these disruptionswill be determined by the community’s ability toabsorb and manage growth, and the rate andscale of local synfuels development.

Automobile production jobs are presently con-centrated in the North-Central region of the Na-tion. The geographical distribution can be ex-pected to change as inefficient plants are closedand new production facilities are established inother parts of the United States. New plants willprovide new employment opportunities with ac-companying community benefits (e. g., tax rev-enues); plant closings in areas heavily orientedtowards the auto industry would deepen the ex-isting economic problems of the North-Centralregion, i.e., high unemployment, rising socialwelfare costs, and declining tax base.


HOW DO THE REGIONAL AND NATIONAL ECONOMIC IMPACTSOF SYNFUELS AND INCREASED AUTO EFFICIENCY COMPARE?

In addition to comparisons on the basis of costper barrel, environmental impacts and local socialimpacts, increased automobile fuel efficiency andsynfuels can be compared on the basis of theirpotential regional and national economic im-pacts. The latter comparison is important becauseeach type of investment implies an alternative na-tional strategy to achieve the goals of price stabil-ity, national economic growth, and equity as wellas oil import reduction.

Regional and national aggregation are also im-portant because both industries are capital-inten-sive. Large blocks of investment must be mobil-ized, with key investment decisions made by arelatively small number of firms, based on veryuncertain longrun predictions about the future.As summarized below, a variety of important na-tional and regional issues are raised by the uncer-tainties and inflexibilities inherent in these deci-sions.

Inflation and Economic Stability

Inflation may be dampened and the economystabilized if either type of investment is successful.In the case of synfuels, if first generation plantsdemonstrate competitive costs, the mere pros-pect of rapid deployment could moderate oil im-port prices and thus help to control what hasbeen one of the major inflationary forces duringthe last decade. In the case of autos, if increasedfuel efficiency helps domestic firms to hold or per-haps increase their market share, this would keepU.S. workers employed and at least stabilize for-eign payments for autos. Higher employment alsotends to reduce Federal transfer payments, whicheither reduces the Federal deficit or lowers taxes.Reductions or stabilization of foreign paymentstends to strengthen the value of the dollar in for-eign exchange markets. Both changes, in the Fed-

eral budget and on foreign accounts, reduce infla-tionary pressure.

On the other hand, attempts to displace oilimports too quickly may be inflationary. Risks ofinflation, technical errors, and market miscalcula-tions all increase with the rate of synfuels deploy-ment and with shortening the time taken to con-vert the domestic auto fleet to high fuel efficiency.

in the case of synfuels, rapid investment growthin the next decade, beyond construction of dem-onstration projects, could cause inflation by creat-ing suppliers’ markets in which prices for con-struction inputs, especially chemical engineeringservices, can rise more rapidly than the generalinflation rate. Deployment prior to definitive test-ing in demonstration plants also compoundspotential losses due to design errors.

In the case of autos, rapid large-scale invest-ments can inflate prices of vehicles as firms at-tempt to amortize capital costs quickly. However,if these attempts fail, presumably because buyersstop buying high-priced domestic autos, thennewly invested capital must be written off pre-maturely, resulting in the waste of scarce re-sources for the firm and the Nation. Furthermore,if rapid fuel-efficiency improvements are forcedby abrupt, real fuel price increases or by ag-gressive foreign auto competition, then thedomestic auto industry and owners of fuel-inef-ficient cars will both be forced to absorb lumpsum losses in the real value of current assets. Lowprices for new cars resulting from competition do,however, benefit purchasers of these cars.

Employment and InternationalCompetition

If improved fuel economy makes domesticallyproduced autos more competitive with imports,


there will be two major national economic pay-offs besides fuel savings. First, this improved com-petitiveness will protect traditional U.S. jobs; sec-ond, it will reduce the drain of foreign paymentsto auto exporting countries as well as to oil ex-porters. Synfuels do not present a similar couplingof economic possibilities.

There are major doubts, though, about thelongrun success of U.S. automakers with foreigncompetition. The United States may not be ableto compete in the mass production of fuel-effi-cient autos for a variety of reasons—such as highwages, low productivity, and inefficient or out-of-date management. All such explanations arespeculative, but together they have raised seriousdoubts about U.S. competitiveness in the con-text of the recent, rapid increase in the marketshare of auto imports. If foreign automakers con-tinue to drive domestics out of the market for fuel-efficient autos, synfuels investments may be pre-ferred over investments in fuel efficiency even ifthe apparent cost per barrel of the former arehigher.

Assuming investment in either industry doeslead to increased U.S. production, employmentopportunities for synfuels and autos can be com-pared based on 1976 data (the most recent avail-able). Synfuels production involves mainly min-ing and chemical processing activities, which in1976 dollars had $59,000 and $55,000 investedper worker respectively. On the other hand, thetransportation equipment sector of the economy(which is dominated by autos) had $27,000 in-vested per worker and auto suppliers such as fab-ricators of metal, rubber, and plastic products hadabout $21,000 per worker. In other words, in therecent past the auto industry created about twiceas many jobs per dollar of investment as industrialactivities similar to synfuels. The current trendtoward automation in automating will undoubt-edly lower its labor intensity, but the auto industryshould continue to employ more workers per unitof investment.

Income Distribution Among Regions

Another question concerns the likely regionaldistribution of incomes from autos and synfuels.An analysis of location factors was not carried out,

but two points can be made. First, to the extentthat the auto industry could use existing plantsor build nearby, current employment patternsand established communities could be main-tained. This would preclude costly relocation andwould tend to favor the North-Central region ofthe United States, which has been losing its in-dustrial base.

Second, new auto plants can be located inmore areas of the country than new synfuelsplants because of the high cost of transportingsynfuels feedstocks, especially oil shale, com-pared with the cost of transporting manufacturedmaterials and parts for automobiles. Transporta-tion costs are likely to concentrate synfuels invest-ments in regions of the Nation with superior shaleand coal reserves. Biomass options are least likelyto be concentrated, because resources are dis-persed, and coal-based options are much moreflexible than shale because coal is more widelydispersed.

Capital Intensity andOwnership Concentration

Finally, both strategies for oil import substitu-tion affect the number of profitable firms in eachindustry. In both industries, the number of com-petitive firms is severely constrained by the sizeof investment outlays and by the acquired knowl-edge of those already in the business.

In liquid fuels, the introduction of syntheticfuels sharply increases the amount of capital in-vestment required per barrel of liquid fuels pro-duction capacity. For example, in the case of onemajor oil company, present capitalized assets peraverage daily barrel of oil equivalent of produc-tion from old reserves of conventional oil and nat-ural gas is less than 20 percent of OTA’s estimateof the similar ratio for oiI shale. * However, newreserves of conventional oil and gas will also re-quire much larger capital outlays than old re-serves, due to depletion of finite natural re-sources.

*Value of assets for Exxon was obtained from its 1980 AnnualReport.


From the investor’s viewpoint, the sequenceof investments for conventional petroleum re-sources is very different from synfuels projects.For conventional petroleum, small operators canexplore for new reserves, at least on shore, withonly a relatively small amount of high-risk moneyand the limited technical staff required to renta drilling rig and to determine where the wildcatwell should be drilled. If a discovery is made, sub-sequent, much larger investments in develop-ment wells and pipelines can be made at relative-ly low risk. In synfuels, a firm simply cannot enterthe business without command of all capital re-quirements up-front, or without a very large staffof technicians and managers.

In summary, investment options to discoverand develop conventional oil and gas will be ex-ploited before synthetics even if estimated totalcapital outlays are the same, because the formerconfine major risks to the front-end of projectsbefore the largest blocks of capital must be com-mitted.

As a result, it is likely that only a very small frac-tion of the hundreds of firms currently produc-

ing conventional oil and gas will have the finan-cial and technical means to produce synfuelswhen conventional resources are depleted.While this growing concentration of ownershipmay not lead to the classical problem of price fix-ing by domestic producers, because oil and gasare traded on worldwide commodity markets, itdoes at least make the industry appear to be moremonolithic, since synfuels project managers willcommand very large blocks of human andmaterial resources.

In domestic automating, ownership may be-come more concentrated because at least twoout of the three major U.S. companies are beingforced, by lack of capital and perhaps by highproduction costs, to curtail the number of differ-ent vehicles made. Although foreign automakersare increasing their U.S. manufacturing activities,the growing dominance of one major U.S. auto-maker over the other two may decrease pricecompetition in certain types of cars and possiblyreduce profitmaking opportunities for domesticsuppliers to auto manufacturing because of themarket leverage of the one dominant buyer.

WHAT DO INCENTIVES FOR INCREASED FUEL ECONOMY IMPLYFOR THE EVOLUTION AND HEALTH OF THE U.S. AUTO INDUSTRY?

The auto industry began a process of structuralchanges in the 1970’s which complicates evalua-tion of how fuel economy policy might affect theindustry, auto manufacturing communities, andthe national economy. Regardless of fuel econ-omy policy, the U.S. auto industry is undergoinga long-term decline in terms of employment, thenumber of domestic firms (including suppliers),and the proportion of global auto productionsited in the United States. Recent consumer de-mand for fuel economy and other auto character-istics have supported these trends by motivatingcostly product changes to meet competition fromforeign firms. Factors such as the relatively fastsales growth in foreign auto markets and lowercosts of labor and capital abroad have inducedU.S. manufacturers to increase investments in for-eign production activities.

Increases in demand and other pressure onU.S. firms to raise fuel economy will reinforce andperhaps accelerate current industry and markettrends. Although large spending needs will moti-vate reductions in the number of independentfirms and, perhaps, the breadth of their opera-tions, the size and financial health of the U.S. autoindustry in the future will depend, to a greatdegree, on its ability to compete with foreignfirms–particularly in the small-car market. Thecompetitiveness of U.S. auto firms depends notonly on product designs and production facilitiesbut also on total manufacturing costs, which re-flect labor costs and the efficiency of production,organization, and management. Incentives for ac-celerating fuel-efficiency increases will not onlydirectly affect the investment requirements, buta combination of high perceived investment

90 . Increased Automobile Fuel Efficiency and Synthetic Fuels: Alternatives for Reducing Oil /reports

needs, possible rigidity in U.S. costs, and a slow-growing competitive U.S. market may discourageU.S. firms from investing in U.S. capacity.

There are some auto company activities thatshould be relatively invulnerable to fuel econ-omy-motivated market changes and should con-tinue in the United States. These activities includeproduction of specialty cars and nonautomotiveprojects such as defense contracting. U.S. autocompanies may continue to conduct some activ-ities in the United States at historic or greaterlevels, while they may reduce the levels of othersor eliminate them entirely.

A decline in U.S. auto production, especiallyone that is not substantially offset by growth inforeign-owned capacity in the United States,

poses a major policy dilemma. On the one hand,the auto industry metamorphosis may result ina more economically efficient domestic industrythat is more competitive with strong import com-petition. On the other hand, the process of indus-try change results in loss of jobs for current auto-workers and loss of employment and businessactivity for local economies, losses which are rela-tively large and regionally concentrated in thealready economically depressed North-Centralregion. These concerns can be dealt with throughindustrial and economic development policies,but it should be recognized that policy to accel-erate fuel economy improvements may aggravatethem. In addition, many of these changes mayoccur even in the absence of strong demand forfuel efficiency, but possibly at a slower rate.

CAN WE HAVE A 75= MPG CAR?

There are no technological barriers to design-ing and building a four-passenger automobile thatcould achieve 75 mpg on the combined 55/45percent highway/city EPA driving cycle. Such acar would take at least 5 years to design and de-velop and might be costly to manufacture, butit is technically feasible. It should be noted, ofcourse, that the appearance of one or a few mod-els that get 75 mpg would have littIe immediateeffect on fleet average fuel economy, or on theNation’s petroleum consumption.

High fuel economy entails tradeoffs and com-promises that affect other features of vehicledesign–carrying capacity, performance, safety,comfort, and related amenities. Technology is acritical factor in managing these tradeoffs. Someroutes to improved passenger car fuel economyalso increase manufacturing costs (diesel engines,more complicated transmissions, lightweight ma-terials). Here again, better technology can helpto improve fuel economy at the least cost.

If a 75-mpg car can be made sufficiently attrac-tive to consumers in terms of the other featuresbeyond fuel economy that affect purchasing deci-sions—including price, but also the variety of lesstangible factors that contribute to perceived val-ue—then automakers will build such cars, confi-

dent that they will find a market. The interplaybetween consumer demand and automotivetechnology will determine when 75-mpg cars willappear. Consumer expectations concerning fuelcosts and the possibility of future shortages of fuel,as well as their judgments of the practicality ofsuch cars, will be important factors affecting therate at which these cars would be introduced.

An automobile designed to achieve 75 mpgmight look much like a current subcompact—e.g., a General Motors Chevette—but, as dis-cussed in chapter 5, would be considerably differ-ent under the skin. It would have to be lighter,and might also be somewhat smaller—with a curbweight of perhaps 1,600 lb as opposed to about2,000 lb for the Chevette. The actual weight de-pends not only on the size of the car, but alsoon the materials from which it is made. By usingmaterials with high strength-to-weight ratioswherever possible—or, where strength or stiffnessare not important, materials of low density—afour-passenger car could weigh, in principle,even less than the 1,600 lb suggested above.Costs are the limiting factors in the use of suchmaterials—both the costs of the materials them-selves and the costs of the required manufactur-ing processes.

Ch. 4—issues and Findings • 91

Photo credit: Volkswagen of America

Artists drawing of the VW2000, a three-cylinder diesel test vehicle that has achieved over 60 mpg instandard fuel-efficiency tests

The other essential element in a 75-mpg car isan efficient powertrain. For a car weighing 1,600lb or less, a relatively small diesel engine–onewith a displacement in the range of 0.9 to 1.3liters–would suffice. The transmission could beeither a manual design or a considerably im-proved automatic, perhaps a continuously vari-able transmission.

To get 75 mpg would also require a great dealof attention to the detailed design of many aspectsof the car—low aerodynamic drag, low rolling re-sistance, use of microprocessor controls, minimalaccessories, and parasitic loads—with careful en-

gineering development throughout the vehiclesystem. None of this depends on technologicalbreakthroughs.

Given equally good design practices, the result-ing car would not be as safe as a larger vehicle.Nor would it be luxurious. It might not have airconditioning. It would probably not be able topull a camping trailer through the Rocky Moun-tains. But it could get 75 mpg. When automobilemanufacturers—here, in Europe, and in Japan—decide that American consumers want such a car,they will build it.


ARE SMALL CARS LESS SAFE THAN LARGE CARS?One of the easiest ways to increase the fuel

economy of passenger cars is to make them light-er. Although it is possible to make cars somewhatlighter without making them smaller, in generalsize and weight go together. Thus, cars with in-creased fuel economy are typically smaller—adownward trend in the size and weight of carssold in the U.S. market began in 1977 and willcontinue through the 1980’s, although gradual-ly leveling off.

Size is the more critical variable for safety, al-though weight also affects the dynamics of colli-sions. A great deal of improvement in the safetyof cars of all sizes is possible through improveddesign–but given best practice design, a big carwill always be safer than a small car in a colli-sion. As a result, making cars smaller to improvefuel economy will, everything else being equal,increase risks to drivers and passengers. Assum-ing no change in the way the cars are driven,there will be more injuries and fatalities thanwould occur with bigger cars embodying equiva-lent design practices and having identical acci-dent avoidance capabilities.

Size affects safety because when an automobilehits another object–whether another car, a truck,or a roadside obstacle—the car itself slows, or isdecelerated (the “first collision”), and the occu-pants must then be slowed with respect to thevehicle (the “second collision”). To minimize thechance of injury, the decelerations of the occu-pants with respect to the passenger compartmentduring the second collision must be minimized,The occupants must also be protected against in-trusion or penetration of the passenger compart-ment from the outside. But controlling decelera-tions during the second collision depends on thedeceleration of the entire vehicle during the firstcollision. Given good design practices, the sever-ity of both the first and the second collision canbe lowered, on the average, if the car is madelarger.

Ideally, the vehicle structure will deform in acontrolled manner around the passenger com-partment during a collision, so that the averagedeceleration of the passenger compartment in the

first collision will be low. The larger the car, themore space the designer can utilize to managethe deformations and decelerations—e.g., byusing a crushable front-end. In a small car, thereis less room for controlled deformation withoutintruding on the passenger compartment. Withinthe passenger compartment, more space meansmore room to control the deceleration of the pas-sengers—using belts, harnesses, padding, andother measures—with less risk of hitting unyield-ing portions of the vehicle structure. More roomalso makes penetration or other breaches of theintegrity of the compartment less likely. Onepathway to increased fuel economy without sacri-ficing collision protection is therefore to makecars lighter by design changes and/or differentmaterials while preserving as much space as pos-sible for managing the energies that must be dis-sipated in the first and second collisions.

Because vehicle size and weight are not theonly significant factors in determining vehiclesafety, and because “all other things being equal”does not apply in actual real-world situations, anyconclusion about the relative safety of large andsmall cars should be tempered with the follow-ing observations:

1.

2.

3.

The recent series of crash tests sponsored bythe National Highway Traffic Safety Adminis-tration demonstrated that vehicles of approx-imately equal size can offer remarkably dif-ferent degrees of crash protection to theiroccupants. In many cases, differences be-tween cars of equal size overshadowed dif-ferences between size classes in the kind ofaccident tested (35-mph collision head-oninto a barrier).Crash-avoidance capabilities of large andsmall cars are unlikely to be the same, andany differences must be factored into anevaluation of relative safety. Unfortunately,the effects of differences in such capabilitiesare difficult to measure because they repre-sent both physical differences in the vehiclesand driver responses to those differences.Available traffic safety data and analysis isoften confusing and ambiguous on the sub-ject of large car/small car safety differences.

Ch. 4—issues and Findings • 93

Although analyses of car-to-car crashes tendto agree that occupants of large cars are ata lesser risk than those of smaller cars, thereis no such firm agreement about the otherclasses of accidents that account for three-quarters of all passenger vehicle occupantfatalities. A probable reason for the ambigu-ity of results is the shortage of consistent, na-tionwide data on accident incidence and de-

tails; only fatal accidents are widely re-corded. Another reason is the multitude offactors other than car size that might affectinjury and fatality rates. Important factors in-clude differences (among different sizeclasses) in driver and occupant age distribu-tion, general types of trips taken, average an-nual mileage, vehicle age distribution, andseatbelt usage.

HOW STRONG IS CURRENT DEMAND FORFUEL EFFICIENCY IN CARS?

An extremely important factor influencing fu- Table 17 presents a compariture new-car average fuel efficiency is the marketdemand for this attribute, relative to the otherfeatures the new-car buyer wants. Although it is,at best, only an approximate measure of futuremarket behavior, examination of recent demandpatterns in the new-car market can provide someinsights about current demand for fuel efficien-cy. I n particular, the importance of fuel efficien-cy as compared with car size, price, and perform-ance is examined for 1981 model gasoline-fueledcars* sold through January 5, 1981.

son of the averagefuel efficiency of new gasoline-fueled 1981 modelcars sold through January 5, 1981, with the fuelefficiency of the most efficient car in each of theEnvironmental Protection Agency’s (EPA) ninesize classes. Also shown are the sales fractionsand the nationality of the manufacturer of themost efficient vehicle, These data show that theaverage fuel efficiency of new cars sold was 25mpg, but if consumers had bought only the mostfuel-efficient car in each size class* (and manu-facturers had been able to supply this demand),

*The results of the analysis would change somewhat if dieselswere included, primarily with respect to nationality of manufac- *lnterestingly, this would also have resulted in U.S. manufac-turers because U.S. manufacturers did not offer diesels in several turers and captive imports capturing over 90 percent of sales, rathersize classes in 1981. U.S. manufacturers, however, are beginning than 74 percent of sales that they actually achieved in this period.to offer diesels in most size categories; but the relevant data are If diesels are included, however, average fuel efficiency could havenot now available. In 1981, about 95 percent of the automobiles been slightly higher than 33 mpg, but less than 60 percent of thesold were gasoline-powered. cars purchased would be domestically produced.

Table 17.—Comparison of Average and Highest Fuel Efficiencyfor 1981 Model Gasoline-Fueied Cars in Each Size Class

Sales-weighted averagefuel efficiency of cars Fuel efficiency of most Nationality of

Sales fraction sold through Jan. 5, 1981 fuel-efficient model in manufacturer of mostEPA size class (percent) (mpg) size class (mpg) fuel-efficient modelTwo-seater. . . . . . . . . . . . . 2 22 30 ItalianMinicompact . . . . . . . . . . . 3 34 45 JapaneseSubcompact . . . . . . . . . . . 30 28 42 United States

(Captive Import)Compact . . . . . . . . . . . . . . 9 27 37 United StatesMidsize . . . . . . . . . . . . . . . 37 23 31 United StatesLarge . . . . . . . . . . . . . . . . . 9 24 United StatesSmall wagon . . . . . . . . . . . 4 30 37 JapaneseMidsize wagon . . . . . . . . . 5 23 30 United StatesLarge wagon . . . . . . . . . . . 1 18 20 United States

Sales-weighted average 25 33SOURCE: Data from J. A, Foster, J. D. Murrell, and S. L. Loos, US. Environmental Protection Agency, “Light Duty Automotive Fuel Economy . . Trends Through 1981,”

SAE papar No. 810388, Februa~ 1981.


the average fuel efficiency would have been 33mpg (a 33 percent increase). Because EPA’s sizeclasses are based on cars’ interior volume, it isclear that demand for large interior volume in carsis not currently preventing a significantly higheraverage fuel efficiency in new cars than is actuallybeing purchased.

Similarly, a comparison of prices shows that themost fuel-efficient 1981 model cars generally hada base sticker price in the middle or lower halfof the price range of cars in each size classifica-tion. * Thus, there is no evidence that price is con-straining the purchase of fuel-efficient cars either.

A further comparison of the average and mostfuel-efficient cars in each size category shows thatthe average cars are heavier and have more pow-erful engines than the most fuel-efficient models.However, OTA’s analysis indicates that the great-er engine power found in the average car soldis, to a large extent, needed simply because thecar is heavier.** There is no indication that the

“Sales-weighted average sticker prices for comparably equippedcars are not currently available.

**For example, in subcompacts and midsized cars (accountingtogether for 67 percent of sales), the average car weighed about33 percent more and had about 50 percent larger engine displace-ment (which is correlated to power) than the most fuel-efficientcar. A further comparison of specific fuel efficiency (ton miles per

most fuel-efficient car in most size categories per-forms (e.g., accelerates) significantly worse thanthe averge 1981 model car actually sold in thatcategory. Although there are exceptions to thisin certain size categories (e.g., two-seaters, largecars, and possibly midsized station wagons), theseexceptions account for only about 10 to 15 per-cent of total sales.

This analysis indicates that interior volume,price, and performance cannot account for thelarge difference between the fuel efficiency ofcars actually sold and what was available. As in’the past, consumers consider features such asstyle, quality, safety, ability to carry or haul heav-ier loads, and energy-intensive accessories to beof comparable importance to fuel economy. Itis probable, therefore, that new-car average fuelefficiency could be significantly increased if con-sumer demand for fuel economy were strength-ened.

gallon, or the mpg of an equivalent car weighing 1 ton) shows that,for midsized cars (37 percent of sales), the difference in fuel effi-ciency between the average and the most fuel-efficient car can beexplained solely on the basis of weight. Thus, there is no indica-tion that the average car has better performance characteristics (e.g.,acceleration) than does the most fuel-efficient model. A similar com-parison of subcompacts indicates that, if anything, one would ex-pect the most fuel-efficient model to perform better than theaverage.

WHAT ARE THE PROSPECTS FOR ELECTRIC VEHICLES?EVs were among the first cars built, but they

had almost vanished from the marketplace by the1920’s, primarily because they could not com-pete with gasoline-powered vehicles in terms ofprice and performance. Due to concern overautomobile emissions, the increasing price of oiland recent oil supply disruptions, however, therehas been a renewed interest in this technology.The advantages of EVs are that they derive theirenergy from reliable supplies of electricity, whichcan be produced from abundant domestic energysources, and they operate without exhaust emis-sions. Their disadvantages are their high cost andpoor performance and the increased sulfur diox-ide emissions that result from increased electricgeneration.

From the consumer’s point of view, the prob-lems with EVs are principally centered on bat-tery technology. Current batteries are expensiveand heavy, relative to the energy they store; theyrequire several hours to recharge; and they mustbe replaced approximately every 10,000 miles ata cost of $1,500 to $3,000. The weight of batterieslimits vehicle range, * performance, and cargo-and passenger-carrying capacity. Because of thecost of batteries and electric controls, a new EVis estimated to cost about $3,000 more than acomparable gasoline-powered vehicle. And re-placing batteries every 10,000 miles because oflimited life would add more than $().10/mile to

*Usually 100 miles or less between recharging.

Ch. 4—Issues and Findings ● 9 5

operating costs, which is equivalent to gasolinecosting more than $4 to $6/gal for a 40- to 60-mpgcar.

Future developments in battery technologycould improve the prospects for EVs, and severalapproaches are being pursued. But the under-standing of battery technology is not adequateto predict if or when significant improvementswill occur.

If battery problems persist, sales of EVs couldbe limited to a relatively few people and firmsthat can afford to pay a premium to avoid trans-portation problems that would arise if liquid fuelsupplies were disrupted. On the other hand, ifgasoline prices increase by more than a factor offour or five or if gasoline and diesel fuel are ra-tioned at levels too low to satisfy driving needseven with the most fuel-efficient cars, then EVscould be favored—provided electricity prices donot also increase dramatically.

Prospects for EVs may also be influenced byGovernment incentives based on national andregional considerations. One such considerationis the oil displacement potential of EVs. EVs aremost nearly a substitute for small cars, which arelikely to be relatively fuel efficient in the 1990’s;but the limited range of current and near-termEVs prevents them from being a substitute for all

of the yearly travel needs supplied by a small gas-oline-driven car. As a result, oil displacement byEVs is likely to be relatively small; probably nomore than 0.1 million barrels per day (MMB/D)with a 10-percent market penetration, even as-suming no oil consumption by those electric util-ities supplying electricity to EVs.

At current levels of utility oil consumption,however, the net oil displacement would be lessthan 0.1 MMB/D (see ch. 5 for details). Futurereductions in utility oil consumption will improvethe oil displacement potential of EVs, while in-creased fuel efficiency in petroleum-fueled carswill reduce any advantage EVs might have in thisconnection.

A final consideration is the reduced automotiveemissions and other environmental effects of EVuse. Because that use would be concentrated inurban areas and the necessary increased electricpower generation would be well outside of theseareas, cities with oxidant problems that replacelarge numbers of conventional vehicles with EVswill significantly improve their air quality. This in-centive could improve the prospects for EV salesand use. Emissions and other impacts of increasedpower generation may cancel some of this bene-fit, but the positive urban effects are likely to beconsidered the most important environmental at-tributes of EVs.

IF A LARGE-SCALE SYNFUELS INDUSTRY IS BUILT . . . WILLPUBLIC AND WORKER HEALTH AND SAFETY AS WELL AS THE

ENVIRONMENT BE ADEQUATELY PROTECTED?It is virtually a truism that all systems designed environmental community has focused on the

to produce large amounts of energy will have the potential damaging effects, while the industry haspotential to adversely affect the environment and focused on the controls and environmental man-human health and safety. It is equally true that, agement procedures available to them. Gainingwith few exceptions, it is technically feasible to a perspective on the correct balance betweenreduce these effects to the point where they are these two points of view—on the likelihood thatgenerally considered an acceptable exchange for some of the potential damages will actually occurthe energy benefits that will be obtained. In cur- —is especially important in the case of synfuelsrent arguments concerning synthetic fuels devel- development because environmental dangersopment, as with many other such arguments, the have become a genuine public concern.


As shown in the evaluation of potential envi-ronmental impacts in this report, many of the im-portant impacts of a large synfuels industry willbe similar in kind to those of coal-fired electricpower generation. The magnitude of these im-pacts (acid drainage and land subsidence fromcoal mines, emissions of sulfur and nitrogen ox-ides and particulate, effects of water use, popula-tion increases, etc. ) is likely to be similar to andin some cases less than the likely impacts of thenew, tightly controlled electric-generating capac-ity projected to be installed in the same timeframe.

A second set of impacts–those associated withthe toxic materials present in the process andwaste streams of the plants and possibly in theirproducts—are not predictable at this time but are,nevertheless, very worrisome. Factors that shouldbe useful in gaging the risk from these impactsinclude the technical problems facing the design-ers of environmental controls, the availability ofadequate regulations and regulatory agency re-sources, past industry and Government behaviorin implementing environmental and safety con-trols, and difficulties that might be encounteredin detecting adverse impacts and tracing them totheir sources. A brief discussion of these factorsfollows:

1. Technical Problems.—Virtually all the con-trols which are planned for synthetic fuelsplants are based on present engineeringpractices in the petroleum refining, petro-chemical, coal-tar processing and powergeneration industries, and industry spokes-men appear confident that they will worksatisfactorily. Problems may be encountered,however, because of differences betweenthese industries and synfuels plants—the Iat-ter have higher concentrations of toxic hy-drocarbons and trace metals, higher pres-sures, and more erosive process streams, inparticular, As yet, few effluent streams havebeen sent through integrated control sys-tems, so it has not yet been demonstratedthat the various control processes will worksatisfactorily in concert, Technical person-nel at the Environmental Protection Agen-cy (EPA) and the Department of Energy

2.

3.

(DOE) have expressed particular concernsabout control-system reliability.

Judging from these indications, it appearspossible that a considerable period of time–possibly even a few years–will be necessaryto solve control problems in the first fewplants and get the environmental systemsworking with adequate performance andreliability. Delays are especially likely fordirect-liquefaction pIants, which have someparticularly difficult problems involving toxicsubstances and erosive process streams.These delays may be aggravated by a poten-tial gap in control technology development.Recent Federal policy has left the develop-ment of environmental controls largely to in-dustry. The major concern of the synfuelsindustry, on the other hand, is to clean upwaste streams so that existing regulationsmay be met. Less emphasis is placed on con-trolling pollutants such as polynuclear aro-matics that are not currently regulated. It ap-pears certain that there will be considerablepressure to regulate these and other pollut-ants, but it is not certain that the industrywill be able to respond quickly to such regu-lations.Detecting and Tracing Impacts. —One of themajor potential dangers of synfuels plantswill be low-level emissions of toxic sub-stances, especially through vapor leaks (pri-mary danger to workers) or ground watercontamination from waste disposal (primarydanger to the public and the general envi-ronment), Current ground water monitoringprobably is inadequate to provide a desirablemargin of safety, although presumablyknowledge of this danger will result in bet-ter monitoring systems. A major problemmay be the long lag times associated withdetecting carcinogenic/mutagenic/teratogen-ic damages—a major concern associatedwith trace hydrocarbons produced underthe physical and chemical conditions pres-ent in most synfuels reactors.Regulation. –The regulatory climate facingan emerging synfuels industry is mixed. Onthe one hand, ambient standards for partic-ulates, sulfur oxides, and other pollutants

Ch. 4—issues and Findings ● 9 7— .-.

associated with conventional combustionsources are in place and should offer ade-quate protection to public health with re-spect to this group of pollutants. A limitednumber of other standards, including thosefor drinking water protection, also are inplace. On the other hand, new source per-formance standards–federally set emissionstandards—have not been determined yet,nor have national emission standards forhazardous air pollutants been set for the vari-ety of fugitive hydrocarbons or vaporizedtrace elements that might escape from syn-fuels plants. Likewise, although Occupation-al Safety and Health Administration (OSHA)exposure standards and workplace safety re-quirements do apply to several chemicalsknown or expected to occur in synfuels pro-duction, the majority of such chemicals arenot regulated at this time.

These regulatory gaps are not surprising,given the limited experience with synfuelsplants, and several of the standards–espe-cially the emissions standards—probablycould not have been properly set at this stageof industry development even had therebeen intensive environmental research.However, the difficulties in detecting im-pacts described above, and some doubts asto the availability of environmental researchresources at the Federal level, lead to con-cern about the adequacy of future regula-tion.

4. Past History. –Given both the potential forenvironmental harm and the potential formitigating measures, the attitude and behav-ior of both the industries that will build andoperate synfuels plants and the agencies thatwill regulate them are critical determinantsof actual environmental risk. Consequent-ly, an understanding of the past environmen-tal record of these entities should be a usefulguide in gaging this risk. Unfortunately, thereis little in the way of comprehensive researchon this behavior. Even the compilation ofdata on compliance with existing regulationsand incidence of deaths and injuries is quiteweak.

For example, to our knowledge EPA hassponsored only one major evaluation ofcompliance with emission regulations; thisrecent study of nine States showed that 70percent of all sources failed to comply fullywith those regulations. Also, Department ofLabor statistics on occupational hazards arecompiled in such a way as to overlook healthproblems that cannot easily be attributed toa specific cause—just the kinds of problemsof most concern to an evaluation of poten-tial synfuels problems. Consequently, occu-pational health and safety statistics that ap-pear favorable to synfuels-related industriesare likely to be an inadequate guide to theactual hazard potential.

Anecdotal evidence, although not an ade-quate basis for evaluating risk, may be usefulas a warning signal of future causes of healthand safety problems. For example, recentstudies have demonstrated that protectivegloves used in the chemical industry fail toprotect workers from several hazardous, andcommonly encountered, chemicals. Thispoints to both an immediate technical prob-lem and an institutional failure in the chem-ical industry itself and its regulating agencies.On the other hand, the tests, which weresponsored by OSHA, also demonstrate theability of the regulatory agency to correctpast failures.

Another example of anecdotal evidencethat may indicate some future problems withindustry performance is that some develop-ers have failed to incorporate separate andmeasurable control systems in synfuels pilotplants. For example, a direct-liquefactionfacility in Texas has its effluents mixed withthose of a neighboring refinery, rather thanhaving a separate control system whose ef-fectiveness at treating synfuels wastes can betested and optimized. This might reflect in-dustry’s lack of priority or, more likely, itshigh level of confidence that no unusualcontrol problems will arise that cannot bereadily handled at the commercial stage.There is considerable disagreement aboutthe validity of this confidence.

98 . Increased Automobile Fuel Efficiency and Synthetic Fuels: Alternatives for Reducing Oil lmports

Finally, there is ample evidence that thechemical industry and its regulators have hadsignificant problems in dealing properly withsubtle, slow-acting chemical poisons. Chem-icals that were unregulated or inadequatelyregulated for long periods of time, andwhose subsequent regulation became ma-jor sources of conflict between industry andGovernment, include benzene, formalde-hyde, vinyl chloride, tetraethyl lead, thepesticide 2,4,5-T, and many others. In somecases, controversy persists despite years oreven decades of research.

An implication of the above discussion is thatadequate environmental management of synfuelsis unlikely to occur automatically when develop-ment begins in earnest. Although OTA believesthat the various waste streams can be adequate-ly controlled, this is going to require a strong in-dustry effort to determine the full range of poten-tial environmental impacts associated with devel-

opment and to devise and implement measuresto mitigate or prevent the important impacts. Atpresent, however, there are indications that mostdevelopers are interested primarily in meetingcurrent regulatory requirements, most of whichare limited in their coverage of potential impacts.And completing the regulatory record, to providethe incentive necessary to stimulate further envi-ronmental efforts, is going to be a difficult andtime-consuming job, particularly if ongoing cut-backs in Government research and regulatorybudgets are not accompanied by promised im-provements in efficiency.

Finally, there are some remaining doubts aboutthe reliability of proposed control systems inmeeting current regulatory requirements. Thesepotential problem areas imply that congressionaloversight of an emerging synfuels industry willneed to be especially vigorous in its coverage ofenvironmental concerns.

WILL WATER SUPPLY CONSTRAIN SYNFUELS DEVELOPMENT?In the aggregate, the water consumption re-

quirements for synfuels development are small.Achieving a synfuels production capability of 2million barrels per day oil equivalent would re-quire on the order of 0.3 million acre-feet/yearor about 0.2 percent of estimated total currentnational freshwater consumption, Nevertheless,synfuels plants are individually large water con-sumers. Depending on both the water supplysources chosen for synfuels development and thesize and timing of water demands from otherusers, synfuels development could create con-flicts among users for increasingly scarce watersupplies or exacerbate conflicts in areas that arealready water-short.

The nature and extent of the impacts of syn-fuels development on water availability are con-troversial. The controversy arises in large part be-cause of the many hydrologic as well as institu-tional, legal, political, and economic uncertain-ties and constraints which underlie the data, andbecause of varying assumptions and assessmentmethodologies used. The importance of the fac-

tors influencing water availability will vary in thedifferent river basins where the energy resourcesare located.

In the major Eastern river basins where energyresources are located (i.e., the Ohio, Tennessee,and the Upper Mississippi), water should general-ly be adequate on the mainstems and larger trib-utaries, without new storage, to support likelysynfuels development. However, localized waterscarcity problems couId arise during the inevita-ble dry periods or due to development on smallertributaries. The severity of these local problemscannot be ascertained from existing data and theyhave not yet been examined systematically. Withappropriate water planning and management, itshould be possible to reduce, if not eliminate, anylocal problems that might arise.

Competition for water in the West already ex-ists and is expected to intensify with or withoutsynfuels development. In the Missouri RiverBasin, the magnitude of the institutional, legal,and political uncertainties, together with the need

Ch. 4—Issues and Findings . 99

for major new water storage projects to average-out seasonal and yearly streamflow variations,preclude an unqualified conclusion as to theavailability of water for synfuels development.The major sources of these uncertainties, whichare difficult to quantify because of a lack of sup-porting information, include Federal reserve wa-ter rights (including Indian water rights claims),provisions of existing compacts, and instreamflow reservations.

In the Upper Colorado River Basin, water couldbe made available to support the level of synfuels

development expected over the next decade.However, the institutional, political, and legaluncertainties make it difficult to determine whichsources would be used and the actual amountof water that would be made available from thesesources. The principal uncertainties concern theuse of Federal storage, the transfer of water rights,provisions of existing compacts and treaties, andFederal reserve rights. The range of uncertaintysurrounding the water availability to the entirebasin after 1990 is so broad that it tends to sub-sume the amount of water that would be neededfor expanded synfuels development.

ARE SYNTHETIC FUELS COMPATIBLE WITH EXISTING END USES?

When introducing new fuels into the U.S. liq-uid fuels system, it is important to determine thecompatibility of the new fuels with existing enduses in order to determine what end-use changes,if any, may be necessary. In this section the com-patibility of various synfuels with transportationend uses is briefly described.

Alcohols

Neither pure ethanol nor pure methanol canbe used in existing automobiles without modify-ing the fuel delivery system, but cars using themcan be readily built and engines optimized forpure alcohol use would probably be 10 to 20 per-cent more efficient than their gasoline counter-parts. New cars currently are being built to becompatible with gasohol (1 O percent ethanol, 90percent gasoline), so potential problems with thisblend are likely to disappear with time.8 Metha-nol-gasoline blends have been tested with mixedresults. Principal problems include increasedevaporative emissions and phase separation ofthe fuel in the presence of small amounts ofwater. These problems can be reduced by blend-ing t-butanol (another alcohol) with the methanol,and such a blend is currently being tested.9 How-ever, due to the corrosive effects of methanol on

afnergy From Biological Processes, Volume 11: Technical and fnvi-ronmenta/ Ana/yses, OTA-E-1 28 (Washington, D. C.: U.S. Congress,Office of Technology Assessment, September 1980).

91bid.

some plastics, rubbers, and metals in some vehi-cles, it probably is preferable to use methanol inits pure form in modified vehicles or to requirethat components in new automobiles be compat-ible with methanol blends.

Shale Oil

Shale oil has been successfully refined at thepilot plant level to products that meet refineryspecifications for petroleum derived gasoline,diesel fuel, and jet fuel.10 The properties of thediesel and jet fuels are shown in tables 18 and19, where they are compared with the petroleumcounterparts. Current indications are that the ma-jor question with respect to compatibility of thesefuels with their end uses is what minimum levelof refining (and thus refining cost) will be neededto satisfy the needs of end users.

Direct Coal Liquids

One of the direct coal liquids, SRC II, has alsobeen successfully refined to products that meetrefinery specifications for gasoline and jet fuel(tables 18 and 19). (The cetane number of theresultant “diesel fuel, ” however, is lower thanthat normally required for petroleum diesel fuel.)The gasoline, because of its aromatics content,

‘OR. A. Sullivan and H. A. Frumkin, “Refining and Upgrading ofSynfuels From Coal and Oil Shales by Advanced Catalytic Proc-esses, ” third interim report, prepared for DOE under contract No.EF-76-C-01-2315, Chevron Research Co., Apr. 30, 1980.


Table 18.–Properties of Selected Jet Fuels Derived From Shale Oil and SRC-II

350° to 500° F 300° to 535° FTypical petroleum hydrotreated hydrocracked 300° to 550° F or 250° to 570° F

(Jet A) shale oil shale oil hydrotreated SRC-11

Gravity, ‘API . . . . . . . . . . . . . . 37-51 42 47 33-36Group type, LV%:

Paraffins. . . . . . . . . . . . . . . 35 55 3-4Cycloparaffins. . . . . . . . . . . 45 40 93-81Aromatics . . . . . . . . . . . . . . <20 20 5 4-15

Smoke points, mm. . . . . . . . . > 20 21 35 20-23Freeze point, 0 F . . . . . . . . . . –40 –42 –65 –75 to –95Nitrogen. . . . . . . . . . . . . . . . . (a) 2 0.1 0.1ain addition, severe stability requirements mean that heteroatom content must be very low(usually the nitrogen content is less than 10 ppm for petroleum-derived

jet fuel).

SOURCE: R. A. Sullivan and H. A. Frumkin, “Refining and Upgrading of SYnfuels From Coal and Oil Shales by Advanced Catalytic Processes,” third interim report,prepared for DOE under contract No. EF-76-C-01-2315, Chevron Research Co., Apr. 30, 19S0.

Table 19.–Propertles of Selected Diesal Fuels Derived From Shale Oil and SRC-II

350° to 600° F350° to 650° F hydrotreated shale oil

Typical petroleum hydrotreated shale oil coker distillate 350° F+ hydrotreated SRC-IIGravity, API . . . . > 3 0 38 41 29Cetane No. . . . . . . > 4 0 46a 48 39Pour point, 0 F. . . <+ 15 – 5 –20 –55Group type, LV%:

P . . . . . . . . . . . . . 37 41 –4-7N . . . . . . . . . . . . 44 41 70-93A . . . . . . . . . . . . 19 18 2-23

Nitrogen, ppmb . . (b) 350 350 0.1-0.5aEatlmated.bHeterOatOrnS rnwt be removed to level required for stability (usually 600 Ppm N for petroleum).SOURCE: R. A. Sulllvan and H. A. Frumkln, “Refining and Upgrading of Synfuels From Coal and 011 Shales by Advanced Catalytic Processes, ” third interim report,

prepared for DOE under contract No. EF-76-C-01-2315, Chevron Research Co., Apr. 30, 1960.

would be used as an octane-enhancing blendingagent in conventional gasoline. Aromatics, suchas benzene and toluene, are currently used forthis purpose in gasoline. Again, a principal ques-tion is what minimum level of refining will beneeded. other direct coal liquids probably aresimilar to SRC Ii liquids.

Indirect Coal Liquids

Other than methanol, which was consideredabove, the principal indirect coal liquids forground transportation are gasolines, although theFischer-Tropsch (FT) processes can be arrangedto produce a variety of distillate fuels, as well.There are no indications that these gasolineswould not be compatible with the existing auto-mobile fleet, either alone or in blends with con-ventional gasoline.

Caveat

Despite the apparent compatibility of hydrocar-bon synfuels with existing end uses, refinery spec-ifications do not uniquely determine all of theproperties of the fuel. The tests used to character-ize hydrocarbon fuels were designed for petro-leum products and may be inadequate indicatorsfor the synfuels. Some potential problems withthe hydrocarbon synfuels that have been men-tioned include:

● Lubrication. –Hydrotreating of synfuels isnecessary to meet refinery specifications.However, the lubricating properties of thesynfuels drop with this hydrotreating. Thisdrop in lubricity could lead to possible prob-lems with fuel-injection nozzles and othermoving parts that rely on the fuel for lubri-cat ion.

Ch. 4—issues and Findings ● 1 0 1——-. - — — - — — — .

● Emissions. —The particulate and nitrogen ox-ide emissions of synthetic diesel fuel couldbe greater than those from an otherwisecomparable petroleum diesel fuel. Automo-bile manufacturers are having difficulty meet-ing emissions standards with conventionaldiesel fuel, and there is some concern thatsynfuels could aggravate these problems.

● Variability. —The direct liquefaction synfuelsfrom coal can vary in composition depend-ing on the coal used.11 Consequently, al-though the synfuel from one coal may becompatible with an end use, the same proc-ess might produce an incompatible synfuelif another coal is used.

In principle, if the exact chemical compositionof synfuels were known, synfuels could beblended from petrochemicals and tested exten-sively for these potential problems before synfuelsplants were built. In practice, however, the chem-ical compositions are so complex, varied, andprocess-dependent that this option is probablynot practical.

The alternative is to wait until sufficient quan-tities of synfuels are available and to conduct ex-tensive field tests of synfuels processed in variousways and from different coals. Until this is done,statements about the compatibility of hydrocar-bon synfuels with current end uses are somewhatspeculative.

1 Ibid,

Chapter 5

Increased Automobile Fuel Efficiency

Contents

Page

Status and Trends . . . . . . . . . . . . . . . . . . . 105

Automobile Technologies . . . . . . . . . . . . . 108Vehicle Size and Weight . . . . . . . . . . . . 110Powertrain . . . . . . . . . . . . . . . . . . . . . . . . 111Fuel Economy–A Systems Problem . . . 113Tradeoffs With Safety and Emissions. . . 113Methanol-Fueled Engines . . . . . . . . . . . . 115Battery-Electric and Hybrid Vehicles. . . 117

Future Automobile Fuel Efficiency . . . . . . 120Technology Scenarios . . . . . . . . . . . . . . 121Projection of Automobile Fleet

Fuel Efficiency . . . . . . . . . . . . . . . . . . . 122

Estimated Fuel Consumption, 1985-2000. 126Projected Passenger-Car Fuel

Consumption . . . . . . . . . . . . . . . . . . . . 126Substitution of Electric vehicles . . . . . . 128Fuel Use by Other Transportation

Modes . . . . . . . . . . . . . . . . . . . . . . . . . 129

Page

Costs of Increased Fuel Efficiency . . . . . . 130Overview . . . . . . . . . . . . . . . . . . . . . . . . . 130Fuel Savings Costs . . . . . . . . . . . . . . . . . 137Consumer Costs . . . . . . . . . . . . . . . . . . . 139Electric and Hybrid Vehicles, . . . . . . . . 141MethanoI Engines . . . . . . . . . . . . . . . . . . 142

Appendix A.–Prospective AutomobileFuel Efficiencies . . . . . . . . . . . . . . . . . . 142

Appendix B. –Oil Displacement Potentialof Electric Vehicles . . . . . . . . . . . . . . . 149

TABLES

Table No. Page

20. Potential Battery Systems for Electric(and Hybrid) Vehicles . . . . . . . . . . . . . 118

21. Prospective AutomobileFuel-Efficiency Increases, 1986-2000 . 122

Table No. Page Page

22. Small, Medium, and Large Size 36. Total Domestic Capital InvestmentsClasses Assumed for 1985-2000 . . . . . 123 for Changes Associated With

23. Automobile Characteristics– Increased Fuel Efficiency. . . . . . . . . . . 136High-Estimate Scenario . . . . . . . . . . . . 124 37. Estimated Capital Investment

24. Automobile Characteristics— Allocated to Fuel Efficiency perLow-Estimate Scenario. . . . . . . . . . . . . 124 Gallon of Fuel Saved. . . . . . . . . . . . . . 138

25. Estimated New-Car Fuel Economy: 38. Capital investment Attributed to1985 -2000 . . . . . . . . . . . . . . . . . . . . . . . 125 Increased Fuel Efficiency Plus

26. Effect on Size Mix on Estimated Associated Development Costs perFuel Economy of the New-Car Sales Barrel/Day of Fuel Saved . . . . . . . . . . 139in the United States. . . . . . . . . . . . . . . 126 39. Plausible Consumer Costs for

27. Baseline Assumptions for Projections Increased Automobile Fuel Efficiencyof Automobile Fuel Consumption . . . 126 Using Alternative Assumptions About

28. Effects of Substituting Electric Variable Cost Increase. . . . . . . . . . . . . 140Vehicles . . . . . . . . . . . . . . . . . . . . . . . . 128

29. Projected Petroleum-Based FuelUse for Transportation . . . . . . . . . . . . 129

30. Post-1985 Automotive Capital CostFIGURES

Estimates. . . . . . . . . . . . . . . . . . . . . . . . 132 Figure No. Page

31. Summary of investment 6. Historical Average New-Car FuelRequirements Associated With Efficiency of Cars Sold in the UnitedIncreased Fuel Efficiency. . . . . . . . . . . 134 States . . . . . . . . . . . . . . . . . . . . . . . . . . 105

32. Average Capital Investments 7. Historical Capital Expenditures byAssociated With Increased Fuel U.S. Automobile Manufacturers . . . . . 108Efficiency by Car Size and by 8. Fuel Use in City Portion of EPAScenario . . . . . . . . . . . . . . . . . . . . . . . . 134 Fuel-Efficiency Test Cycle . . . . . . . . . . 109

33. Average Capital Investments 9. Sales-Weighted Average New-CarAssociated With Increased Fuel Fleet Specific Fuel Efficiency. . . . . . . . 125Efficiency by Car Size and by 10. Projected Passenger-Car FuelScenario . . . . . . . . . . . . . . . . . . . . . . . . 135 Consumption . . . . . . . . . . . . . . . . . . . . 127

34. Average Capital Investments 11. Cumulative Oil Savings FromAssociated With Increased Fuel Increased Automobile Fuel Efficiency Efficiency by Car Size and by Relative to 30 MPG in 1985, NoScenario . . . . . . . . . . . . . . . . . . . . . . . . 135 Change Thereafter . . . . . . . . . . . . . . . . 128

35. Percentage of Capital Investments 12. Real and Nominal U.S. Car Prices,Allocated to Fuel Efficiency . . . . . . . . 136 1960-80 . . . . . . . . . . . . . ... , . . . . . . . . 139

Chapter 5

Increased Automobile Fuel Efficiency

STATUS AND TRENDSUntil the 1970’s, fuel economy was seldom im-

portant to American car buyers. Gasoline wascheap and plentiful, taxes on fuel and on car sizelow. In contrast with automobile markets in mostof the rest of the world, automakers had few in-centives to build small cars, or consumers to pur-chase them. The typical American passenger car–large, comfortable, durable–evolved in relativeisolation from design trends and markets in otherparts of the world. Fuel economy was a minorconsideration. (This was not the case for heavytrucks, where fuel costs have always been a sub-stantial component of operating expenses.)

In the late 1960’s and early 1970’s, Federalemissions control standards worked at the ex-pense of fuel economy. But fuel economy pres-sures were also building—signified by gasolineshortages, the sudden rise in oil prices, and thepassage of the Energy Policy and ConservationAct of 1975 (EPCA). EPCA set fleet average mile-age standards for automobiles sold in the UnitedStates beginning in 1978. The combination ofCorporate Average Fuel Economy (CAFE) stand-ards and market forces stimulated rapid changesin the design of American cars, followed by asharp swing in consumer demand during 1979and 1980 toward small, economical imports.

A similar upsurge in small-car demand had fol-lowed the 1973-74 oil shock. Over the 1965-75period, American-made cars averaged about 14to 15 miles per gallon (mpg). * For model year1980, the average for cars sold in the domesticmarket was up to 21 mpg, 1 mpg above the EP-CA requirement. For 1981 models, domestic carssold through January 5, 1981, averaged almost24 mpg, or almost 2 mpg above EPCA require-ments. (See also fig. 6.)

Most of the increases in fuel economy havecome from downsizing—redesigning passengercars so that they are smaller and lighter, and canuse engines of lower horsepower. Other changes

*Based on EPA’s combined test cycles (55 percent city and 45percent highway cycle).

Figure 6.—Historical Average New-Car FuelEfficiency of Cars Sold in the United States

30,

25 -

20 -

15

1 0c

1970 1972 1974 1976 1978 1980 1982

YearSOURCE: J. A. Foster, J. D. Murrell, and S. L. Loaa, “Light Duty Automotive Fuel

Economy. . Trends Through 1981 ,“ SAE Paper 810386, February 1981.

in vehicle designee. g., decreased aerodynamicdrag and rolling resistance, improved automatictransmissions and higher rear axle ratios, elec-tronic engine controls, greater penetration ofdiesels–have also helped to reduce fuel con-sumption.

While the latest technology is used in these re-designs, technology itself is not–and has notbeen–a limiting factor in passenger-car fuel econ-omy in any fundamental sense. The limiting fac-tor is what the manufacturers decide to buildbased on judgments of future consumer demand.Decisions on new models may also be con-strained by the costs of new capital investment.Technology does have a vital role in managingthe many tradeoffs among manufacturing and in-vestment costs, fuel economy, and the other at-tributes that affect consumer preferences—qual-ity, comfort, carrying capacity, drivability, andperformance. Technology is also critical in man-aging possible tradeoffs involving fuel economy,emissions, and safety.

American automakers are still incrementallydownsizing their fleets–in the process convert-

105


ing to front-wheel drive, which helps to preserveinterior space. These redesigns have been large-ly paced by three interrelated factors: 1) CAFEstandards, which require fleet averages of 27.5mpg by 1985; 2) each manufacturer’s estimatesof future market demand in the various sizeclasses (until 1979, market demand for smaller,more fuel-efficient cars lagged behind the CAFEstandards, but it has now outstripped them—sev-eral recent projections point toward fleet averagesof more than 30 mpg by 19851); and, 3) the cap-ital resources of U.S. firms, which affect both theirability to design and develop new small cars andtheir ability to invest in new plant and equipmentfor manufacturing them.

The gradual downsizing of the U.S. automobilefleet has been accompanied by an intensive de-velopmental effort aimed at maximizing the fueleconomy of cars of a given size, consistent withthe need for low pollutant emission levels andoccupant safety—both also matters of Govern-ment policy. Foreign manufacturers, who in 1980accounted for about one-quarter of sales in theUnited States, are also improving the fuel econ-omy of their fleets, but they can concentrate ontechnical improvements rather than new small-car designs because their product lines are al-ready heavily oriented toward cars that are smallin size and low in weight.

Because the U.S. automobile fleet now con-tains over 100 million cars, increases in new-carfuel economy take time to be felt. Typically,about half the cars of a given model year are stillon the road after 10 years; it takes about 17 yearsbefore 99 percent are retired. Thus, while new-car fuel economy for the 1980 model yearreached about 21 mpg, the average for the U.S.fleet in 1981 is still only about 16 to 17 mpg, alegacy of the big cars of earlier years. if new carsaverage 30 to 35 mpg by 1985—a target that iseasily attainable from a technological stand-point–the average fuel efficiency of cars on theroad would reach only about 22 mpg by 1985.While more than half the annual fuel savings asso-

ciated with the 1985 CAFE standards will beachieved by 1985, the full benefit of the 30-mpgnew cars of 1985—and of further improvementsin later years—will not come until the end of thecentury.

Of course, 30-mpg CAFE standards are possibleright now, and 50-mpg cars are currently beingsold. Proportionately higher fuel economy figureswill in principle be attainable in the future, asautomotive technology progresses. But todayonly a portion of consumers want such vehicles—because fuel economy often comes at the ex-pense of comfort, accommodations for passen-gers and luggage, performance, luxury, conven-ience features and accessories, and other attri-butes more commonly found in larger cars–andmanufacturers try to plan their future product mixto appeal to a broad range of consumer tastes.

The sudden shift in market demand towardsmall, fuel-efficient cars in 1974-75, followed bya resurgence in large-car sales during 1976-78 andanother wave of demand for fuel efficiency in1979 and 1980, illustrates the unpredictable na-ture of consumer preferences. The 1979 marketshift has outpaced the CAFE standards. The 27.5-mpg requirement set by EPCA for the 1985 modelyear remains in effect for subsequent years unlessmodified by Congress. If world oil prices againstabilize, and supply exceeds demand—as oc-curred through much of 1981 —the risks and un-certainties facing U.S. automakers could multi-ply, a particularly worrisome situation given theirprecarious financial situations and the large cap-ital outlays necessary for redesign and retooling.Recently, American automakers have been reas-sessing their commitments to rapid downsizingand new small-car lines—both because of cashflow shortfalls and because of uncertainty overfuture market demand.2

The 14-mpg U.S. fleet average of 1975 is a use-ful baseline for estimating recent and near-termfuel savings resulting from the combination ofEPCA standards and market forces. For the period1975-85, OTA estimates that fuel economy in-

IW. G. Agnew, “Automobile Fuel Economy Improvement,” Gen-eral Motors Research Publication GMR-3493, November 1980; TheU.S. Automobile Industry, 1980: Report to the President From theSecretary of Transportation, DOT-P-1 O-81-O2, January 1981. inde-pendent estimates by OTA are similar.

ZJ. Holusha, “Detroit’s Clouded Crystal Ball; Gasoline Glut SpursReview of Small Cars,” New York Times, July 28, 1981, p. Dl;j. Holusha, “For G. M., a Fresh Look at Spending Strategy,” NewYork Times, Nov. 10, 1981, p. D1.

Ch. 5—Increased Automobile Fuel Efficiency • 107

creases in passenger cars alone will have savedslightly more than 0.8 MMB/DOE (million bar-rels per day, oil equivalent) on the average—much less in the earlier years, and about twiceas much near the end of this 10-year period, Con-sidering only passenger cars, the cumulative sav-ing through 1985 will be approximately 3 billionbarrels (bbl) of petroleum–about 75 percent ofthe total U.S. crude oil imports during 1979 and1980. For the period 1985-95, by the end ofwhich the average efficiency of all cars on theroad should be 30 mpg or more, the daily sav-ings would be at least 3 million barrels per day(MMB/D), giving an additional cumulative sav-ing of at least 10 billion bbl compared with a 14-mpg fleet. Thus, for the 20-year period 1975-95,the total savings from increased passenger-car fueleconomy would be over 13 billion bbl—equiva-Ient to 8 years of crude oil imports or about 7years of net petroleum imports at the 1981 rate.

These estimated reductions in petroleum con-sumption could be larger if fuel-economy im-provements proceed faster than assumed. Fuel-economy improvements in trucks, particularlylight and medium trucks, will also save significantamounts. Nonetheless, the U.S. passenger-carfleet would still consume 3.6 MMB/D in 1985 and3 MMB/D in 1995–compared with 4.3 MMB/Din 1980—if the passenger-car fleet grows as ex-pected and cars continue to be driven at the his-torical rate of 10,000 miles per year, on the aver-age. If automobile travel is reduced, fuel con-sumption would be decreased proportionately.

The savings in petroleum consumption–whichrepresent a direct benefit to consumers as wellas an indirect benefit because of the expectedimprovement in the U.S. balance of payments—also carry costs. These will generally take the formof higher purchase prices for new cars, eventhough these cars will be smaller. Costs will behigher because the redesign and retooling for adownsized U.S. fleet requires capital spendingat rates significantly higher than the historicalaverage for American manufacturers. Increasedcapital spending—which, along with the salesslump of 1979-81, shares responsibility for theover $4 billion lost by U.S. automakers during1980–is passed along at least in part to purchas-

ers. To the extent that competitive forces allow,importers will also raise prices even though theircapital spending rates may not have gone up.

Many of the technological roads to improvedfuel economy also carry higher direct manufac-turing costs. A familiar example is the diesel en-gine—which, for comparable performance, canincrease passenger car fuel economy, and de-crease operating costs, by as much as 25 percent,but at a substantial penalty in purchase price. Inthis case, the higher costs stem largely from anintrinsically expensive fuel injection system, butalso from the greater mechanical strength andbulk required in a diesel engine. Beyond eco-nomic costs and benefits, smaller cars cannot bedesigned to be as safe as larger cars (given bestpractice design in both) –thus, risks of death andinjury in collisions could go up.

For the 10-year period 1968-77, average annualcapital investment by the big three U.S. automak-ers in constant 1980 dollars was $6.68 billions(AMC and, in later years, Volkswagen of Americaadd only small amounts to these averages). Overthis period, production fluctuated considerably,but with only a slight upward trend; thus, theaverage expenditure is primarily that for normalredesign and retooling as new models are intro-duced and existing product lines updated, ratherthan for increases in production capacity. Notethat the period 1968-77 includes investments as-sociated with the introductions of several newsmall cars around 1970 (Pinto, Maverick, Vega),as well as later subcompact designs (Chevette,Omni/Horizon). The figures also include overseasinvestments by the three U.S. firms.

The 2 years with the highest investments dur-ing the 1968-77 period were 1970 ($7.67 billion)and 1977 ($7,78 billion). In 1978, investment roseto $9.21 billion, and in 1979 it reached $10.5 bil-lion (still in 1980 dollars) –half again as much asthe historical level. (See fig. 7.) Estimates of invest-ment for the 5-year period 1980-84 reach close

JThese investment figures were tabulated from annual reports by

R. A. Leone, W. J. Abernathy, S. P. Bradley, and J. A. Hunker, “Reg-ulation and Technological Innovation in the Autombile Industry, ”report to OTA under contract No. 933-3800.0, May 1980, pp. 2-92.Conversions to 1980 dollars are based on the implicit price defla-tor for nonresidential fixed domestic investment.

708 ● Increased Automobile Fuel Efficiency and Synthetic fuels: Alternatives for Reducing Oil Imports

Figure 7.—Historical Capital Expenditures byU.S. Automobile Manufacturers

SOURCE: G. Kulp, D. B Shonka, and M. C, Holcomb, “Transportation EnergyConservation Data Book: Edition 5,” Oak Ridge National Laboratory,ORNL-5765, November 1981.

to $60 billion.4 5 Such estimates have generallybeen based on fleet redesigns to meet EPCA re-quirements through 1985. While it is doubtful that

Weneral Accounting Office, “Producing More Fuel-Efficient Auto-mobiles: A Costly Proposition, ” CED-82-14, Jan. 19, 1982.

‘Fuel Economy Standards for New Passenger Cars After 1985(Washington, D. C.: Congressional Budget Office, December 1980),p. 56. Other estimates are generally comparable but may differ asto whether development costs or only fixed investment are included.

they include much spending for new models tobe introduced in the immediate post-1985 period,they do include substantial overseas expenditures—perhaps one-quarter or more.

The $60 billion estimate represents about $12billion per year in 1980 dollars, nearly double thehistorical spending level. It is a clear indicationof the market pressures (for smaller cars as wellas for greater fuel economy in vehicles of all sizes,including light trucks) being placed on domesticmanufacturers. Component suppliers also facehigher-than-normal investments.

The remainder of this chapter treats the factorsthat determine automobile fuel consumption inmore detail—both the technological factors andmarket demand—as well as the net savings in fuelconsumption that may accrue from increases inautomobile fuel economy. While consumer pref-erences—as judged by manufacturers in planningtheir future product lines–are dominant, technol-ogy is vitaI in maximizing the fuel economy thatcan be achieved by cars of given size and weight,as well as given levels of performance, emissions,and occupant safety.

AUTOMOBILE TECHNOLOGIES

Fuel consumed by an automobile (or truck) de-pends, first, on the work (or power) expendedto move the vehicle (and its passengers and car-go), and second, on the efficiency with which theenergy contained in the fuel (gasoline, diesel fuel)is converted to work. The power requirementsdepend, in essence, on: 1) the driving cycle–the pattern of acceleration, steady-state opera-tion, coasting, and braking–which is affected bytraffic and terrain, but otherwise controlled bythe driver; 2) the weight and rolling resistance ofthe vehicle; and 3) the aerodynamic drag, whichdepends on both the size and shape of the vehi-cle and is also a function of speed. The designercontrols weight, aerodynamic drag, and to someextent the rolling resistance, but can affect the

driving cycle only indirectly –e.g., through thepower output and gear ratios available to thedriver.

The fuel consumed in producing the power tomove a vehicle of given size and weight is againa function of vehicle design, primarily engine de-sign. The efficiency with which the engine con-verts the energy in the fuel to useful work de-pends on the type of engine as well as its detaildesign; diesel engines are more efficient thanspark-ignition (gasoline) engines, but not all dieselengines have the same efficiency.

Furthermore, an engine’s efficiency varies withload. For example, when a car is idling at a traf-fic light, the engine’s efficiency is virtually zero

Ch. 5—Increased Automobile Fuel Efficiency ● 109

because the energy in the fuel is being used onlyto overcome the internal friction of the engineand to power accessories. Engines are more effi-cient at relatively high loads (accelerating, driv-ing fast, or climbing hills), but such operation willstill use fuel at a high rate simply because thepower demands are high–hence the justificationfor the 55-mph speed limit as an energy-conserva-tion measure.

Efficiency–the fraction of the fuel energy thatcan be converted to useful work—cannot be 100percent in a heat engine for both theoretical andpractical reasons. For a typical automobile en-gine, peak efficiency may exceed 30 percent, butthis is attained for only a single combination ofload and speed. Average efficiencies, character-istic of normal driving, may be only 12 or 13 per-cent, even lower under cold-start and warmupoperation. To illustrate this point, figure 8 showsenergy losses for the drivetrain in one 1977 mod-el car in the Environmental Protection Agency’s(EPA) urban cycle. This figure should not be taken

too literally because losses vary considerably fromcar to car and numerous design changes havebeen implemented since 1977, but the figuredoes serve to illustrate approximate magnitudes.

The design of the engine affects the amount offuel consumed during the driving cycle in twobasic ways. First, the size of the engine fixes itsmaximum power output. In general, a smaller en-gine in a car of given size and weight will givebetter fuel economy, mainly because the smallerengine will, on the average, be operating moreheavily loaded, hence in a more efficient part ofits range, There are practical limits to enginedownsizing, however, because a heavily loaded,underpowered engine provides poor perform-ance and can suffer poor durability.

Second, the designer can directly affect driving-cycle fuel economy through the transmission andaxle interposed between the engine and wheels.Significant gains in fuel economy over the pastfew years have come from decreases in rear axle

Figure 8.—Fuel Use in City Portion of EPA Fuel’ Efficiency Test Cycle (2,750 lb/2.3 liter)

Alternator

Fan0.7%

Coolant42.3%

0.5%mExhaust

r 22%

/Friction

1 1 2 . 2 %

I

Heat a x l e0.8%

P o w e r steering 1

“ “ ” ” Delivered to wheels

11.5%0

SOURCE Serge Gratch, Ford Motor Co., prwate communication, 1982.


ratios* so that engines are operating at higher effi-ciencies during highway operation, and fromchanges in transmission ratios** to better matchdriving needs to engine efficiency (in earlier yearstransmission ratios were often chosen to maxi-mize performance rather than fuel economy).Adding more speeds to the transmission—wheth-er manual or automatic—serves the same objec-tive. The optimum would be a continuously vari-able, stepless transmission allowing the engineto operate at all times at high loads where its ef-ficiency is greatest. (Such transmissions can bebuilt today, but require further development forwidespread use in cars.)

Changes in many other areas of automobile de-sign can help increase fuel economy, but the pri-mary factors are size (which is one of the fac-tors that determines aerodynamic drag), weight(which determines the power needed for acceler-ation, as well as rolling resistance), and power-train characteristics (engine plus transmission).These are discussed in more detail below, in thecontext of the driving cycle—itself a critical vari-able in fuel economy—followed by brief discus-sions of emissions and safety tradeoffs, methanol-fueled vehicles, and electric vehicles (EVs).

Vehicle Size and Weight

On a sales-averaged basis, the inertia weightof cars sold in the United States during 1976 (in-cluding imports) came to slightly over 4,000 lb.6

This corresponds to an average curb weight***of 3,700 to 3,800 lb. The average inertia weightfor 1981 is expected to be about 3,100 lb, andmay further decrease to around 2,750 lb by 1985.Although the lightest cars sold here still have iner-tia weights close to 2,000 lb–as they did in 1975—the distribution has shifted markedly toward thelower end of the range. Many heavier models

*Rear axle ratio is the ratio of the drive shaft speed to the axlespeed.

**Transmission ratio is the ratio of the engine crankshaft speedto the drive shaft speed.

6J. A. Foster, j. D. Murrell, and S. L. Loos, “Light Duty Automo-tive Fuel Economy . . . Trends Through 1981 ,“ Society of Automo-tive Engineers Paper 810386, February 1981. Inertia weight is a representative loaded weight—equal to curb weight, which includesfuel but not passengers or luggage, plus about 300 lb–used by EPAfor fuel economy testing.

***Curb weight is the weight of the car with no passengers orcargo.

have disappeared; consumers are now selectingsmaller and lighter vehicles—downsized or newlydesigned U.S. models as well as imports.

While size is a primary determinant of weight,newer designs typically make greater use of light-weight materials such as plastics and aluminumalloys, as well as substituting higher strengthsteels—in thinner sections—for the traditionalsteels. Materials substitution for weight reductionwill continue, but is constrained by the highercosts of materials with better strength-to-weightor stiffness-to-weight ratios. As production vol-umes go up, costs of at least some of these mate-rials will tend to decline.

Weight is a fundamental factor in fuel econ-omy because much of the work, hence energy,needed for a typical driving cycle is expendedin accelerating the vehicle. The fuel consumedin stop-and-go driving is directly related to theloaded weight of the car (including passengersand payload) and the inertia of its rotating parts.Everything else being the same, it takes twice asmuch energy to accelerate a 4,000-Ib car as a2,000-lb car over the same speed range. Urbandriving, in particular, consists largely of repeatedaccelerations and decelerations; thus, weight iscritical to fuel consumption. (This also points upthe potential that smoothing flows of traffic of-fers for gasoline savings.) Lighter cars consumeless fuel even at constant speeds because theyhave less rolling resistance.

Although the weights of cars can be reducedby making them from lightweight materials andby shifting from separate body and frame designsto unitized construction, cars can always be madelighter by making them smaller.

Small cars can also have lower aerodynamicdrag, because drag depends on frontal area aswell as on the shape of the vehicle. Drag can bereduced by making cars lower and narrower, aswell as by streamlining the vehicle. Drag reduc-tion has become at least as important as stylingin recent years; working primarily with wind tun-nel data, automakers have reduced typical dragcoefficients* from 0.5 to 0.6, characteristic of the

*The drag coefficient is a measure of how aerodynamically “slip-pery” the car is. The aerodynamic drag is proportional to the dragcoefficient, the frontal area and the velocity squared.


Photo credit Genera/ Motors Corp

The shape of this experimental car is designed for low aerodynamic drag

early 1970’s, to 0.4 to 0.5 at present, with a num-ber of models being under 0.4. Values in therange of 0.35 will eventually be common.

It takes only 15 or 20 horsepower to propel atypical midsized car at a steady 55 mph—that isall that is needed to overcome rolling resistanceand aerodynamic drag. The remainder of the en-gine’s rated power is used for acceleration, climb-ing hills, and other demands. The low power re-quirements for constant-speed driving–typically15 to 20 percent of the engine rating—emphasizethe importance of the fuel used in start-and-stopdriving (and the influence of weight) in determin-ing driving-cycle fuel economy.

Powertrain

The engine (or other vehicle prime mover) con-verts the energy stored in fuel (or, for an EV, inbatteries) to mechanical work for driving thewheels. The efficiency of the engine—as well asthe efficiency of the transmission—determines theproportions of the energy in the fuel which are,respectively, used in moving the car and lost as

waste heat. Under most operating conditions,transmissions are much more efficient than theengine.

The efficiency-work output divided by energyinput—of any engine depends on both detail de-sign and fundamental thermodynamic limitations.The temperatures at which the engine operatesplace practical constraints on the efficiencies ofsome types of engines—e.g., gas turbines—butnot on others—e.g., spark-ignition (SI) (gasoline)and compression-ignition (Cl) (diesel) engineswhere the combustion process is intermittent.The components of the latter need not withstandtemperatures as high as those where combustionis continuous.

Many other factors besides efficiency enter intothe choice of engine for a motor vehicle; untilrecently, efficiency was often of secondary impor-tance. Cars and trucks have been powered bygasoline or diesel engines because these have fa-vorable combinations of low cost, compact size,light weight, and acceptable fuel economy. Nei-ther demands for improvements in exhaust em is-


sions nor for better fuel economy have yet re-sulted in serious challenge to these engines—which have been dominant for 70 years. At leastthrough the end of the century, most passengercars are likely to be powered by reciprocating SIor diesel engines.

SI and Cl or diesel engines have peak efficien-cies generally in the range of 30 to 40 percent.However, their efficiency at part-load can bemuch less; the farther the engine operates fromthe load and speed for which its efficiency isgreatest, the lower the efficiency. In typical ur-ban driving, the average operating efficiency isless than one-third of the peak value–e.g., in therange of 10 to 15 percent.

Part-load fuel economy remains a more criticalvariable for an automobile engine than maximumefficiency because of the light loading typical ofmost driving. Such a requirement favors Cl en-gines, for example, but works against gas tur-bines. Cl engines have good part-load efficiencybecause they operate unthrottled, thereby avoid-ing pumping losses. They also have high com-pression ratios–which, up to a point, raises effi-ciency under all operating conditions.

Various modifications to SI engines can in-crease part-load efficiencies. This is one of theadvantages of stratified-charge engines—whichuse a heterogeneous fuel-air mixture to allowoverall lean operation, ideally without throttlingas in a diesel—and also of SI engines that burnalternative fuels such as alcohol or hydrogen.Smaller, more heavily loaded SI engines also tendto have greater driving-cycle fuel economy be-cause the higher loads mean the engine is run-ning with less throttling. Among the steps that canbe and are being taken to give greater fuel econ-omy are:

●

●

●

●

using the highest compression ratio consist-ent with available fuels;refined combustion chamber designs, partic-ularly those optimized for rapid burning oflean mixtures, one of the routes to highercompression ratios;minimizing engine friction;optimizing spark timing consistent with emis-sions control;

●

●

●

�

minimizing exhaust gas recirculation consist-ent with emissions control;precise control of fuel-air ratio, both overalland cylinder-to-cylinder, particularly undertransient conditions such as cold starts andacceleration—again consistent with emis-sions control; andminimizing heat losses.

Transmission efficiencies also depend on load,but much less so than engines; transmission effi-ciencies are also much higher in absolute terms.For manual transmissions, more than 90 percentof the input power reaches the output shaft ex-cept at quite low loads. Because they have moresources of losses, automatic transmissions are lessefficient, particularly those without a lockuptorque converter or split-path feature. In theseolder transmissions, all the power passes throughthe torque converter, even at highway cruisingspeeds. The resulting fuel-economy penalty, com-pared with a properly utilized manual transmis-sion, is typically in the range of 10 to 15 percent.By avoiding the losses from converter slippageat higher speeds, split-path or lockup designs cutthis fuel-economy penalty approximately in half,four-speed transmissions offering greater im-provement than those with only three forwardgears.

One function of the transmission is to keep theengine operating where it is reasonably efficient.Although automatic transmissions are less effi-cient than manual designs, they can sometimesincrease overall vehicle efficiency by being“smarter” than the driver in shifting gears. Fur-ther benefits are promised by improved electron-ic control systems for automatics. These micro-processor-based systems can sense a greaternumber of operating parameters, and are thusable to use more complex logic, perhaps in con-junction with an engine performance map storedin memory. Such control systems could also beused with manual transmissions—e.g., to tell thedriver when to shift. A continuously variabletransmission would be better still. As mentionedabove, these can be built now, but they have notbeen practical because of problems such as highmanufacturing cost, low efficiency, noise, andlimited torque capacity and durability.

Ch. 5—increased Automobile Fuel Efficiency ● 113

Fuel Economy—A Systems Problem

An automobile is a complex system; design im-provements at many points can improve fueleconomy. Even if each incremental improvementis small, the cumulative effect can be a big in-crease in mileage. Interactions among the ele-ments of the system (engine, transmission, vehi-cle weight) and the intended use of the vehicleare among the keys to greater system efficiency.At the same time, as the state of the art improves,further increases in efficiency tend to becomemore difficult unless there are dramatic technicalbreakthroughs–and OTA thinks such break-throughs are improbable. This is a mature tech-nology in which, as a general rule, radicalchanges are few and far between.

Making cars smaller and lighter helps in manyways to reduce the power needed, hence fuelconsumed. Front-wheel drive preserves interiorspace while allowing exterior size and weight tobe reduced. Reductions in the weight of the bodystructure mean that a smaller engine will giveequivalent performance, while also allowing light-er chassis and suspension members, smaller tiresand brakes, and related secondary weight sav-ings. Among other steps taken in recent yearshave been the adoption of thinner, hence lighter,window glass—and even redesigned window liftmechanisms.

Once major decisions have been made con-cerning overall vehicle design parameters—size,engine type, etc.—subsystem refinement and sys-tems integration become the determining factorsin the fuel economy achieved in everyday driv-ing. Some of these refinements decrease the needfor power, as by reducing friction or making ac-cessories more efficient; others increase the effi-ciency of energy conversion, as by using three-way exhaust catalysts and feedback control of thefuel-air ratio to limit emissions while preservingfuel economy.

Tradeoffs With Safety and Emissions

Government policies to increase automobilefuel efficiency, reduce pollutant emission levels,and improve passenger safety involve significanttradeoffs. Measures to control auto emissions can

impair fuel efficiency. Reducing the size of carsto increase fuel economy makes them intrinsicallyless safe. Meeting regulatory goals also affectsmanufacturing costs. Tradeoffs like these have notalways been fuIly recognized in the formulationof Federal policy,7 but will continue to be impor-tant as policy makers focus on questions of post-1985 fuel economy.

The issues include whether Government poli-cies are to be directed at further improvementsin mileage, such as by a continued increase inCAFE standards, or by a gasoline tax, and whetheremissions standards are to be tightened or re-laxed. The tradeoffs will involve manufacturingcosts, as always—but the relationship of fueleconomy to safety will perhaps be most critical.

Safety

The tradeoffs between fuel economy and occu-pant safety are largely functions of vehicle size—therefore of weight as well. Although many char-acteristics of the car are important for occupantsafety, protection in serious collisions dependsquite substantially on the crush space in the vehi-cle structure and on the room available withinthe passenger compartment for deceleration.Penetration resistance is also vital. Design re-quirements are based on a “first collision” be-tween vehicle and obstacle, and a “second colli-sion” between occupants and vehicle. In the“first collision,” the more space available for thestructure to crush—without encroaching on thepassenger compartment-the slower the averagerate of deceleration that the passenger compart-ment and the passengers experience. More crushspace translates directly to lower decelerations.

Space, hence vehicle size, is also importantwithin the passenger compartment. The morespace available inside, the easier it is to preservethe basic integrity of the structure and the slowerthe occupants can be decelerated during the“second collision. ” Seatbelts, for example, canstretch to lower the decelerations the restrainedoccupants experience, but only if there is nothing

7U.S. Industrial Competitiveness. A Comparison of Steel Elec-tronics, and Automobiles, OTA-ISC-135 (Washington, D. C.: U.S.Congress, Office of Technology Assessment, June 1981), pp.120-122.


rigid for the occupants to hit; deformable anchorsfor seatbelts are thus a further method for improv-ing safety.

Because of their larger crush space and interiorvolume, big cars can always provide more pro-tection for their occupants in a collision—givenbest practice design. However, not all cars em-body best practice designs, and the crashworthi-ness of autos in the current fleet does not improveuniformly and predictably with vehicle size. Fur-thermore, vehicle safety depends on avoidingcrashes as well as surviving them, and thereforeon factors such as braking and handling as wellas driver ability. These and other factors relatedto the potential effects on auto safety of increas-ing fuel efficiency are discussed in chapter 10.

Emissions Control

Fairly direct tradeoffs exist between engine effi-ciency and several of the measures that can beused to control the constituents of exhaust gasesthat contribute to air pollution. The three majorcontributors in the exhaust of gasoline-fueled ve-hicles, all regulated by the Clean Air Act and itsamendments, are hydrocarbons (HC), carbonmonoxide (CO), and nitrogen oxides (NOX).8Emissions control measures have frequentlyworked against the fuel economy of cars sold inthe United States since 1968, when manufactur-ers began to retard spark timing to reduce HCemissions. Although the costs of emissions con-trol measures—as reflected in the purchase priceof the vehicle—have often been disputed and re-main controversial, there have also been operat-ing cost penalties because fuel economy has beenless than it would otherwise have been.

Mileage penalties were more severe in the mid-1970’s than at present, but efficiency increaseshave come at the expense of higher first cost.Ground is periodically lost and regained, buteven with best practice technology at any giventime, the engineering problems of balancingemissions and fuel economy at reasonable costhave forced many compromises. One recent esti-mate of the net effect of emissions control

through 1981 finds a 7.5-percent fuel-economypenalty. 9

“The single change with the greatest continuingeffect has been reduced compression ratios re-sulting from the changeover to unleaded gaso-line. Cl engines require high compression ratios;SI engines, in contrast, suffer from a form of com-bustion instability termed detonation (i.e., the en-gine “knocks”) if the compression ratio is toohigh for the octane rating of the fuel. Thus, de-creases in the already lower compression ratiosof SI engines—to values in the range of 8:1 ver-sus ratios as high as 10:1 in the early 1970’s—haveled to significant decreases in fuel economy.

Lead compounds, formerly added to gasolineto raise the octane, have been removed to pre-vent poisoning (deactivation) of catalytic convert-ers—themselves adopted to control, first, HC andCO, and later NOX as well–and also because ofconcern over the health effects of lead com-pounds. While electronic engine control systems,including knock detectors, have allowed com-pression ratios to be increased somewhat, onlya portion of the ground lost can be regained inthis way.

Methanol with cosolvents can be used as anoctane-boosting additive to gasoline that doesnot interfere with the catalytic converter. In addi-tion, compact, fast-burn combustion chambersmay help. By burning the fuel fast enough thatthe preflame reactions leading to detonation donot have time to occur, fast-burn combustion sys-tems might allow compression ratios to be in-creased by several points. This latter approach,however, increases HC and NOX emissions, andit is not yet clear how much compression ratioscan be raised while maintaining emissions withinprescribed limits.

Related measures used to control emissions—and/or to limit detonation—can also degrade en-gine efficiency. Retarding ignition timing—to limitdetonation, and in some cases help control HCand CO emissions by promoting complete com-bustion of the fuel–hurts fuel economy. Othertechniques adopted in the early 1970’s to con-

6D. j. Patterson and N. A. Henein, Emissions From CombustionEngines and Their Control (Ann Arbor, Mich.: Ann Arbor SciencePublishers, 1972).

‘L. B. Lave, “Conflicting Objectives in Regulating the Automo-bile,” Science, May 22, 1981, p. 893.


trol HC and CO—such as thermal reactors, whichwere likewise intended to drive the combustionprocess toward complete reaction of HC and CO—also led to increased fuel consumption.

Unfortunately, while more complete combus-tion decreases HC and CO pollutants, NOX in theexhaust is increased under conditions leading tomore complete combustion. Thus, not only doescontrol of exhaust emissions conflict with fueleconomy, but there are also potential conflictsbetween control of HC and CO on the one hand,and NOX on the other. The NOX standards of themid-1 970’s could be met by adding exhaust gasrecircuIation (EGR) to the repertoire of measuresused for HC and CO control. Although EGR ini-tially carried a substantial fuel economy penalty,and also impaired driveability, improved controlsystems—which recirculate exhaust only whenneeded—have improved both economy anddriveability substantially. Still, the drawbacks ofother methods of NOX control, * together withthe more stringent NOX standards of later years,have led to the most common current controltechnique—the three-way catalytic converter,which reduces levels of all three pollutants. Thisgives fuel economy comparable to an uncon-trolled engine, though at higher first cost.

Compression ratios of diesel engines are oftentwice those for SI engines; at these high levelssmall changes in compression ratio have relativelylittle effect on efficiency. For this reason and be-cause of the different set of emissions standardsapplied, Cl engines have faced fewer conflicts be-tween emissions control and fuel economy. Thisadvantage has helped them to compete with SIengines for passenger cars, although the situationmay change in the future, as the diesel standardsbecome tougher. Particulate (bits of unburned,charred fuel) are the most difficult of diesel emis-sions to control, although NOX also poses prob-lems. However, future regulations for particulatein diesel exhaust are not yet definite. This createsuncertainty not only about the control technol-ogies that might be needed, but also about the

*It should be noted, however, that burning a very lean fuel/airmixture also reduces NOX emissions substantially. Because metha-nol has considerably wider flammability limits than does gasoline,the use of methanol opens new opportunities for controlling NOX.

future penetration of Cl engines in passengervehicles.

Nonetheless, as will be seen below, OTA re-mains cautiously optimistic about diesels. Theirhigher intrinsic efficiency at both full- and part-Ioad makes them quite attractive in terms of fueleconomy, and there is considerable scope for fur-ther improvements in their driveability and re-lated characteristics that are more important forpassenger cars than for trucks and other uses inwhich diesels have been more common. Thelong developmental history of Cl engines pro-vides a useful foundation for passenger-car appli-cations.

Methanol-Fueled Engines

There are basically two routes to higher SI en-gine efficiency via alternate fuels: lean operation,which cuts pumping and other thermodynamiclosses, and higher compression ratio.10 Fuels varyin the extent to which these factors operate andthere are a number of secondary effects, but alco-hols and hydrogen have excellent potential forboth lean burning and higher compression ratios,with possible driving-cycle economy improve-ments in the range of 10 to 20 percent. Furtherengineering development—but no breakthroughs—would be needed before alcohols or other alter-nate fuels could be used in U.S. cars, but the pro-duction and distribution of such fuels are moresignificant barriers.

Diesels, like SI engines, can operate on a varietyof alternative fuels, although perhaps needingspark-assisted combustion. Powerplants such asopen-chamber stratified-charge engines and con-tinuous combustion engines can often toleratequite broad ranges of fuels with minimal designchanges.

Because methanol from coal is an attractivesynthetic fuel, methanol-burning engines for pas-senger cars are discussed in more detail below.Unlike ethanol, which will probably be used pri-marily as a gasoline extender (e.g., in gasohol),

10J. A. Alic, “Lean-Burning Spark Ignition Engines–An Overwew,”Proceedings, 2nd Annual UMR-MECConference on Energy, Rolla,Me., Oct. 7-9, 1975, p. 143.

sufficient quantities of methanol could be pro-duced to consider using it as the only or principalfuel for some automobiles.

If methanol-fueled engines receive intensive de-velopment aimed at maximizing fuel economyand driveability, driving-cycle fuel-efficiency im-provements (on a Btu basis) of 20 percent or moreshould be possible, compared with a well-devel-oped but otherwise conventional SI engine burn-ing gasoline. Most of the improvement stems fromthe higher octane rating of methanol, whichwould permit compression ratios in the range of11 or 12:1—perhaps even higher, depending onwhether preignition is a serious limiting factor—aswell as the somewhat leaner air-fuel ratios pos-sible.

The engineering of vehicles to run on methanol—or other alchohols—is rather straightforward.11

Indeed, a good deal of experience has alreadybeen accumulated. Despite the greater efficien-cy possible with methanol, vehicles fueled with

11’’CH30H: Fuel of the Future?” Automotive Engineering,December 1977, p. 48.

it probably will require larger fuel tanks to achieveacceptable cruising ranges, because methanolhas significantly less energy per gallon than gaso-line or diesel fuel. Methanol corrodes some ofthe materials commonly used in gasoline-fuel sys-tems, which must be replaced by more corrosion-resistant components.

Because alcohols have much higher heats ofvaporization than gasoline and therefore do notvaporize as easily, alcohol-fueled engines aremore difficult to start in cold weather. Driveabilityduring warmup also tends to be poor. Fuel injec-tion is one approach to mitigating such difficul-ties. Another solution is to start and warm up theengine on a different fuel. In Brazil, where manycars and trucks run on 100 percent ethanol, en-gines are typically started on gasoline via an aux-iliary fuel system. A lower cost alternative mightbe to blend in a small fraction–5 to 10 percent–of a hydrocarbon to aid in starting and warmup.Fuel blends could be tailored seasonally just asgasolines are.

Methanol also offers advantages in reducingheat losses and thus raising fuel efficiency. Al-


though its high specific heat and high heat ofvaporization can cause starting and warmupproblems, these characteristics also mean that thefuel can, in principle, be used to help control in-ternal engine temperatures and heat flows so asto reduce heat losses.

Test programs with alcohol fuels have some-times shown abnormally high engine wear—par-ticularly piston ring and bore wear. While thecauses have not yet been fully determined, corro-sion, perhaps associated with wall-washing andcrankcase dilution during cold-start, are possiblecontributing factors.12 If this is the case, solvingthe cold-start and warmup difficulties would alsobe expected to cut down on wear. Oil additivepackages specially tailored for alcohols shouldbe a further help.

Emissions from methanol-burning automobilescan be controlled with many of the same technol-ogies used for gasoline engines. However, be-cause of the differing fuel chemistries, standardsdeveloped for gasoline-burning vehicles are notnecessarily appropriate for alcohols. Aldehydes,for example, may need to be controlled.

Battery-Electric and Hybrid Vehicles

The automobile powerplants considered byOTA for increased fuel efficiency are all heat en-gines–i.e., they convert the energy (heat) pro-duced when a fuel burns into mechanical work.Passenger cars can also be powered from energystored in forms other than fuel—e.g., by mechan-ical energy drawn from a spinning flywheel.Among these alternative storage media are re-chargeable batteries that convert chemical energyinto electrical energy. The electric energy canthen drive a direct current (DC) (or sometimes,through an inverter, an alternating current (AC))motor. Many of the first automobiles built, aroundthe turn of the century, used battery-electric pow-er.

In an extension of the battery-electric concept—called a hybrid—a conventional heat engine

12T. w. RYan, III, D. w. Naegeli, E. C. OWens, H. w. MarbaChtand J, G. Barbee, “The Mechanism Lending to Increased CylinderBore and Ring Wear in Methanol-Fueled S.1. Engines, ” Society ofAutomotive Engineers Paper 811200, 1981.

drives a generator (or alternator) which can thensupply power to an electric motor directly, chargebatteries, or both–depending on the instantane-ous needs of the driving cycle. A parallel hybridis designed so the heat engine can also powerthe wheels directly, through a transmission (theengine turns the generator and the drive wheelsin parallel). A series hybrid, in contrast, has nodirect mechanical connection between heat en-gine and drive wheels. Diesel-electric submarinesprovide examples of both series and parallelhybrid powertrains, but automobiles have neverbeen mass produced with either arrangement.

Whether or not a battery-electric or hybridautomobile would have an overall energy conver-sion efficiency greater or less than a more con-ventional SI- or Cl-powered car depends on manyvariables, including the sources of the electrici-ty used to charge the batteries. In the context ofthis report, the potential of battery-electric or hy-brid vehicles as substitutes for petroleum-basedfuels is more important than the net energy con-version efficiency. If the electricity for chargingthe batteries comes from a coal or nuclear gener-ating plant—or any other nonpetroleum energysource—widespread sales of such cars could helpconserve liquid petroleum.

At present, however, the limitations of practicalelectric and hybrid vehicles far outweigh any ad-vantages that might be gained from their petro-leum-displacing effects.13 Battery-electric cars willhave very limited applications until the perform-ance of batteries (as measured, for example, bythe quantity of energy that can be stored per unitof battery weight), increases roughly fivefold—or unless petroleum availability declines muchmore rapidly than now expected. Hybrids sharemany of the disadvantages of battery-electric carsand—although they offer theoretically promisingenergy conversion efficiencies—are dependenton fuels. Their current prospects are even dim-mer than for battery-electrics, in large part be-cause hybrid vehicles would be expensive andcomplex—the duplication in the powertrain is aformidable cost barrier.

13R. L, Graves, C. D, West, and E. C. Fox, “The Electric car—is

It Still the Vehicle of the Future, ” Oak Ridge National LaboratoryReport ORNLITM-7904, August 1981.


Despite the widespread publicity given to bat-tery-electric automobiles over the past 20 years—most recently, the attention garnered by GeneralMotors’ announcement of production plans forthe mid-1980’s—progress in EVs remains severelylimited by battery performance. This is true bothfor battery systems that are available now and forthose that appear to have possibilities for near-term production. Power and energy densities ofavailable batteries remain too low for practicaluse in other than highly specialized automotiveapplications. Power density (watt per pound orW/lb) measures the rate at which the battery cansupply energy. Energy density (watt hours perpound or Whr/lb) measures the total amount ofenergy that can be stored and then withdrawnfrom the battery. For some battery systems, to getall the energy out requires a slow rate of with-drawal, limiting the instantaneous delivery ofpower. Because of the transient demands of auto-motive driving cycles, power density is almost asimportant for vehicle applications as energy den-sity, which determines the operating range beforethe batteries need to be recharged.

In general, battery systems that are near-termcandidates for automotive applications sufferfrom both low power density and low energydensity. Table 20, which includes several batteriesthat are still in rather early stages of development,gives typical values. Energy and power densitytend to be inversely related, a particular problemfor the familiar lead-acid battery; the inverse rela-tion means that—for any given battery system—the designer can choose higher energy densityonly at the sacrifice of power density. Limitedpower density restricts the acceleration capabil-ities of current EVs to low levels—for some driv-ing conditions, to the detriment of safety. The en-

Photo credit: Electric Vehicle Council

A view of the battew-pack configuration in ademonstration electric vehicle

ergy density, in contrast, limits the total amountof energy that can be carried, therefore, the rangeof the vehicle before the batteries must be re-charged. Recharging is a time-consuming proc-ess—as much as 10 hours for some, though notall, batteries. If power density and energy densi-ty are low, then the vehicle must carry more bat-teries. This makes it heavier, increasing the de-mands for power and energy and compoundingthe design problems.

As a rule-of-thumb, and assuming reasonablecosts, an energy density in the vicinity of 100

Table 20.-Potential Battery Systems for Electric (and Hybrid) Vehicles

Energy density Power densityBattery (Whr/lb) (W/lb) Status

Lead-acid . . . . . . . . . 15-20 5-20 AvailableNickel-zinc . . . . . . . . 30-40 40-80 Available, but expensiveZinc-chlorine . . . . . . ~ 3 5 ~ 5 0 Experimental; potentially inexpensiveAluminum-air . . . . . . 100-200 ~ 8 0 Experimental; cannot be electrically

recharged (requires periodicadditions of water and aluminum)

Sodium-sulfur. . . . . . ~100 ~100 Prospective; high-temperature;potentially inexpensive



Whr/lb, along with a power density of about 100W/lb, would suffice for a practical general-pur-pose vehicle. With these characteristics, 400 lbof batteries would give about 100 miles of travelbetween battery recharging and produce about55 horsepower. Urban or commuter cars mightget by with somewhat lower figures. Table 20shows that currently available battery systemseither cannot achieve such figures, or—as withnickel-zinc batteries-are too expensive for wide-spread use.

Battery-electric cars–often production vehiclesconverted by replacing the engine and fuel sys-tem with an electric motor and lead-acid batteries(like the storage batteries used in golf carts) -havebeen built in prototype or limited-productionform for years. At present, a four-passenger elec-tric car with lead-acid batteries would weighabout twice as much as a conventionally pow-ered car, cost twice as much, and have a rangeof less than 50 miles before recharging. The bat-tery pack alone would weigh 1,000 lb or more,and would have to be replaced several times dur-ing the life of the vehicle, adding to the operatingcosts.

In addition to the nickel-zinc batteries men-tioned above, there are a number of other candi-date battery systems for EVs–of which table 20includes three as examples—the zinc-chlorine,aluminum-air, and sodium-sulfur batteries. Theseshare the advantage of relatively inexpensive rawmaterials, but have other drawbacks: for exam-ple, the zinc- chlorine battery has low energydensity; the aluminum-air system is “recharged,”not by an inward flow of electricity, but bymechanical replacement of materials (in consum-ing materials to produce electricity the aluminum-air battery is like a fuel cell, but fuel cells arecontinuous-flow devices); the sodium-sulfur bat-tery operates at temperatures greater than 5000F. All of these batteries are experimental, andnone has been developed as rapidly as oncehoped; the same is true of many other candidate

battery systems with theoretically attractive char-acteristics for EVs and/or hybrid vehicles.14

Not only are battery-electric cars severely lim-ited in range and performance by the energy andpower densities of available batteries, but pro-duction costs would also be high, at least initial-ly. A further and serious disadvantage is the lim-ited life of many prospective battery systems.Often, the batteries would need to be replaced–at high cost—before the rest of the vehiclereached the end of its useful life. Battery-electriccars also pose new and different safety problems,such as spills of corrosive chemicals in the eventof an accident,

Battery-electric powertrains may have a placein local delivery trucks, and perhaps for small,specialized commuter cars. More widespread usedepends on large improvements (a factor of atIeast 5 in battery performance, particularly energydensity). Although research and development(R&D) on battery systems for EV applications willcontinue, there seems little likelihood of signifi-cant production—i. e., hundreds of thousands ofvehicles per year—before the end of the century.“Breakthroughs” in batteries are improbable;slow incremental progress is more likely to char-acterize R&D on battery systems, and hence EV(and hybrid) vehicles. Moreover, by the time bat-tery performance is improved sufficiently for prac-tical application, progress in fuel-cell technologymay make the latter a more attractive option.(Fuel cells convert a fuel, now generally hydrogenbut potentially a hydrocarbon or methanol, di-rectly to electricity.)

Hybrids also are limited by battery perform-ance, but the on board charging capacity meansthat not as many batteries are needed, so the bat-

IAA. R. Lancfgrebe, et al., “Status of New Electrochemical Storageand Conversion Technologies for Vehicle Applications, ” Proceec/-ings of the 16th Intersociety Energy Conversion Engineering Confer-ence, Atlanta, Ga., Aug. 9-14, 1981, Vol. 1 (New York: AmericanSociety of Mechanical Engineers, 1981), p. 738.

120 ● Increased Automobile Fuel Efficiency and Synthetic Fuels: Alternatives for Reducing Oil imports

tery pack is lighter. However, hybrids must carrya complete heat engine, as well as a generatoror alternator, and an electric drive motor. Al-though the engine can be small, because it neednot be capable of powering the vehicle by itself,the cost and complication of the hybrid power-train are prohibitive, at least at present. The com-plication comes not only from the duplicateenergy conversion and drive systems but also

from the control system. While the performancerequirements for the control system are not unu-sual, the need for a reliable, mass-produced sys-tem at reasonable cost does create a demandingset of constraints. The added weight of the bat-teries and duplicate drivetrain, and the efficien-cy losses associated with recharging the batteries,also tend to counteract the theoretical advantagesof hybrids in fuel economy.

FUTURE AUTOMOBILE FUEL EFFICIENCYAutomobile technology is not a major con-

straint on fuel economy. Small cars can be de-signed today—indeed, are on the market—withmileage ratings twice the current new-car aver-age. Technology is important for increasing thefuel economy of the larger, more powerful, andmore luxurious cars that many Americans still de-sire. Evolutionary improvements will continue toincrease the mileage of both large and small cars,but the pacing factor at the moment is marketdemand.

Because consumer demand is unpredictable,estimates of post-1985 fuel economy are uncer-tain; these estimates largely reflect expectationsof the importance consumers will place on sizeand gas mileage. Projections of the fuel economythat the U.S. new-car fleet will achieve vary wide-ly, but most now tend to be optimistic. Only 2or 3 years ago, American automakers viewed theCAFE standards, correctly, as pushing their prod-uct lines away from the sorts of cars that mostconsumers still demanded. Now many of thosesame consumers are buying cars with averagefuel economies above the CAFE requirements.

EPA statistics indicate that average domesticnew-car fuel economy averaged almost 24 mpgfor 1981 models sold through January 5, 1981.15If imports are included, the figure is about 25mpg. A few predictions are as high as 90 mpgfor 1995 or 2000, although such projections areusually exhortations rather than realistic attemptsto project future trends. While the technology toachieve such efficiencies will exist, fleet averages

15’’ Light Duty Automotive Fuel Economy . . . Trends Through1981 ,“ op. cit.

are likely to remain well below the economy rat-ings that the best performers will be able toachieve.

The primary differences among the many pro-jections of automobile fuel economy for the yearsahead arise from varying assumptions of futuremarket demand. Different assumptions for therate of introduction of new technology are alsocommon. A constraint for American manufactur-ers may be the ability to generate and attract cap-ital for R&D and for investment in new plant andequipment, particularly if movement towardsmall, high-mileage cars and introductions of newtechnology are more rapid than domestic firmshave been anticipating. Many foreign automakersalready produce cars that are smaller and lighter—and get better fuel economy—than those theynow sell in the United States.

Although the fuel economy achieved by thenew-car fleet in future years will depend strong-ly on market demand and the health of the autoindustry, technology is also important. Both thetiming of new vehicle designs and their ultimatecosts—whether routine downsizing and materialssubstitution, or more demanding tasks such asimproved powerplants—depend on extensiveprograms of engineering development. Thesetake time and talent, as well as money. Completesuccess can never be guaranteed. Some projectswill have more satisfactory outcomes than others.

To distinguish these technological dimensionsfrom questions of market demand, the discussionbelow first outlines two scenarios for future devel-opments in automobile technology. Designatedthe “high-estimate scenario” and the “low-esti-

Ch. 5—Increased Automobile Fuel Efficiency . 121

mate scenario, ” they represent plausible upperand lower bounds for fleet average passenger-carfuel economy in future years. Of course, amongthe cars on the market in any year, some wouldhave mileage ratings considerably below, someconsiderably above, the fleet averages for thatyear. These scenarios are independent of marketdemand for cars of various size classes, and aresimply based, respectively, on optimistic and pes-simistic expectations for rates of advance in auto-mobile technology as these affect fuel economy.Using these scenarios, later sections of the chap-ter discuss the effect of market demand on thefuel economy of the U.S. auto fleet.

Technology Scenarios

Both the high- and low-estimate technologyscenarios take as a baseline the new-car fueleconomy now expected for 1985. This baselineincludes a “number of technical advances, as wellas further downsizing, compared with 1982 mod-el cars. While the product plans of individualmanufacturers for 1985 are not known in detail,the broad outlines of 1985 passenger-car technol-ogies can be easily discerned. The scenarios thencover the period 1985-2000. The high estimateassumes:

●

●

●

The

●

●

that engineering development projectsaimed at improving fuel economy are gen-erally successful;that these technological improvements arequickly introduced into volume production;andthat they produce fuel economy improve-ments at the high end of the range that cannow be anticipated.

low estimate assumes, in contrast:

that development projects are not as success-ful–e.g., that technical problems decreasethe magnitude of fuel-economy gains,lengthen development schedules, and/orresult in high production costs;the pace of development is slower thanwould result from the vigorous efforts to“push” automotive technology assumed forthe high estimate scenario; and

● the resulting fuel-economy improvementsare at the low end of the range that can nowbe anticipated.

From a technological perspective, the vehiclesubsystem most critical for fuel-economyimprovements is the powertrain-i.e., the engineand transmission. Here, as in other aspects ofautomotive technology, more-or-less continuousevolutionary development can be expected. Butmajor changes in powertrains have also been oc-curring—e.g., new applications of diesel enginesto passenger cars.

The pace of development may vary for otheraspects of automobile technology—aerodynam-ics, downsizing and weight reduction, powerconsumption by accessories—but individual inno-vations with large impacts on fuel economy areunlikely. Engine developments, in contrast, de-pend more heavily on successful long-term R&Dprograms; fundamental knowledge–e.g., of com-bustion processes–is often lacking, and the risksas well as the rewards can be large. In contrast,development programs aimed, for instance, atfriction reduction, are likely to be more straight-forward–and less costly.

Table 21 presents OTA’s high and low estimatesfor improvements in fuel economy by categoryof technology, based largely on informed techni-cal judgments. * Relative to an assumed 1985 carwhich gets 30 mpg (EPA rating, 55 percent city,45 percent highway driving cycle), table 21 indi-cates that gains of 35 percent in fuel economymay be possible from engine redesigns, but thatpercentage improvements in transmissions andvehicle systems are likely to be smaller. Nonethe-less, the cumulative improvements in fuel econ-omy can be quite large.

*Alternative methodologies for estimating future fuel economy—e.g., the use of learning curves, or analytical modeling of the vehi-cle system—generally lead to comparable results. All approachesto projecting fuel economy have their limitations. The methodadopted by OTA does not always do the best job of evaluating thesystems effects of combining different technologies—i.e., open-chamber diesel engines combined with four-speed lockup torqueconverters. Learning curves, based on historical trends, do not takeexplicit account of new technologies. Analytical modeling is a valu-able tool for comparing alternative technologies, but models must

be validated by comparison with hardware results before the modelcan be used with confidence.


Table 21.–Prospective Automobile Fuel-Efficiency Increases, 1986-2000

Percentage gain in fuel efficiency

High estimate Low estimateTechnology 1986-90 1991-95 1996-2000 1986-90 1991-95 1996-2000EnginesSpark-ignition (SI) . . . . . . . . . . . . . . . . . . . . . . . . 10 10-15 15 5 5-1o 5-1oDiesel:

Prechamber. , . . . . . . . . . . . . . . . . . . . . , . . . . . 15 15 15 15Open chamber . . . . . . . . . . . . . . . . . . . . . . . . . 25 35 35 20 20 25

Open chamber (SI) stratified charge (SC) . . . . 15 20Hybrid diesel/SC . . . . . . . . . . . . . . . . . . . . . . . . . 35TransmissionsAutomatic with lockup torque converter . . . . . 5 5 5 5 5 5Continuously variable ((XT). . . . . . . . . . . . . . . . 10 15 10Engine on-off . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 10 10Vehicle systemWeight reduction (downsizing and

materials substitution) . . . . . . . . . . . . . . . . . . 8 13 18 4 8 10Resistance and friction (excluding engine)

Aerodynamics. . . . . . . . . . . . . . . . . . . . . , . . . . 2 3 4 1 2 3Rolling resistance and lubricants . . . . . . . . . 1 2 3 1 1 2

Accessories . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 3 4 1 1 2a improvements in fuel efficiency are expressed as percentage gain in mpg compared with an anticipated average 1985 passenger car. The 1985 average car used as

a reference has an Inertia weight of about 2,500 lb, is equipped with spark-ignition engine, three-aped automatic transmission, and radial tires, and has an EPA mileagerating (55 percent city, 45 percent highway) of 30 mpg. The fuel efficiencies of the individual baseline cars, which are used to calculate future fuel efficiencies ineach size class, are given In tables 23 and 24. Percentages are given on an equivalent Btu basis where appropriate—e.g., for diesels, which use fuel having higherenergy content per gallon than gasoline, the percentage gain refers to miles per gallon of gasoline equivalent, which Is 10 percent less than miles per gallon of dieselfuel. The table does not include efficiency improvements from alternate fuels such as alcohol.


None of the figures in this table should be inter-preted as predictions. Rather, they illustrateranges in fuel-economy improvement, based onOTA’s judgment of what is likely to be technicallypractical. The projected improvements are notdirectly associated with the developmental pro-grams of specific automobile manufacturers—either domestic or foreign. As changes in auto-mobile technology occur, older designs will coex-ist with new—just as, recently, older V-8 SI en-gines have remained in production alongside re-placements such as diesels and V-6 SI engines.New engines and transmissions are typically intro-duced with the presumption that they will remainin production for at least 10 years. These ratherslow and gradual patterns of technologicalchange are likely to continue unless market con-ditions force an acceleration. For this to happen,the market pressures would have to be rather in-tense, if only because of the limited capital re-sources available at present to the domestic auto-makers.

Table 21 lists the net improvements in automo-bile components that could be expected, on theaverage, for the high and low estimates during

each 5-year period. Note that the technologieslisted in the table are not in every case compati-ble with one another, nor can any simple com-bining procedure yield net figures that have clearand direct meaning for particular hypotheticalvehicles. This is because different technologiescombine in different ways. For example, the poorpart-load efficiencies of throttled SI engines meanthat continuously variable transmissions and en-gine on-off will yield greater improvements thanfor partially throttled open-chamber stratified-charge engines or diesels. Thus, the choice ofcost-effective technologies cannot be inferredfrom such a table alone, but must depend onmore detailed analysis, and finally on testing.

The technologies listed in table 21 are discussedin more detail in appendix 5A at the end of thischapter.

Projection of AutomobileFleet Fuel Efficiency

Based on the technological scenarios in table21 and several assumptions about the size mixof new cars, OTA has constructed a set of pro-


jections for the fuel economy of passenger carssold in the U.S. market in future years. As empha-sized earlier in this chapter, market demand—not technology—is the key factor in determiningthe mileage potential of the new-car fleet, Marketdemand is particularly critical in determining thesize mix of new-car sales.

The automobile technologies listed in table 21are more important as tools for increasing the fueleconomy of the larger, more Iuxurious cars thatmany American purchasers still demand than forcars that are small and Iight—e.g., nearly all cur-rent imports, as well as the new generations ofAmerican-made subcompacts. Improved power-trains and the use of materials with high strength-to-weight ratios will lead to improved fuel econ-omy in cars of all sizes. But a 10-percent increasein gas mileage for a big car—with mileage thatis initially low—saves more fuel than a 10-percentimprovement to a small car that is already morefuel efficient.

This is not to say, however, that a given tech-nological development will necessarily give thesame percentage improvement for cars of allsizes—or even be applicable to all types of cars.Continuously variable transmissions (CVTs) havebeen in limited use for many years in small cars,and would no doubt be applied widely in sub-compacts before finding their way into heaviervehicles. The reason is simply that the mechanicaldesign problems for a CVT are simpler if the levelsof torque that must be transmitted are low.

On the other hand, if gas turbines becomepractical as automobile powerplants, they arelikely to be used first–and perhaps exclusively–in big cars, because turbine engines are more effi-cient in larger sizes.

Any projection of fleet fuel economy will de-pend on the assumed weight (size) mix of new-car sales in the years ahead. For its analysis, OTAadopted a simplified description of this mix,based on three size classes–small, medium, andlarge. This allows possible market shifts to be ana-lyzed in terms of the assumed proportion of new-car sales by size class—for each of which the av-erage fuel economy has been estimated. This isa considerable abstraction from the real situation–-one in which the spectrum of curb weightsfrom which consumers select extends from lessthan 2,000 lb to over 4,000 lb. For any givenweight—now and in the future—there will alsobe a range of fuel economies, depending on vehi-cle design. The convenience of the descriptionin terms of only three size classes, for which othercharacteristics are averaged, comes at the ex-pense of the richness and variety that will actuallyexist in the marketplace.

New-Car Fuel Efficiency by Size Class

Table 22 describes the small, medium, andlarge size classes on which OTA’s projections arebased. The scheme is similar to current EPA prac-tice for fuel-economy ratings—grouping cars ofsimilar passenger capacity and interior volume.However, the designations of car sizes in table22 differ from some current designations becausethey are intended to reflect future vehicle charac-teristics rather than the past; in other words, OTAprefers to call a future small car just that, not a‘‘m in i compact .“ Each class in the table encom-passes a considerable range of possible vehicledesigns. Under either the high or low estimatescenarios, curb weights of cars in the U.S. fleetare expected to decrease over the period 1985-2000.

Table 22.–Small, Medium, and Large Size Classes Assumed for 1985-2000

Curb weighta (lb) 1981 equivalents

Class High estimate Low estimate Interior volume (ft3) Passenger capacity Size class Typical models

Small . . . . . . . . 1,300-1,600 1,400-1,700 < 85 2-4 Minicompact, Honda Civictwo seaters Toyota Starlet

Medium. . . . . . 1,600-2,000 1,700-2,000 80-110 4 Subcompact, VW Rabbitcompact Chrysler K-Car

Large. . . . . . . . 2,200-3,000 2,500-3,000 100-160 5-6 Intermediate, GM X-Carlarge, luxury Ford Fairmont

aCurb weight is the weight of the car without passengers or cargo.



Tables 23 and 24 expand on the descriptionsin table 22. For these tables, OTA has estimatedweight averages, engine alternatives, and averagefuel economies at 5-year intervals through 2000for the two technology scenarios. Again, theseestimates reflect informed technical judgment butshould not be viewed as predictions. The curbweights are averages expected for each of thethree size classes; rather broad ranges in actualweights are likely, especially in the medium andlarge classes. The fuel economy estimates are like-wise averages with considerable spread antici-pated. Fuel economy projections are given interms of current EPA rating practice (combinedcity-highway figures)—which overestimate actualover-the-road mileage by as much as 20 percent.The EPA rating basis has been adopted for easeof comparison with fuel economy ratings for thecurrent fleet; in later sections, to estimate actualfuel consumed, EPA ratings are adjusted down-ward to more realistic values.

The average fuel economy estimates in tables23 and 24 for the high- and low-estimate technol-ogy scenarios are grouped together in table 25so that the differences by size class and technol-

ogy level can be more easily compared. Table25 illustrates the importance of size and weightfor fuel economy. By 2000, the low-estimate aver-age efficiency for medium-size cars is the sameas the high-estimate efficiency for big cars—bothare 50 mpg. Large cars show the greatest percent-age improvements because more can be doneto improve fuel economy before diminishing re-turns become severe.

One way to abstract the effects of downsizingand weight reduction from other technologicalimprovements is to examine specific fuel econ-omy—by normalizing to ton-mpg, or the milesper gallon that would result for an otherwise sim-ilar car weighing 1 ton. Ton-mpg values have ex-hibited an upward trend overtime as automotivetechnologies have improved.16

Figure 9 shows the gradual increase–with con-siderable year-to-year fluctuations—that has char-acterized average fuel economy in ton-mpg forthe U.S. new-car fleet over the past decade, to-gether with estimates through 2000 based on the

16"Powerplant Efficiency Projected Via Learning Curves, ” Auto-motive Engineering July 1979, p. 52.

Table 23.—Automobile Characteristics- High-Estimate Scenario

Small Medium Large1965 1990 1995 2000 1965 1990 1995 2000 1985 1990 1995 2000

Average curb weight (lb) . . . . . . . . . . . . . . . . . 1,600 1,500 1,400 1,300 2,000 1,800 1,700 1,600 3,0002,6002,4002,200Engine type (percent):

Spark ignition . . . . . . . . . . . . . . . . . . . . . . . . 100 95 70 60 90 70 50 30 75 30 30 25Prechamber diesel . . . . . . . . . . . . . . . . . . . . — 5 — — 10 10 — — 25 40 — —Open chamber diesel or open

chamber stratified charge . . . . . . . . . . . . — 30 40 20 50 70 – 30 70 75Fuel economya (mpg) . . . . . . . . . . . . . . . . . . . . 48 62 74 84 39 51 61 71 27 37 43 49aCombined EPA clty/highway fuel-economy rating, baaed on 55 percent city and 45 percent highway driving.SOURCE: Office of Technology Assessment.

Table 24.—Automobile Characteristics-Low-Estimate Scenario

Small Medium Large1965 1990 1995 2000 1985 1990 1995 2000 1985 1990 1995 2000

Average curb weight (lb) . . . . . . . . . . . . . . . . . 1,7001,6001,500 1,400 2,0001,9001,600 1,700 3,0002,6002,6002,500Engine type (percent):

Spark ignition . . . . . . . . . . . . . . . . . . . . . . . . 100 100 100 60 90 60 70 70 75 60 50 40Prechamber diesel . . . . . . . . . . . . . . . . . . . . — — — — 10 20 30 25 40 20Open chamber diesel . . . . . . . . . . . . . . . . . . — — — 40 — — — 3 0 - – – 3 O G

Fuel economya (mpg) . . . . . . . . . . . . . . . . . . . . 45 52 57 65 35 41 45 50 23 28 31 34aCombined EPA city/highway fuel-economy rating, baaed on 55 percent city and 45 percent highway driving.SOURCE: Office of Technology Assessment.

Ch. 5–Increased Automobile Fuel Efficiency • 125

Table 25.—Estimated New-CarFuel Economy: 1985-2000

Average new-car fuel economya

Size class 1985 1990 1995 2000Large:

High estimate . . . . . . . 27 37 43 49Low estimate . . . . . . . 23 28 31 34

Medium:High estimate . . . . . . . 39 51 61 71Low estimate . . . . . . . 35 41 45 50

Small:High estimate . . . . . . . 48 62 74 84Low estimate . . . . . . . 45 52 57 65

aCombined EPA city/highway fuel-economy rating, based on 55 percent city and45 percent highway driving.


Figure 9.—Sales.Weighted Average New-Car FleetSpecific Fuel Efficiency

z - 1975 1980 1985 1990 1995 2000

Year


high and low scenarios in table 21. As for theearlier tables, figure 9 aggregates both domesticautomobiles and imports. Using such projections,the fuel economies of future new-car fleets ofvarious size (weight) mixes can be estimated.

As the figure shows, average specific fuel con-sumption for new cars sold in the United Stateshas increased from less than 30 ton-mpg in theearly 1970’s to roughly 39 ton-mpg in 1981, a30-percent improvement. The most efficient 1981cars sold in this country gets 50 ton-mpg. * By

1990, the average should equal the current best.By 2000, the average could be as high as 65ton-mpg.

Based on the projections in tables 23-25, or al-ternatively those in figure 9, the effects of changesin the size mix of the new-car fleet can be esti-mated. In the mix of new 1981 cars sold throughJanuary 5, 1981, small cars made up only 5 per-cent of the market; the rest was almost evenlydivided between medium cars (48 percent) andlarge cars (47 percent). By 1985, the share of smallcars may remain at 5 percent, but the share ofmedium cars is expected to go up at least to 60percent, dropping the large-car share to 35 per-cent or less. Even in the unlikely event that the60:35, ratio remains unchanged beyond 1985–that medium cars show no further sales gains overlarge cars–the average fuel economy of the new-car fleet in 2000 would be 62 mpg in the high-estimate scenario, 43 mpg in the low-estimatescenario. * (See table 26, “no mix shift” case.)These figures represent a substantial improve-ment over the 25 mpg expected in 1981 and the30 to 35 mpg expected for 1985. A further shiftin consumer preference toward smaller andlighter cars would increase the expected fleet-average fuel economy even more.

To illustrate the effects of a continuing shift to-wards smaller and lighter cars, table 26 also givesaverage fuel economies at 5-year intervals for a“moderate” mix shift-leading to 35 percentsmall cars, 50 percent medium cars, and 15 per-cent large cars by 2000-and for a “large-scale”mix shift. The latter assumes 70 percent smallcars, 25 percent medium cars, and only 5 per-cent large cars in 2000. As the table shows, thelarge-scale mix shift could give a new-car fleetaverage fuel economy of 60 to 80 mpg by 2000.Whether market demand will lead to such a mixshift depends on factors such as price differen-tials between large and small cars, and the com-promises in other vehicle characteristics that ac-company smaller cars, as well as the pricing andavailability of fuel.

*These are diesels, for which the ton-mpg rating has been ex-pressed on a gasoline-equivalent basis; the value based on dieselfuel would be about 55 ton-mpg. The best current SI engine mod-els sold in the United States have ton-mpg ratings about 10 per-cent lower, or roughly 45 ton-mpg.

*The corresponding numbers for the 1981 mix are 59 mpg inthe high estimate and 41 mpg in the low estimate.


Table 26.—Effect on Size Mix on Estimated Fuel Economy of the New-Car Sales in the United States

Estimated average new-car fuel economya (mpg)

No mix shiftb Moderate mix shiftc Large-scale mix shiftd

Technology scenario 1985 1990 1995 2000 1985 1990 1995 2000 1985 1990 1995 2000

High estimate . . . . . . . . . . . . . . . . . . . . . . . . . . 34 45 54 62 34 48 59 70 37 53 65 78Low estimate . . . . . . . . . . . . . . . . . . . . . . . . . . . 30 36 39 43 30 38 43 51 33 43 49 58a55 percent city, 45 percent highway EPA rating.

CThe moderate mix shift assumption is as follows:bThe no mix shift case assumes:

‘The large-scale mix shift assumption is:Sales mix (percent) Sales mix (percent)

Size class Sales mix (percent) for all yearsSmall . .

Size class 1985 1990 1995 2000 Size class 1985 1990 1995 20005

Medium . . . . . 60Small . . . 5 15 25 35 Small . . . . . . . 10 30 50 70Medium . . . . . 60 60

Large . .55 50

35Medium . 75 65 45 25

Large . . . . . . . 35 25 20 15 Large . . . . . . . 15 5 5 5


ESTIMATED FUEL CONSUMPTION, 1985–2000

In this section estimates of the fuel consumedby the U.S. passenger-car fleet are based on thevarious assumptions and projections of car sizemix and efficiency discussed above, but assumethat gasoline and diesel fuel continue to powerpassenger vehicles, with no significant penetra-tion of alternative fuels such as methanol.

In 1975, when Federal fuel economy standardswere enacted, passenger cars consumed an aver-age of about 4.3 MMB/D of fuel. (Trucks areomitted from the calculations in this chapter, butmany light trucks and vans are used interchange-ably with passenger cars and add about 1.1MMB/D to average consumption). Passenger-carfuel consumption rose to 4.8 MMB/D in 1978,but has since declined to 4.3 MMB/D–about the1975 Ievel.17 OTA projects that passenger-car fuelconsumption will continue to decline—to about3.6 MMB/D in 1985, as the automobile fleet be-comes more fuel-efficient. This estimate assumesthat the fleet will grow from about 107 millioncars in 1980 to 110 million in 1985, and that theaverage car will continue to accumulate about10,000 miles per year.

Projected Passenger-CarFuel Consumption

The baseline chosen for discussing fuel con-sumption by passenger cars past 1985 is outlinedin table 27. Growth in the automobile fleet—which depends on both sales levels for new cars,and the rates at which older cars are scrapped—is

Table 27.—Baseline Assumptions for Projectionsof Automobile Fuel Consumption

1980 1985 1990 1995 2000 2005 2010

Vehicle miles of travel:Trillion

miles/yr . . . 1.14 1.17 1.20 1.23 1.26 1.28 1.31Also assumes 47 percent of fleet vehicle miles traveled(VMT) by cars less than 5 years old, 38 percent of VMTby cars 5 to 10 years old, and 15 percent of VMT by carsolder than 10 years.

New-car fuel economy (combined EPA ratings; 55 percentcity, 45 percent highway)

1985 base case (low-high estimate)Small cars: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45-48 mpgMedium cars: . . . . . . . . . . . . . . . . . . . . . . . . . . 35-39 mpgLarge cars: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23-27 mpg

Fleet baseline average efficiency30 mpg, 1985-2010

High- and low-estimate scenariosSee table 25

lzDerived from j. K. pollard, et al., “Transportation Energy Out-

look: 1985-2000,” Transportation Systems Center, U.S. Departmentof Transportation, DOT-TSC-RS-1 12-55-81-6, September 1981, pp.4-21; and “Monthly Energy Review, ” Energy Information Agency,U.S. Department of Energy, DOE/EIA-0035 (81/10), October 1981.

On-the-road fuel efficiency:10 percent less than EPA rated fuel efficiency

Size mix:See table 26SOURCE: Office of Technology Assessment.


projected to average less than 2 percent peryear.18 Cars are assumed to be driven an averageof 10,000 miles per year, with newer cars drivenmore and older cars driven less, on the average(see table 27). The projections for new-car fueleconomy and future size mix are taken from thetables in earlier sections. Because those fuel-economy projections were based on EPA ratings,which overestimate actual on-the-road mileage,fuel economy has been adjusted downward 10percent to compensate.

If neither fuel economy nor size mix were toadvance past a 1985 baseline average of 30 mpg,fuel consumption by passenger cars would stilldecline slowly for 10 years, reflecting the largerfraction of cars in the fleet with fuel economiesat this baseline value, Between 1985 and 1995,passenger-car fuel consumption would decline–even with a status quo in fuel economy and sizemix—from about 3.6 MMB/D in 1985 to 2.7MMB/D in 1995. Thereafter, the upward trendwould resume because of increases in the totalsize of the fleet.

But of course, automobile technology will con-tinue to improve (table 21 ), and a continuing shifttoward smaller cars is also probable (table 26).Therefore, under almost any realistic set of as-sumptions, passenger-car fuel consumption willcontinue to decrease during the post-1995 peri-od. At some point it may still turn upward be-cause of increases in fleet size, this turning pointdepending on both technology and size mix. Inany case, as figure 10 shows, the decline in pas-senger-car fuel consumption will level off byabout 2005 (unless growth in the fleet is slowerthan projected in table 27 or cars are driven fewermiles per year).

Figure 10 gives fuel-consumption projectionsto 2010 based on these assumptions. The influ-ence of technological improvements is striking.For example, even without a mix shift toward carssmaller than in the 1985 baseline mix, the highestimate gives fuel savings greater than those forthe low estimate with a large-scale shift towardssmaller cars. But such a mix shift would also cre-ate substantial fuel savings. For the cases plotted

~au. S. /ndustrja/ Competitiveness: A Comparison of Stee( elec-tronics, and Automobiles, op. cit., pp. 140-141.

Figure 10.—Projected Passenger-CarFuel Consumption

5 . 0

4.030-mpg new-car

3.5 average 1985-2000 0 3.0 2.5

2.0 Low estimate,k *

●

large mix shift 1 . 5 - - -

- - - -

1 . 0High estimate,

‘- - -

no mix shift High estimate. 0.5 large mix shift

0 . 0 1 I I 1 1 1

1975 1980 1985 1990 1995 2000 2005 2010

Year


in this figure, passenger-car fuel consumptionstays well below 2.5 MMB/D during the earlyyears of the next century. The figure also showsthe potential benefits if technical success is ac-companied by a strong shift to smaller cars. Thedifference in 2010 between the low estimate withno mix shift (about 2.0 MM B/D), and the high esti-mate with a large mix shift toward smaller cars(1.1 MMB/D), is nearly a factor of 2. The lowerend of this range is about one-fourth the currentlevel of fuel consumption. Where within thisrange the actual fuel consumption would fall islikely to depend–as emphasized earlier–on mar-ket demand for fuel-efficient vehicles, and/or con-tinuing Government policies designed to encour-age the manufacture and purchase* of fuel-effi-cient cars. Changes in vehicle miles traveledwould also change the fuel consumption propor-tionately.

*An illustration of the importance of new-car sales can be de-rived as follows: In 1980, 47 percent of the vehicle-miles traveled(VMT) were by cars 0 to 4 years old, 38 percent by 5 to 9 year oldcars, and 15 percent by cars 10 years old and older. Call this thebase case. A persistent 20 percent depression in new car sales couldchange the VMT distribution by 1995 to: 40 percent by cars O to4 years old, 35 percent by cars 5 to 9 years old, and 25 percentby cars 10 years old and older. Call this the “low” car sales case.If VMT are held constant at the 1980 level, fuel consumption underthe base case would be up to 0.3 MMB/D (or nearly 20 Percent)lower than fuel consumption in the “low” car sales case in 1990,everything else being equal. And cumulative oil savings could beover 1 billion bbl during the period 1981-2000. The base case, how-ever, probably would be accompanied by higher VMT than the“low” sales case, and much of this savings could be lost.


Total projected oil savings are bracketed by thecurves in figure 11. These show the fuel con-served relative to the 1985 baseline case of new-car efficiencies of 30 mpg between 1985 and2010. Clearly, the high-estimate technology sce-nario, accompanied by a continuing shift towardsmaller cars, leads to large fuel savings. By2010–when virtually the full benefit of fuel sav-ings from cars sold in the period 1985-2000 wouldbe realized–the cumulative savings (relative toa 30-mpg fleet) could be as high as 10 billion bblof oil equivalent. This is equivalent to 6 years sup-ply of passenger-car fuel at the 1980 rate of con-sumption. The fuel economy increases expectedbetween now and 1985 would add about 14 bil-lion bbl to this cumulative savings between nowand 2010 (relative to 1980 fuel consumption).Thus, between now and 2010, the total savingspossible is about 24 billion bbl relative to 1980passenger-car fuel consumption–an amountabout equal to proven U.S. oil reserves, whichwere 26.5 billion bbl as of 1980.19

Substitution of Electric Vehicles

The estimates above are based on a passenger-car fleet for which energy comes from a fuel car-ried onboard—e.g., gasoline or diesel fuel. In the

19Wof/d Petro/eum Availability 19802(XXL Technical Memoran-dum, OTA-TM-5 (Washington, D. C.: U.S. Congress, Office of Tech-nology Assessment, october 1980.)

Figure Il.—Cumulative Oil Savings From IncreasedAutomobile Fuel Efficiency Relative to 30 MPG in

1985, No Change Thereafter

0

1985 1990 1995 2000 2005 2010

Year


“Battery-Electric and Hybrid Vehicles” section,the prospects for EVs were briefly discussed, withthe conclusion that major improvements in bat-tery performance were necessary before EVs (orhybrids) would be practical in any but very spe-cialized applications. If, however, these improve-ments are achieved—or if acute shortages oftransportation fuels occur in the future—EVsmight be sold in sufficiently large numbers to af-fect petroleum consumption.

The result would be to replace some of the pe-troleum consumed in the transportation sectorby electric power generation. To the extent thatthis electricity was produced from nonpetroleumfuels–e.g., natural gas, coal, nuclear–the cumu-lative oil savings shown in figure 10 would in-crease (see app. 56). Table 28 illustrates theresults for a highly optimistic level of EV substitu-tion. Note that this again is not a prediction; sub-stantial penetration by electric and/or hybridvehicles (EHVs) before the end of the century isunlikely, and doubtful even thereafter. The tablesimply shows what might happen if battery im-provements occur more rapidly than OTA ex-pects, or if other factors combine to increase theattractiveness of EHVs. Table 28 assumes thatEHVs represent 5 percent of the total U.S. passen-ger-car fleet by 2000, and 20 percent by 2010.This would require EHV production and sales atlevels of several million per year during the lastfew years of the century.

Table 28 shows that penetration of EHVs at highenough rates could begin to replace meaningfulvolumes (14 percent) of transportation fuels dur-

Table 28.—Effects of Substituting Electric Vehicless

Passenger-carComposition of fuel consumedpassenger-car or replacedfleet (million) (MMBID)

2000 2 0 1 0 2000 2010Conventional cars . . . . . 133 124 1.7 1.4EVs . . . . . . . . . . . . . . . . . . 7 31 0.04 0.2Percent EVs . . . . . . . . . . 5 20 — —Percent fuel

consumptionreplaced. . . . . . . . . . . . — — 2 14

qf battery Improvements occur more rapidly than OTA expects, or if other factorscombine to increaae the attractiveness of electric vehicles.



ing the first decade of the next century. Nonethe-less, the savings in petroleum would be relative-ly small in absolute terms–because EHVs are bestsuited as replacements for small cars which al-ready get good mileage.

Comparing the estimated fuel savings in table28–only 0.2 MM B/D even for optimistic assump-tions of EHV penetration—with the fuel-consump-tion trends projected in figure 9, demonstratesthat improvements in automobile technology,particularly if combined with more rapid mixshifts toward smaller cars, offer much greaterpotential. Thus, the primary apparent advantageof EVs during the next 30 years is that they wouldnot depend on petroleum supplies—an impor-tant factor if severe absolute shortages develop–rather than any potential for saving petroleum.

Fuel Use by OtherTransportation Modes

Thus far, the discussion of fuel consumptionhas been restricted to passenger cars, although,as pointed out earlier, many light trucks—i.e.,vans and pickups—are used primarily for passen-ger travel. In addition, medium and heavy trucks,buses, motorcycles, and airplanes–plus rail andmarine transportation and military operations—consume petroleum-based fuels. All of these

transportation modes depend predominately onheat engines for power, although SI engines arenot so widely used as in passenger cars. Dieselshave already replaced SI engines in almost allheavy trucks, and rates of installation in medium-duty trucks are going up rapidly. Diesels are alsocommon in rail and marine applications, al-though some large ships rely on gas turbines orsteam power. Commercial aircraft are generallypowered by turbine engines.

Table 29 summarizes the projected oil con-sumption for transportation between 1980 and2000. The projections for automobiles are derivedin this chapter, while those for other transporta-tion modes are taken from the “market trend”base case in a recent Department of Transporta-tion study.20 The projections for fuel consump-tion by trucks in table 29 assume that many ofthe technological improvements discussed abovefor passenger cars will also be applicable to lighttrucks. However, the specific technologies dis-cussed elsewhere in this chapter are more gener-ally appropriate to pickup trucks and vans thanto medium and heavy trucks.

Zoj. K. Pollard, et al., “Transportation Energy Outlook: 1985-2000,”Transportation Systems Center, U.S. Department of Transportation;DOT-TSC-RS-1 12-55-81-6, September 1981.

Table 29.-Projected Petroleum-Based Fuel Use for Transportation

1980a 1990a 2000a

Mode M M B / Db P e r c e n t M M B / Db P e r c e n t M M B / Db P e r c e n t

Passenger car . . . . . . . . . . . . . 4.3 49 2.4-2.9 35 1.3-2.1 23Light trucks . . . . . . . . . . . . . . . 1.1 13 0.9 12 0.8 11Other trucks. . . . . . . . . . . . . . . 1.1 13 1.2 16 1.4 19Other highway (buses,motorcycles, etc.) . . . . . . . . . . 0.1 1 0.2 3 0.2 3

Total highway . . . . . . . . . . . 6.6 75C 4.7-5.2 65C 3.7-4.5 5 5C

Air . . . . . . . . . . . . . . . . . . . . . . . 0.8 9 1.1 14 1.5 20Marine. . . . . . . . . . . . . . . . . . . . 0.7 8 0.8 10 0.9 12Rail . . . . . . . . . . . . . . . . . . . . . . 0.3 3 0.4 5 0.4 5Pipelines d . . . . . . . . . . . . . . . . . 0.1 0.1 1 0.1 1Military Operation . . . . . . . . . . 0.3 3 0.3 4 0.4 5

Total nonhighway . . . . . . . . 2.2 25C 2.7 35C 3.3 45C

Total . . . . . . . . . . . . . . . . . 8.8 100 7.4-7.9 100 7.0-7.8 100aA\\ fuel consumption numbers, except for passenger cars, from J. K. Pollard, C. T. Phillips, R. C. Ricci, and N. Rosenberg,

“Transportation Energy Outlook: 1985-201M,” U.S. Department of Transportation, Transportation Systems Center, Cambridge,Mass., DOT-TSC-RS-112-55-81%, September 1981. Passenger-car fuel consumption from this study.

blB - 5 . 9 MMBtu.csum~ may not agr~ due to round-off errors.dDoes not include naturai 9as.


130 • Increased Automobile Fuel Efficiency and Synthetic fuels: Alternatives for Reducing Oil Imports

Some of the fuel economy gains for other trans-portation modes—as for passenger cars—will beoffset by growth in miles traveled. Annual growthrates of 1 to 3 percent per year are expected formost modes of transport, although sales of lighttrucks have recently dropped to such an extentthat mileage traveled by such vehicles may de-crease in the years ahead. In total, fuel consumedfor transportation is projected to decline from 8.8MMB/D in 1980 to the range of 7 to 8 MMB/Dby 2000, and then to rise slowly as diminishingreturns set in.

Because of the differing growth rates for the var-ious transportation modes and the differing mag-nitudes of the fuel economy improvements ex-pected, the distribution of fuel use by mode willchange. Passenger cars now account for half ofall the fuel used in transportation. Their share willdecrease to about 25 percent by the early partof the next century. Medium and heavy truckscurrently consume 12 percent of all transporta-tion fuel, a figure that could rise to 20 percentby 2000. Likewise, the percentage of transportfuels used by aircraft could nearly double.

COSTS OF INCREASED FUEL EFFICIENCY

Overview

Automobile manufacturers spend for manypurposes–R&D, investment in plant and equip-ment, labor, materials, marketing, and adminis-tration. How much a particular manufacturerspends depends on the firm’s financial capabili-ties, the rate of technology change, initial charac-teristics of the product line, and the state of ex-isting manufacturing facilities.

R&D on vehicle designs and manufacturingprocesses is an important spending area. Devel-opment—on which most R&D money is spent,research expenditures being small by comparison—creates new product designs and productionmethods. Growth in R&D activity, required tosupport rapid changes in vehicle design, will raiseboth the total costs of automobile production andthe proportion of development and other prepro-duction expenses.

In 1980, the four major domestic manufacturersspent almost $4.25 billion (1980 dollars) on R&D.For individual firms, this spending amounted to2 to 5 percent of sales revenues. in addition, ma-jor parts and equipment suppliers spent about$293 million on automotive R&Din 1980. Togeth-er, major automobile manufacturers and suppli-ers spent over $4.5 billion on R&D for automo-biles and other vehicles.21

Capital investment levels are even greater.These expenditures, which go hand-in-hand withdesign and development activities, are the largestsingle category of spending in automobile manu-facturing. Major categories of capital goods in-clude factory structures, production equipmentsuch as machine tools and transfer lines, and awide variety of special tools such as dies, jigs, andfixtures.

Manufacturers today are making investmentsto improve product quality, as well as increaseproductivity and cut costs. Flexible manufactur-ing is also becoming an increasingly attractive in-vestment. Such sophisticated facilities are relative-ly expensive but may yield low operating costsand other long-term benefits. General Motors(GM), Ford, and Chrysler spent $10.8 billion(1980 dollars) on property, plant, equipment, andspecial tooling in 1980.22

Financing is an important aspect of capital in-vestment. Historically, the automobile industryhas financed capital programs with retained earn-ings, except during recessions when low salesgenerated inadequate revenues. Several current,and possibly enduring, factors—declining profit-ability, high inflation, slow market growth, marketvolatility, and consumer resistance to real priceincreases—have eroded manufacturers’ ability tofinance major capital programs from earnings (orby issuing stock), leading them to borrow funds

2]’’ R&D Scoreboard: 1979,” Business Week, July 7, 1980; “R&DScoreboard: 1980, ” Business Week, JuIy 6, 1981.

22Annual Reports for 1980.

Ch. 5—Increased Automobile Fuel Efficiency . 131

and therefore to face potentially higher costs ofcapital. GM and Ford each borrowed over $1 bil-lion during 1980; they may together borrow asmuch as $5 billion by the mid-1980’s.23

Although domestic manufacturers have recent-ly borrowed from foreign and other nontradition-al sources, they are able to borrow only a limitedportion of their capital needs (at acceptable inter-est rates). Borrowing in the United States has re-cently become more costly to automobile manu-facturers because their bond ratings have beenlowered, in recognition of the low profitability,high spending levels, and high risks that charac-terize today’s auto market. Consequently, theyare obtaining cash by restructuring their physicaland financial operations (e.g., by selling assetsand changing the handling of accounts receiv-able), engaging in joint ventures, and selling taxcredits (under Economic Recovery Tax Act of1981 leasing rules).

Automotive fuel economy improvement affectsother costs as well, although not to the same ex-tent that it affects R&D and capital investment,Costs for labor and materials depend on vehicledesign and on production volumes and proc-esses. For example, automated equipment re-duces labor content; small cars require less ma-terial; and lighter body parts and more efficientengines may require new, relatively expensivematerials (high-strength steels, aluminum alloys)and processes (heat treatments, longer weld cy-cles, a greater number of forming operations,slower machining). Reductions in the amountsof labor and materials used per car help offsetinflationary and real increases in their costs. Laborcosts, however, are slow to change in the shortterm because they are subject to union negotia-tions, and because contractual provisions con-strain layoffs and require compensation pay-ments.

Finally, spending on marketing and administra-tion is not directly related to technological changeor to production; although these expenses maybe cut back to facilitate spending in other areas.During 1980, for example, auto manufacturersmade large cuts in white collar staffs to lower ad-

Z3AIS0 see “producing More Fuel-Efficient Automobiles: A CoSt-

Iy Proposition, ” op. cit.

ministrative costs. However, marketing activitiesmay increase because of heightened competitionor the introduction of new products.

The remainder of this chapter focuses on capi-tal costs, because they are the critical componentof the overall costs of changing automobile de-signs. On a per car basis, however, labor and ma-terials costs will remain higher than capital costsbecause of the ways different types of costs areallocated. The costs of capital goods (includingfinancing) are recovered throughout their servicelife in vehicle prices. Since capital goods are usedto produce many vehicles over many years (atleast 30 years for plants, 12 years for much pro-duction equipment, and 3 to 5 years for specialtooling), each vehicle bears a relatively small per-centage of the costs of capital to produce it. Incontrast, labor services and materials are effec-tively bought to manufacture each car.

Relative to other manufacturing costs, capitalcosts are expected to undergo the greatest per-centage increase as manufacturers increase theiroutput of fuel-efficient vehicles. Moreover, capitalcosts are becoming proportionately greater be-cause capital goods are being purchased at fasterrates and at higher real prices than historically,and because automotive production is becomingmore capital-intensive as automation proceeds—i.e., more capital equipment is being used to pro-duce automobiles relative to labor, materials, andother inputs. Consequently, capital costs willhave an especially pronounced influence on thefinancial health of automobile manufacturers overthe next two decades.

Investments to Raise Fuel EconomyTechnology-Specific Costs

Table 30 presents capital cost estimates pre-pared by OTA for the technologies describedearlier in this chapter, based on discussions withindustry analysts and the most recently publishedanalyses. However, they draw on the experienceand expectations of the mid and late 1970’s,when limited consumer demand for fuel econ-omy led manufacturers to make conservative pro-jections of vehicle design changes and high pro-jections of costs. Because of recent surges in thedemand for fuel economy and small cars, manu-facturers now expect to make substantial im-


Table 30.—Post-1985 Automotive Capital Cost Estimates

$M/500,000 units Associated costs

Platform changeWeight reduction, redesign . . . . . . . . . . . . . 500-1,000 R&Da, redesignMaterial substitution . . . . . . . . . . . . . . . . . . . 100-400 R&Da, materials, laborEngine changeImproved Sib, diesel . . . . . . . . . . . . . . . . . . . 50-250New Sib, diesel, DISCC. . . . . . . . . . . . . . . . . . 400-700

R&Da, redesign

Transmission changeImprove contemporary drivetrains. . . . . . . . 100-400 R&Da, redesignNew drivetrains—CVTd, energy storage,

engine on-off. . . . . . . . . . . . . . . . . . . . . . . . 500-700 R&Da, redesign%apital costs for accessory and iubricant Improvements and aerodynamic and rolling resistance reductions are not Includedseparately. They may in total cost about S50M1500,000, an amount within the range of error impiied by the above estimates.Note: aerodynamic improvements will be carried out with weight-reducing design changes; iubricant end tin changes arealready being made by suppliers and may continue as a regular aspect of their businesses; and some accessory Improvements

. are made regularfy and as accompaniments to engine redesigns.%park ignition.cDirect injection stratified charge.d~ntinuoualy variable transmission.


provements in fuel economy by the mid-1980’sby accelerating technological change schedulesand by increasing the proportion of small cars inthe sales mix.

Note that increasing production volume for ex-isting models is much cheaper than introducingnew models; U.S. manufacturers could probablydouble their output of many existing models. Thecosts of design changes in the post-1985 periodnow seem even more uncertain, because earlierachievements will leave fewer and generallymore costly options available.

Projecting costs for specific design changes isdifficult, for several reasons. First, the redesignof any one vehicle component or subsystem oftennecessitates related changes elsewhere. Second,such changes may require new production pro-cesses. Third, actual costs to individual manufac-turers are technology-specific and sensitive to sev-eral factors—technological development, produc-tion volume, vertical integration, the rate at whichchanges penetrate the fleet, and available manu-facturing facilities. These factors are discussedbelow.

Technological Development.–Many technolo-gies are inherently expensive due to materials re-quirements or complexity of design or manufac-ture. The diesel engine for passenger cars is agood example. Over time, experience with a newtechnology may lead to some cost reduction.

Production Volume.–Costs vary with produc-tion volume because equipment and processesare designed such that average product cost islowest once a threshold production volume isachieved.24 Because this minimum volume orscale grows as the production process becomesmore highly automated, the rising capital inten-siveness of automobile production increases thesensitivity of unit costs to production volume.Operating costs (comprised of labor, materials,and allocated marketing and administration costs)per unit are sensitive to production volume in theshort term. For example, Ford’s operating costsper dollar of sales were estimated to be under$0.90 in the first quarter of 1979, but subsequentsales declines brought them close to $1.05 by thefourth quarter of 1980.25

The cost estimates in table 30 assume uniform500,000-unit capacities. * Cost estimates for uni-form or optimal capacity levels provide a bettermeasure for spending levels for the industry asa whole than for individual manufacturers be-cause individual firms acquire different levels ofcapacity at different costs according to their finan-

Z4S& K. Bhasker, “The Future of the World Motor Industry” (NewYork: Nichols Publishing Co., 1980.) The optimum productionvolumes may change with manufacturing technology, however.

‘s’’ Ford’s Financial Hurdle: Finding Money is Harder and Harder,”Business Week, February 1981.

*This procedure was aiso employed in the Mellon Institute study(Ref. 32) which drew on data provided by automobile manufac-turers.


cial ability, sales volume, and technological op-tions. The costs of acquiring more or less thanoptimal capacity, however, are not linearly re-lated to the level of capacity.

Vertical Integration. —Vertical integration re-fers to the degree to which a manufacturer is self-sufficient in production or distribution. integra-tion can reduce costs in two ways: 1 ) by eliminat-ing activities and costs associated with the transferof goods between suppliers and distributors (deal-ers), and the automakers themselves; and 2) byenabling manufacturers to optimize the flow ofproduction and distribution. Major U.S. automo-bile manufacturers are highly integrated com-pared with firms in many other industries, al-though they are much less integrated than oilcompanies. Various U.S. automobile firms makesteel, glass, electronic components, and robots,but overall they buy about ha If of their materialsand other supplies. GM’s greater vertical integra-tion relative to other U.S. automobile manufac-turers is one reason for its lower manufacturingcosts. The high effective degree of vertical integra-tion among Japanese auto manufacturers (over80 percent for some firms) helps to make autoproduction in Japan cheaper than in the UnitedStates. *

Rate of Change.–The rate at which new tech-nology is incorporated in automobiles influencescost in three important ways. First, the faster adesign is implemented, the shorter are the prod-uct development, product and process engineer-ing schedules, and the less likely is productionto be at minimum cost, given scale. Second, in-creasing the rate of technological change raisesthe number and magnitude of purchases fromsuppliers. Third, a faster rate of change can makefacilities and processes technologically obsolete,necessitating investments in replacements beforeoriginal investments are recovered.

Available Facilities.—Opportunities for manu-facturers to redesign their product lines areshaped by the characteristics of their base vehi-cles and existing production facilities. The techno-logical scenarios described at the beginning of

*These conditions reflect peculiarly close relationships betweenJapanese manufacturer and supplier firms, even in the absence offormal linkages.

this chapter illustrate how paths of change maydiffer.

Estimates of manufacturing costs require evalu-ation of the requirements for implementing eachcombination of new technologies. Investment bydifferent manufacturers to produce the same ve-hicle will differ because their initial facilities andvehicle designs provide different bases forchange, and because manufacturers have choicesin the timing and extent of major facility renova-tions, in balancing plant renovation and new con-struction, and the selection of new productionequipment—e.g., degree of automation. Differentproduction bases make rapid change more costlyfor some manufacturers than for others.

The variability in actual facilities costs is illus-trated by recent projects associated with newvehicle designs. Chrysler spent over $50 millionto renovate its Newark assembly plant to produce1,120 K-cars per day. * Chrysler made similar al-terations to its Jefferson Avenue (Detroit) assem-bly plant to enable K-car production at the samerate as at the Newark plant, but at a cost of$100million. GM plans to spend $300 million to $500million to build a new Cadillac assembly plant(replacing two old ones) on the Chrysler DodgeMain site in Michigan. The differences in thesespending levels reflect different starting points anddiffering objectives. Since automation, qualitycontrol, and nonproduction aspects of the aboveprojects contribute to other goals in addition tohigher fuel economy, these examples illustratehow difficult it is to infer the specific costs of in-vestments to raise fuel economy.

New Car and Fleet Investments

The incremental investments manufacturersmake to raise automobile fuel efficiency will af-fect the costs of producing new cars and new-car fleets. To gage the effect of changing automo-tive technology on industry investment require-ments, the costs of capacity associated with thehigh- and low-estimate scenarios were estimated

*The project entailed new plant layout; body shop renovation;

conveyor system replacement and rearrangement; assembly tool-ing replacement; installation of automatic and computerized ma-chine welding, transfer, and framing equipment; installation of newpainting, front-end alignment, trim and cushion assembly equip-ment and additional quality control systems.


by weighting and adding the costs of specificchanges. This procedure produces crude esti-mates of the investments necessary to producecars at fuel efficiencies projected in the scenar-ios. * Table 31 presents investment projections byscenario and by 5-year periods, and tables 32-34present the derivation of table 31 in somewhatmore detail.

*Assuming that each technology is applied across the fleet at effi-cient volume, the calculations can be performed on a per-500,000unit basis and scaled up or down to determine overall or impliedper unit investments. The average investment to produce each sizeclass in each period with projected technological characteristicsmay be calculated by weighing the cost of each technology (table30) by its proportion of application and summing the weightedinvestments.

Adding the costs of specific technologies takenseparately—for which cost data are available—isan imprecise way of estimating the costs of tech-nology combinations embodied in new automo-biles and fleets, because it does not capture thecosts of implementing changes together. Very ac-curate investment estimates can be made by eval-uating for specific new automobile designs theplant-by-plant changes in costs (for everythingfrom property and construction to engineeringand equipment to taxes), accounting for variouseconomies (concurrent and sequentially intro-duced technologies may share plant, equipment,even special tooling) and extra costs for minorchanges to the car during production.

Table 31.—Summary of Investment Requirements Associated With Increased Fuel Efficiency

High estimatea Low estimatea

Year Units Large Medium Small Large Medium Small1985-90 . . . . . . . . . . . . $Mil./5OO,OOO 900-1,740 820-1,660 780-1,600 490-1,000 480-1,000 450-1,000

$/car 180-350 160-330 150-320 100-200 100-200 90-2001990-95 . . . . . . . . . . . . $Mil./5OO,OOO 570-1,100 570-1,100 610-1,200 520-940 520-930 500-900

$/car 110-230 110-230 120-240 100-190 100-190 100-1801995-2000 . . . . . . . . . . $Mil./5OO,OOO 350-950 370-980 350-050 480-860 520-930 500-900

$/car 70-190 70-200 70-190 100-170 100-190 100-180aSee table 23 and 24 for definitions of these scenarios.


Table 32.—Average Capital Investments Associated With IncreasedFuel Efficiency by Car Size and by Scenario (1985=90)

Percent of production facilities that incorporatenew technologies or are redesigned

High estimate Low estimateLarge Medium Small Large Medium Small

EnginesSIE a $50-250M/500,000 . . . . . . . . . . . . . . . . . . . . 30 70 95 60 80 100Prechambe b $400-700 M/500,000. . . . . . . . . . . . 15 — 5 15 10Open chamberC $400-700M/500,000 . . . . . . . . . 30

—20 —

TransmissionsFour-speed auto and TCLUd

$300-500M/500,00 . . . . . . . . . . . . . . . . . . . . . . 70 70 70 50 50 50PlatformVarious e $500-1,000 M/500,000 . . . . . . . . . . . . . . 100 100 100 50 50 50Capital costs for technology changes

weighted average)Total $M/500,000. . . . . . . . . . . . . . . . . . . . . . . . . $905-1,740 $825-1,665 $778-1,623 $490-1,005 $480-1,020 $450-1,000Per car (total÷500,000÷10)f . . . . . . . . . . . . . . . $181-348 $165-333 $156-325 $98-201 $96-204 $90-200%ark-ignition engine.bPrechamber diesel.Cown chamber dlegel or open chamber stratified charge.dFour.g~~ automatic with torque converter lockup.weight reduction, material aubstitutlon, aerodynamic and rolling resistance reductions, improved lubricants end assessories.fvehicle change inve9tment9 are divided by 10 tO approximate amortization practices. Forty percent of capital spending goes for plant (30 year) and equipment (12

years), which may together be summarized as “fecllities” and amortized over 15 years (Ford Motor Co. interview) 0.4x3+0.6x 15 = 10.2 or about 10 years.



Table 33.—Average Capital Investments Associated With IncreasedFuel Efficiency by Car Size and by Scenario (1990-95)


High estimate Low estimate

Large Medium Small Large Medium SmallEnginesSIE $400-700 M/500,000 . . . . . . . . . . . . . . . . . . . . 15 25 35 25 35 50Prechamber $400-700 M/500,000 . . . . . . . . . . . . – — — 30 20Open chamber $400-700 M/500,000 . . . . . . . . . . 40

—30 30

TransmissionsCVT, four-speed auto and TCLU

$500 -700 M/500,00 . . . . . . . . . . . . . . . . . . . . . . 35 35 35 50 50 50Engine on-off $500-700 M/500,000 ... , . . . . . . . 15 15 15PlatformVarious $100-400 M/500,000 . . . . . . . . . . . . . . . . 100 50 50100 100 50

Capital costs for technology changesTotal $M/500,000 . . . . . . . . . . . . . . . . . . . . . . . . . $570-1,135 $570-1,135 $610-1,205 $520-935 $520-935 $500-900Per car(total =500,000 +10). . . . . . . . . . . . . . . . $114-227 - $104-187 $104-187 $100-180


Table 34.—Average Capital Investments Associated With IncreasedFuel Efficiency by Car Size and by Scenario (1995.2000)



Large Medium Small Large Medium Small

EnginesSIE $400-700 M/500,000 . . . . . . . . . . . . . . . . . . . .Open chamber $400-700 M/500,000 . . . . . . . . . .

1535

1540

3515

1530

2530

1040

TransmissionsCVT improved $100-400 M/500,00 . . . . . . . . . . .CVT $500-700 M/500,000 . . . . . . . . . . . . . . . . . . .Engine on-off $500-700M/500,000 . . . . . . . . . . .PlatformVarious $100-400 M/500,000 . . . . . . . . . . . . . . . .

50——

50——

50——

3515

3515

3515

100 100 100 50 50 50Capita/ costs for technology changesTotal $M/500,000 . . . . . . . . . . . . . . . . . . . . . . . . .Per car (total÷500,000÷ 10). . . . . . . . . . . . . . . .

$375-985$74-197

$350-950$70-190

$350-950$70-190

$480-865$96-173

$520-935$104-187

$500-900$100-180


The Transportation Systems Center of the U.S. engineering changes that are beyond the scopeDepartment of Transportation, for example, has of this report, and speculative for the 1990’s.developed a “surrogate plant” methodology todo this. The accuracy of this approach is based Note that investment figures presented here ap-on detailed consideration of vehicle designs and ply only to investments required to raise thethe corresponding equipment and plant needs. fuel economy of cars sold in the United States.The methodology requires specific projections of Total capital spending reported by U.S. manufac-

98-281 f) - 82 - 10 : ;~ 3


turers also includes investment in nonautomobileprojects (such as truck and military equipmentproduction), investments in foreign subsidiaries,and spending for normal replacement of worn-out capital and for capacity improvement and ex-pansion.

Note also that not all of a given investment maybe associated with fuel economy improvement.For example, engines, transmissions, and car bod-ies are redesigned periodically. Changes of thissort cannot always be distinguished from thosemade for increased fuel efficiency, so the full costof the investments shown in table 31 should notbe allocated solely to fuel economy improve-ments.

The difficulty of allocating costs when a singleinvestment produces several distinct results is awell-known problem of accounting,26 and thereis no fully satisfactory method for making the allo-cations. For the purposes of the fuel savings costanalysis in the next sections, it is assumed thatthe percentages shown in table 35 represent theshare of investment costs attributable to increasesin fuel economy. Engine and body redesigns aremade for many reasons other than fuel efficien-cy. On the other hand, most of the advancedmaterials substitution (to plastics and aluminum)

Z6A ~. _fho-mas, “The Allocation Problem in Financial Account-

ing Theory” (Sarasota, Fla.: American Accounting Association,1969), pp. 41-57, and A. L. Thomas, “The Allocation Problem: PartTwo” (Sarasoto, Fla.: American Accounting Association, 1974.)

assumed in the scenarios, or automatic enginecutoff, probably would not be incorporated intocars by 2000 without the impetus for increasedfuel efficiency. Advanced transmissions representan intermediate case between these extremes.

The total capital investment associated with theproduction of fleets of given size mix is calculatedby taking an appropriately weighted sum of in-vestments by size class. Assuming that U.S. new-car sales average 11.5 million units in 1985-90,11.7 million units in 1990-95, and 12.1 millionunits in 1995-2000 (conforming to growth ratesprojected earlier in this chapter); and assumingthat imports throughout the 1985-2000 period av-erage 25 percent of all sales (near recent levels),foIlowing the high-estimate scenario for the 15-year period may require $30 billion to $70 billionin investments and R&D expenditures. Follow-ing the low-estimate scenario may require about$25 billion to $50 billion (see table 36). if new-carsales are lower due to continued recession andconsumers’ stagnant real disposable income, thenthe investments would be proportionately small-er. For example, if domestic sales remain at 8 mil-lion vehicles per year (6 million domestically pro-duced) between 1985 and 2000, then capital in-vestments would be about two-thirds as large asshown in table 36 (but R&D costs could remain

Table 36—Total Domestic Capital investments forChanges Associated With Increased Fuel

Efficiencya (billion 1980 dollars)

Table 35.—Percentage of Capital InvestmentsAllocated to Fuel Efficiency

Percentage of investmentCategory allocated to fuel efficiency

Engine. . . . . . . . . . . . . . . . . . . . 50Transmission:

CVT, four- and five-speedauto and TCLV . . . . . . . . 75

Energy storage andengine on-off . . . . . . . . . . 100

Platform:Weight reduction

(body redesign) . . . . . . . . 50Materials

substitution . . . . . . . . . . . 100Composite of all efficiency

related investments:1985-90 . . . . . . . . . . . . . . . . . 55-851990-95 . . . . . . . . . . . . . . . . . 70-801995-2000 . . . . . . . . . . . . . . . 65-75


Time of investment High estimateb Low estimateb

High car salesa

1985-90 . . . . . . . . . . . . . . .1990-95 . . . . . . . . . . . . . . .1995-2000 . . . . . . . . . . . . .

Total. . . . . . . . . . . . . . .

Low car sales1985-90 . . . . . . . . . . . . . . .1990-95 . . . . . . . . . . . . . . .1995-2000 . . . . . . . . . . . . .

Total . . . . . . . . . . . . . . .

14-2910-20

7-18

8-179-16916

31-67

10-207-144-12

21-46

26-49

6-126-116-11

18-34assumptions about car sales:

High car sales1985-90 . . . 11.5 million cars/yr1990-05 . . . . . . 11.7 million cars/yr1995-2000. . . . 12.1 million cars/yr

Low car sales1985-2000. . . . 8 million cars/yr

Estimates also assume that imports average 25 percent of total car salesbetween 1985 and 2000.

bWithin the uncertainties, the Investment requirements are the same for all threesales-mix scenarios.


Ch. 5—Increased Automobile Fuel Efficiency Ž 137— .

the same). If this happens, total investments plusR&D would be reduced by about 25 percent be-low those for the high car sales case.

The greater investments (per vehicle) associatedwith the high-estimate scenario reflect the factthat the scenario contains more extensivechanges more often than the low-estimatescenario. In either case, however, there is nosignificant difference in the total investment forincreased fuel efficiency for the different size-mix scenarios, since the rate of capital turnoverfor increased fuel efficiency is probably adequateto accommodate the mix shifts. *

OTA estimates of cumulative investments im-ply that manufacturers would make capital invest-ments of $2 billion to $5 billion per year (1980dollars) over about 15 years to implement thehigh-estimate scenario and about $2 billion to $3billion per year to implement the low-estimatescenario in the case of high car sales. The corre-sponding figures would be about $1.5 billion to$3 billion and $1billion to $2 billion, respective-ly, for low car sales.

Actual added capital spending levels by manu-facturers are likely to be lower than indicatedbecause some investments in technologies toraise fuel economy will take the place of invest-ments in more conventional technologies thatwould normally be made as plant and equipmentwear out. In fact, deducting the cost of changesthat would have been made under normal cir-cumstances,** but are obviated by or could beincorporated in the investments shown in tables32-34, could reduce the added investment costof implementing the scenarios by two-thirds inthe high estimate and by about 80 percent in thelow estimate, leading to capital investments aver-aging $0.3 billion to $0.7 billion per year for thelow estimate (high car sales) and $0.6 billion to$1.5 billion per year for the high estimate (highcar sales) above “normal. ”

*On the average, over 50 percent of engines, transmissions, andbodies are being redesigned during each 5-year period for increasedfuel efficiency, whereas the mix shift requires 10 to 20 percentchange during each 5-year period.

**Assuming “normal” capital turnover is: engines improved after6 years, on average, redesigned after 12 years; transmissions sameas engines; body redesigned every 7.5 years; no advanced materialssubstitution.

Spending by the automakers will be reducedto the extent that they buy rather than make vari-ous items; to the extent that U.S. suppliers pro-vide purchased items, the total investment levelscan be viewed as spending estimates for U.S.automakers and suppliers together. However,joint ventures with foreign firms, erection offoreign plants with foreign government aid andrelatively labor-intensive designs, and purchasesof parts and knocked-down vehicle kits fromoverseas would all lower investment costs to U.S.firms. So would an increase in import penetra-tion. Finally, note that future levels of normalcapital spending, however they are determined,may be higher than past levels if competition fromforeign manufacturers makes it “normal” fre-quently spend to improve fuel economy and tomodernize facilities.

Fuel Savings Costs

To compare the costs and gains of saving fuelby raising automobile fuel economy with thecosts and gains through other means, it is usefulto express costs in terms of a common measuresuch as dollars per quantity of oil (gallons or bar-rels-per-day) saved. To measure the total dollarsper quantity of oil saved implied by raising fuelefficiency requires estimating changes in variable(labor and materials), fixed capital and R&D costs.

OTA was unable to obtain or develop reliablevariable cost figures for technologies discussedin this report, because information about variablecosts, which vary considerably between compa-nies, is proprietary and speculative for the 1990’s.Four general observations about variable costsof raising fuel economy can be made: First, im-plementing some new technologies, includingcertain weight-reduction measures (e.g., smallerengines and body frames), will lower variablecosts by reducing labor and materials require-ments. Second, automation will lower labor re-quirements. Third, using some new technologies,such as four- and five-speed transmissions andalternative engines, will raise variable costs be-cause they are inherently more complex thanconventional technologies. Fourth, use of newmaterials will raise variable costs. The net changein variable costs is uncertain and will dependheavily on basic materials costs and the success


of adapting the new designs to mass production.Note that variable costs have been about threetimes the level of fixed costs for the average caror light truck.

Based on the percentages shown in table 35,however, table 37 shows the capital investmentattributable to increased fuel efficiency per gallonof gasoline equivalent saved, assuming the aver-age car is driven 100,000 miles and the averageservice life of the investment is 10 years. In allcases, the investment cost is less than $1.00 pergallon saved. If, however, accelerated capitalturnover reduces the useful service life of the in-vestment to 5 years, the costs would be twicethose shown in table 37. Conversely, ifautomobiles are kept longer and driven furtherin the future, then the cost per gallon is reduc-ed. For example, if cars are driven 130,000 milesover their lifetime, on the average, the costswould be 75 percent of those shown in table 37.

The final cost category considered here is theproduct development cost. Although it is difficultto make detailed predictions of the costs of devel-opment, U.S. automobile manufacturers spentfrom 40 to 60 percent as much on R&D (mostly

development) during the 1970’s as they spent oncapital investments.27 During 1978 and 1979,R&D averaged about 40 percent of capital invest-ments.

In order to compare the investments for in-creased fuel efficiency in automobiles with thosefor synfuels, it is convenient to express them asthe investment cost attributable to fuel efficiencyplus the associated R&D expenditures per bar-rel per day oil equivalent saved by these invest-ments. Assuming that development expendituresare 40 percent of capital investment, the COStsin table 37 can be converted to the investmentsshown in table 38 for individual cars and fleet av-erages between 1985 and 2000. The combinedR&D and capital costs appear to increase some-what from the 1985-90 period to the 1990-95 pe-riod. However, technical advances by the early1990’s could prevent further increases during thelate 1990’s.

27G, Kulp, D. B. Shonka, and M. C. Halcomb, “TransportationEnergy Conservation Data Book: Edition 5,” oak Ridge NationalLaboratory, ORNL-5765, November 1981.

Table 37.—Estimated Capital Investment Allocatedto Fuel Efficiency per Gallon of Fuel Saved

High estimate Low estimateCar size: Car size:

Large Medium Small Large Medium Small1985Mpg . . . . . . . . . . . . . . . . . . . . . . . 27 39 48 23 35 45

1990Mpg . . . . . . . . . . . . . . . . . . . . . . . 37 51 62 28 52Gallons saved/yra . . . . . . . . . . . 910 550 430 705 380 270Investment ($/car)b . . . . . . . . . . 100-190 90-180 90-180 60-110 60-110 50-110Dollars per gallonc . . . . . . . . . . 0.11-0.21 0.17-0.34 0,21-0.42 0.084,16 0.15-0.30 0.19-0.411895Mpg . . . . . . . . . . . . . . . . . . . . . . . 43 61 31 45 57Gallons saved/yra . . . . . . . . . . . 340 240 310 200 150Investment ($/car)b . . . . . . . . . . 80-180 80-180 90-180 70-130 70-130 70-130Dollars per gallonc . . . . . . . . . . 0.24-0.51 0.29-0.60 0.37-0.77 0.22-0.42 0,35-0.67 0.44-0.83

Mpg . . . . . . . . . . . . . . . . . . . . . . . 50 85 35 50 65Gallons saved/yra . . . . . . . . . . . 260 210 150 260 200 190Investment ($/car)b . . . . . . . . . . 50-150 50-150 50-150 70-130 70-140 70-140Dollars per gallonc . . . . . . . . . . 0.18-0.56 0.24-0.71 0.33-0.99 0.27-0.50 0.36-0.68 0.36-0.68aFuel consumption of car relative to fuel consumption of comparable car 5 years earlier. Assumes 100,000 miles driven over

life of car and on-the-road fuel efficiency 10 percent less than the EPA rated mpg shown.%he investment attributed to fuel efficiency assuming an average life of 10 yeara for the investment. Also assumes production

at rated piant capacity during the 10 years. Does not include R&D costs, which wouid add about 40 percent to the cost.cThe investment ~r car divided by the fuel saved OVer the life of the car.



Table 36.-Capital Investment Attributed toIncreased Fuel Efficiency Plus Associated

Development Costs per Barrel/Day of Fuel Saved

Capital investment plusNew-car fuel associated developmentefficiency at costs (thousand 1980$end of time per B/D oil equivalent

Car size period a (mpg) fuel saved)b

1985-90Large. . . . . . . . .Medium . . . . . .Small. . . . . . . . .Average Ac . . . .Averge Bc . . . . .

1990-95Large. . . . . . . . .Medium . . . . . .Small. . . . . . . . .Average A= . . . .Average Bc . . . .

1995-2000Large. . . . . . . . .Medium . . . . . .Small . . . . . . . . Average Ac . . . .Average Bc . . . .

28-3741-5152-6238-4843-53

31-4345-6157-7443-5949-65

34-4950-7165-8451-7058-78

19-5135-81

47-1oo21-57d

21-60d

53-12069-16089-20058-120d

64-140 d

44-13057-17078-24048-150d50-150d

“EPA rated 5W45 city/highway fuel efficiency of average car in each size class.bAsgume9 development costs total 40 percent of capital investments and that

a car is driven 10,000 miles per year on averge. A barrel of oil equivalent contains5.9 MMBtu.

c Averages A and B are based on the moderate and large mix shift scenarios,respectively.

d Averages are calculated by dividing average investment fOr technologicalimprovements by fuel savings for average carat end of time period relative toaverage car at beginning of time period. The resultant average cost per barrelpar day is lower than a straight average of the investments for each car sizebecause of mathematical differences in the methodology (I.e., average of ratiosv. ratio of averages) and because extra fuel is saved due to demand shift tosmaller cars. The averaging methodology used is more appropriated forcomparisons with synfuels because it relates aggregate investments toaggregate fuel savings. It should be noted that the cost of adjusting to the shiftin demand to smaller sized cars is not Inciuded. Only those investments whichincrease the fuel efficiency of a given-size car are included. However, giventhe rate of capitai turnover assumed for the scenarios, adjustments to the shiftIn demand probably can be accommodated within the investment costs shown.


Note that the costs and fuel savings benefit ofimproving fuel economy are incurred at differenttimes by different parties. Manufacturer (and sup-plier) investments are made prior to production:30 percent of capital spending occurs 12 to 24months before first production, 65 percent oc-curs within the 12 months preceding production,and 5 percent occurs after production begins.28

R&D costs may occur 5 to 7 years before first pro-duction. Fuel savings begin only after a vehicleis purchased, and they accrue over several years.Fuel savings benefit the consumer directly andthe industry only indirectly.

ZBHarbridge House, Inc., Energy Conservation and the passengerCar: An Assessment of Existing Pub/ic Po/icy, Boston, July 1979.

Consumer Costs

Automotive fuel economy improvements canaffect costs to consumers through changes in realcar purchase prices and changes in real costs ofmaintaining and servicing cars.

Prices

Trends in average car prices are easier to pre-dict than trends in prices for specific car classesor models, because manufacturers have flexibilityin pricing models and optional equipment. Realcar prices (on which consumers base their expec-tations) have been relatively stable over the last20 years (see fig. 12), although nominal car priceshave risen steadily since the mid-1960’s, becauseof general inflation. Labor and materials cost in-creases are not necessarily passed on to consum-ers. For example, the General ManufacturingManager of GM’s Fisher Body Division observedin a recent interview that although raw materialsand labor costs have been rising about 11 to 12percent annually, only 7 to 8 percent of thoseincreases have been recovered through price,with improvements in productivity helping tocontrol costs.

Figure 12.— Real and Nominal U.S. CarPrices, 1960-80

9

7Price in 1980 dollars(corrected

6 with Consumer Price-lrtdex) A

5al

m 4

3

2

1

n1960 1965 1970 1975 1980

Year

SOURCE: Bob Clukas, Bureau of Economic Analysis, U.S. Department of Com-merce, private communication, 1981.


W/here expenses increase faster than prices,manufacturers can still make profits by charginghigher prices for those options or car models forwhich consumer demand is relatively insensitiveto price. This flexibility is eroded when large pro-portions of automotive costs increase due to rapidand extensive change. Some industry analysts ex-pect that through the mid-1980’s, large capitalspending programs and real increases in laborand materials costs will lead to increases in realcar prices of up to 2 percent per year (approxi-mately half of which reflects capital costs); morerapid automotive change might lead to evengreater increases.29

Two percent of today’s average car price(about $8,000) is about $160, although prices ofindividual cars will rise by greater and lesseramounts. OTA’s scenario analysis suggests thatinvestment costs alone for 5-year periods couldrange fro-m $50 to $350 per car (assuming 10-yearamortization periods for plant and equipment).The Congressional Budget Office (CBO), for com-parison, concluded that average automobile pro-duction costs may increase by about $560 be-tween 1985-95 because new technologies willcost that much (per car, on average) to imple-ment. 30 These analyses may overstate the averageamount of capital cost increase, because whennew technologies replace old ones, the capitalcosts charged to old “technologies” should drop

zgMaryann N. Keller, Status Report: Automobile Monthly Vehi-c/e Market Review, Paine Webber Mitchell Hutchins, Inc., NewYork, February 1981.

JoCongressional Budget O f f i c e , h./e/ ~COf10n7Y sta~~a~~s ‘or

Passenger Cars Afier 1985, 1980.

out of the vehicle cost calculation (unless oldequipment is made obsolete prematurely). Ac-tual cost increases will also depend on changesin variable costs, as illustrated below.

Table 39 shows two plausible estimates of con-sumer costs (per gallon of fuel saved) for in-creased fuel efficiency, based on the analysis inthis chapter. The lower costs are calculated as-suming that labor and material (variable) costs areno higher for more fuel-efficient cars than for carsbeing produced in 1985.31 The higher costs in-clude variable cost increases that are twice aslarge as the capital charges associated with in-creasing fuel efficiency .32

Although the range varies from costs that areeasily competitive with today’s gasoline prices tolevels much above those prices, OTA does notbelieve that future variable costs can be predictedwith sufficient accuracy to warrant more detailedestimates of variable costs. Table 39 should there-fore be viewed as illustrative; actual consumercosts will depend on many factors, including thesuccess of new production technologies.

31 Richard L, Strom botne, Director, Office of Automotive FuelEconomy Standards, National Highway Traffic Safey Administra-tion, U.S. Department of Transportation, private communication,1981.

JzRichard H. Shackson and H. James Leach, “Maintaining Auto-motive Mobility: Using Fuel Economy and Synthetic Fuels to Com-pete With OPEC Oil,” Energy Productivity Center, Mellon Institute,Arlington, Va., Interim Report, Aug. 18, 1980. Variable cost changeswere deduced from the estimates of capital investment and changesin consumer costs by assuming an annual capital charge of 15 per-cent of the investment and deducting this capital charge from theconsumer cost estimate.

Table 39.—Plausible Consumer Costs for Increased Automobile FuelEfficiency Using Alternative Assumptions About Variable Cost Increase

Consumer costa ($/gal gasoline saved)

Assuming no variable costAverage fuel efficiency at increase relative to 1985 Assuming variable cost increase equal

Time period Mix shift end of time period (mpg) variable costs of production to twice the capital charges1985-90 . . . . . . Moderate 38-48

Large 43-53 0.15-0.40a 0.40-1 lob1990-95 . . . . . . Moderate 43-59

Large 49-65 0.35-0.85 b 1.10-2.60 b

1995-2000. . . . Moderate 51-70Large 58-78 0.30-0.95 b 0.90-2.80 b

“A”~u~e~ annual ~Wlt~ ~harwa of 015 times ~aplt”l investment ~loc”t~ to fuel efficiency, no discount of future savings, Md car drl~n 1(K),000 miles durtng itS lifetime.bwlthin the uncertalntieg the costs are the same for each mix shift.

SOURCE: Office of Technology Asseaament.


Individual car prices will not necessarily changein proportion to their costs, in any case. Spe-cifically, there are three reasons why it is difficultfor U.S. manufacturers to finance automotivechanges through price increases: competitionfrom lower cost imports, the relationship betweennew and used car prices, and limited consumerwillingness to tradeoff high car prices againstlower gasoline bills. First, because high fuel-economy imports from Japan cost about $1,000(1980 dollars) less than American-made cars,33

U.S manufacturers have little freedom to raiseprices without losing sales volume to imports (allthings equal). International cost differences maynarrow in the future, however, as foreign laborcosts rise and if U.S. productivity increases.

Second, the effective price of new cars for mostbuyers includes a trade-in credit on an older car.Decline in demand for used cars, which mightoccur if older cars were significantly less fuel effi-cient than new ones and if maximum fuel econ-omy were in demand, would effectively raise theprice of new cars. This phenomenon would hin-der a rapid mix shift.

Third, consumers may resist high prices for fuel-efficient cars because they tend to discount suchfuture events as energy cost savings rather heav-ily, by perhaps 25 percent or more,34 Discount-ing at high rates would cause consumers to de-mand relatively large amounts of fuel savings inreturn for a given increase in price. Each gallonsaved seems to cost more if consumers discountat high rates, because discounting reduces theperceived number of gallons saved over the lifeof the car.

This phenomenon can be illustrated as follows:The undiscounted lifetime fuel savings from rais-ing a car’s fuel economy from 45 to 60 mpg is557 gal; at a 25 percent discount rate the dis-counted savings is 227 gal (using a decliningschedule of yearly fuel consumption) or about

33LJ. S, ]ndustria/ Competitiveness: A Comparison of Stee( Ek-tronics, and Automobiles, op. cit.

34 Evidence of high consumer discount rates for energy- efficient

durable goods is presented in an article by J. A. Hausman, ‘Jlndivid-ual Discount Rates and the Purchase and Utilization of Energy-UsingDurables,” Be//Journa/ of Economics, vol. 10, No. 1 (spring 1979),and in a “Comment” article by Dermot Gately in the same jour-nal, vol. 11, No. 1 (Spring 1980).

40 percent of the actual savings. If it costs $150to $350 per car to raise the fuel economy from45 to 60 mpg, the cost per gallon saved wouldbe $0.27 to $0.63 without discounting but about2.5 times as much, $0.66 to $1.54, if fuel savingsare discounted at 25 percent. Consumer behaviormay be at odds with the national interest, becausefuture savings of oil have a relatively low “social”discount rate for the Nation.

Maintenance

Automotive maintenance and service costs mayincrease with vehicle design change but theamount of increase depends on institutional aswell as technological change. Manufacturers aremodifying car designs to make servicing less fre-quent and less expensive, but—with more com-plex and expensive components in cars—thereis a definite potential for increased repair costs.Also, use of new equipment, including electronicdiagnostic units, may lead to higher real costs forservice. For smaller shops, in particular, lack offamiliarity with new technologies and problemswith multiple parts inventories (necessary for serv-icing new- and old-technology cars) could addto consumer service costs. Because dealershipsand larger service firms are in a better positionto adjust to changing technology, they are likelyto gain larger shares of the service market.

Available estimates of service cost changes forfuture cars are very speculative. For instance,CBO has estimated that maintenance and servicecosts associated with transmission improvements,adding turbochargers, and altering lubricantscould raise discounted lifetime maintenance costsof new cars by $40 to $90 on average (assuminga 10 percent discount rate). The actual changesin maintenance costs, however, will dependheavily on the success of development workaimed at maintaining automobile durability withchanging technology.

Electric and Hybrid Vehicles

Costs of producing electric (EV) and hybrid ve-hicles (EHV) will differ from those of producing

conventional vehicles. EVs substitute batteries,motors, and controllers for fuel-burning enginesand fuel tanks. Hybrid vehicles include most or


all of the components of conventional vehiclesas well as EVs, but in modified forms. The com-ponents required for electric propulsion, whichcontribute directly to vehicle cost, further add in-directly to cost because they change the struc-tural requirements of the vehicle. The size andweight of batteries, in particular, increase theneed for space and structural strength,35 necessi-tating changes in vehicle design and weight in-creases.

Batteries are a major source of both direct andindirect cost. They may comprise 25 percent oftotal cost, depending on type, size, and capacity. *Batteries available for electric vehicles by 1990may be priced (1980 dollars) at $1,700 to $2,700(corresponding to production cost of $1,300 to$2,100) while advanced batteries available by2000 may be priced below $2,000 (with produc-tion cost around $1,500).36

Electric motors are smaller, lighter, and simplerthan internal combustion engines. Motor control-lers, however, are relatively bulky and may bemore expensive than the motors themselves, de-pending on their design. Motor-controller combi-nations likely to be available by 1990 may costaround $1 ,000. *

JSCBO, op. cit.*Batteries may comprise about 25 to 30 percent of the weight

of an electric vehicle and about 20 to 25 percent of the weight ofa hybrid vehicle. The electric motor and controller may compriseabout 10 percent of an electric or hybrid vehicle’s weight.

3qA/. M. Carriere, W. F. Hamilton, and L. M. Morecraft, General

Research Corp., Santa Barbara, Calif., “The Future Potential of Elec-tric and Hybrid Vehicles,” contractor report to OTA, August 1980.

*Battery size is a function of the number and size of constituentcells. Battery capacity—and therefore vehicle driving range—is afunction of the amount of energy deliverable by each pound ofbattery.

Estimates given in the report cited in footnote36 suggest that near-term EVs and EHVs wouldcost at least sO percent more than comparableconventional vehicles. Technological advancesin battery development could reduce electric andhybrid vehicle costs, however. Note that manu-facturers may initially set the prices of EVs closeto those of conventional cars to enhance theirappeal to consumers.

Methanol Engines

Production of automobiles designed to run onmethanol entails only minor modifications of theengine and fuel system. Consequently, the costincrease for engines designed to use methanolare minor. However, the cost of modifying an en-gine to operate efficiently on a fuel for which itwas not designed can be more significant. Oneestimate is that retrofitting a gasoline-fueled vehi-cle for methanol use would cost $600 to $900,and redesigning an engine for methanol combus-tion would cost $50 to $100 per vehicle.37 Ford,which is converting several Escorts to methanolcombustion for the Los Angeles County EnergyCommission, estimates that necessary modifica-tions cost about $2,000 per vehicle, although theywould cost less if larger numbers of cars wereconverted.38

J~illiam Agnew, G.M. Research Laboratories, private communi-

cation.36’’Ford Converts 1.6L Escorts for Methanol,” Ward’s Engine Up

date, Feb. 15, 1981.

APPENDIX A.–PROSPECTIVE AUTOMOBILE FUEL EFFICIENCIES

Table 5A-1 summarizes the prospective automobile The table indicates increases in fuel economy of 15fuel-efficiency increases used in OTA’s analysis. The percent at most for vehicles using improved S1 enginestechnologies involved are described in more detail compared with the baseline 1985 car—everything elsebelow. In addition, alternative heat engines are dis- remaining the same. Sources of such improvementscussed and the reasons for not including them in the include:projections to 2000 are explained. ●

More or Less Conventional Engines

There is more diversity among the engine technol- ●

ogies listed in table 5A-1 than in any other category.

smaller engines, because lighter cars will not re-quire as much power and because the enginesthemselves will also continue to decrease inweight;decreases in engine friction–e. g., from new pis-ton ring designs, smaller journal bearing diame-


Table 5A-1.– Prospective Automobile Fuel= Efficiency Increases, 1986.2000

Percentage gain in fuel efficiency—


Technology 1986-90 1991-95 1996-20000 1986-90 1991-95 1996-2000

EnginesSpark-ignition (S1) . . . . . . . . . . . . . . . . . . . . . . . . 10 10-15 15 5 5-1o 5-1oDiesel:

Prechamber. . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 15 15 15Open chamber . . . . . . . . . . . . . . . . . . . . . . . . . 25 35 35 20 20 25

Open chamber (S1) stratified charge (SC) . . . . 15 20Hybrid diesel/SC . . . . . . . . . . . . . . . . . . . . . . . . . 35

TransmissionsAutomatic with lockup torque converter . . . . . 5 5 5 5 5 5Continuously variable (CVT). . . . . . . . . . . . . . . . 10 15 10Engine on-off . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 10 10

Vehicle systemWeight reduction (downsizing and

materials substitution) . . . . . . . . . . . . . . . . . . 8 13 18 4 8 10Resistance and friction (excluding engine)

Aerodynamics. . . . . . . . . . . . . . . . . . . . . . . . . . 2 3 4 1 2 3Rolling resistance and lubricants . . . . . . . . . 2 3 1 1 2

Accessories . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 3 4 1 1 2a Improvements in fuel efficiency are ~~pre~sed a9 percentage gain in mpg compared with an anticipated average 1985 passenger car. The 1985 average car used as

a reference has an inertia weight of about 2,500 lb, is equipped with spark-ignition engine, three-speed automatic transmission, and radial tires, and has an EPA mileagerating (55 percent city, 45 percent highway) of 30 mpg. The fuel efficiencies of the individual baseline cars, which are used to calculate future fuel efficiencies ineach size class, are given in tables 23 and 24. Percentages are given on an equivalent Btu basis where appropriate—e.g., for diasels, which use fuel having higherenergy content per gallon than gasoline, the percentage gain refers to miles per gallon of gasoline equivalent, which is 10 percent less than miles per gallon of dieselfuel. The table does not include efficiency improvements from alternate fuels such as alcohol.

SOURCE: Office of Technology Assessment

●

●

●

ters, increases in stroke-to-bore ratios, improvedengine oils;new combustion chamber designs, particularlyfast-burn chamber geometries that permit leanoperation at higher compression ratios;further refinements to electronic engine controlsystems (although most of the possible gains willhave been achieved by 1985); anddecreases in heat losses, consistent with allow-able thermal loadings of internal engine parts andthe octane ratings of available fuels.

Friction, which goes up with displacement, is a ma-jor source of losses in piston engines. Smaller enginescut friction losses, and also operate with less throt-tling—another source of losses—under normal driv-ing conditions. Turbocharging is one way to make theengine smaller, improving fuel economy without sacri-ficing performance-albeit at rather high cost. Addinga turbocharger can help a small engine meet transientpeak power demands and improve the driving-cyclefuel economy of both S1 and Cl engines by perhaps5 to 10 percent—provided economy and not perform-ance is the goal. Further applications are possible ifthe benefits perceived by consumers outweigh theprice increases.

Bigger gains over the 1985 baseline S1 powered carare possible with Cl engines (table 5A-1). In the past,most efforts on diesels have been directed at heavy-

duty applications such as trucks. Although the efficien-cy advantage of Cl engines relative to S1 engines de-creases as engines become smaller, considerablescope remains for improving the driving-cycle efficien-cy of passenger-car diesels. In particular, all dieselengines now used in passenger cars are based on a“prechamber” design (also termed indirect injection).The combustion chambers in such engines consist oftwo adjoining cavities, with fuel injected into thesmaller prechamber. At present, prechamber engineshave several advantages for passenger vehicles. Theyare quieter than open-chamber (or direct injection)diesels, have wider ranges of operating speeds, andlower-cost fuel injection systems; in addition, emis-sions control is easier and smoke limitations are notas serious. 39

As development of open-chamber diesels for pas-senger cars continues, substantial fuel-economy im-provements can be expected–perhaps 15 percentabove the levels that might be achieved with precham-ber diesels, themselves of course considerably betterthan S1 engines (table 5A-1 )—assuming NO X and par-ticulate emissions can be controlled, and noise heldto acceptable levels. The efficiency advantages of the

open chamber engine stem largely from higher volu-metric efficiency, lower heat losses, and more rapid

39’’ Future Passenger Car Diesels May Be Direct injection, ” Auto-motive Engineering, June 1981, p. 51.


combustion. Estimates40 indicate that a 1.2-liter openchamber diesel in an automobile with an inertiaweight of 2,000 lb should be able to achieve a 55/45EPA fuel-economy rating of 60 to 65 mpg (with a man-ual transmission). Such estimates assume emissionsstandards for Cl engines that do not severely com-promise efficiency. Standards for NO X and for partic-ulates—which, besides making diesel exhaust smoky,are health hazards—are the most difficult to meet. Ingeneral, measures that reduce NO X increase par-ticulate emissions, and vice versa. To some extent,diesel engines will probably face continuing sacrificesin fuel economy to meet emissions standards.

To emphasize efficiency gains rather than differ-ences in the energy content of various fuels, the dieselengine improvements listed in table 5A-1 are all basedon miles per gallon of gasoline equivalent. Becausediesel fuel contains more energy (Btu) per gallon thangasoline, miles per gallon of diesel fuel would be 10percent greater than miles per gallon of gasolineequivalent. For example, a 1990 prechamber dieselis expected to be about 10 percent more efficient thanan S1 engine in the low estimate, but about 20 per-cent better in terms of miles traveled per gallon of fuel.

Stratified-charge (SC) engines (table 5A-1 ) are S1 en-gines that have some of the advantageous features ofdiesels–such as potentially higher efficiency and po-tentially easier control of emissions, although lowemissions levels have been difficult to achieve in prac-tice—as well as the disadvantages of diesels, such ashigher production costs. 41 SC engines, like diesels,operate with a heterogeneous distribution of fuel andair in the combustion chamber. But unlike diesels, theSC engines now in production burn gasoline and usespark plugs. SC engines, again like diesels, come intwo varieties–prechamber, such as the Honda CVCCengine that has been sold in the United States since1975, and open-chamber (also called direct injection).Prechamber engines have shown little if any fuel econ-omy advantage, while open chamber SC enginespromise good efficiencies in theory but have not yetbeen successfully reduced to practice. As with open-chamber diesels, it has proven difficult to achievegood response and smooth operation over the rela-tively wide range of loads and speeds needed for pas-senger cars. Moreover, open-chamber SC engineshave the most potential in larger engine sizes; for asmaller engine operating at a higher load level—a situ-ation now more prevalent—one of the major advan-tages of the SC engine, its lower throttling require-ment, is less of a factor. Such an engine would be

401bid.41J, A. Alic, op. cit.

more costly to produce than a conventional S1 engine,though less expensive than a diesel.

Another potential advantage of open-chamber SCengines is their tolerance for a wide range of fuels—including both gasoline and diesel, as well as alcoholsand other energy carriers not necessarily based onpetroleum. The broad fuel-tolerance of SC engines hasled to a good deal of work directed at military applica-tions. Further, the low-emissions potential of SC en-gines provided early stimulus for R&D directed at auto-motive applications. Open-chamber SC engines couldfind a place in passenger cars during the 1990’s if theremaining problems are overcome.

Another possible path leads to a merging of dieseland SC engine technologies (table 5A-1). This mightbe visualized as a diesel with spark-assisted ignition.Spark-ignition would increase the tolerance of the en-gine to fuels with poor ignition quality (i.e., to fuelswith a low cetane numbers such as gasoline or alco-hols), but the combustion process would be morenearly a constant pressure event, as in a diesel.

Gas Turbine, Brayton,and Stirling Engines

Prospects for other “alternate engines” remain dim;in particular, most alternatives to S1 and Cl enginesare poorly suited to small cars. Candidates include gasturbines (i.e., those operating on a Brayton cycle), orthe Stirling cycle powerplants that have also beenwidely discussed for automotive applications. At pres-ent, such alternatives to S1 and Cl engines suffer manydrawbacks. Gas turbines, for example, would needceramic components to achieve high efficiencies atlow cost–most critically in the power turbine, be-cause high turbine inlet temperatures are needed toraise the efficiency. Ceramics are inherently brittle,and a great deal of work remains to be done beforedurable and reliable engine parts can be mass-pro-duced from materials such as silicon nitride. The tech-nical problems are more severe for highly stressedmoving parts such as turbine rotors than for the appli-cations such as combustor heads envisioned for Stirl-ing engines. While the problems of developing toughceramics for high-temperature applications in energyconversion devices are receiving considerable R&Dsupport, success cannot be guaranteed. Even if the ce-ramics can be developed successfully this will not nec-essarily suffice to make gas turbines (or Stirling-cyclepowerplants) practical for use in passenger cars.

With or without ceramic components, automotivegas turbines would, at least initially, be high in cost;and beyond high costs, they suffer a number of otherdisadvantages as automobile engines. Although gas

Ch. 5–Increased Autornobile Fuel Efficiency ● 745

turbines are highly developed powerplants in the largesizes used for stationary power or for marine and air-craft applications (500 hp and above) and ceramiccomponents would allow higher operating tempera-tures and theoretically high efficiencies, turbineengines do not scale down in size as well as recipro-cating engines. Both compressors and power turbineslose efficiency rapidly as their diameters decrease to-ward the sizes needed for smaller cars (75 hp andbelow). Brayton-cycle powerplants also have generallypoor part-load fuel economy–which is a severe dis-advantage in an automobile, where low-load opera-tion is the rule. Furthermore, they need complex trans-missions because the power turbine runs at speedsmuch higher than those of reciprocating engines.Fixed-shaft turbines, in particular, pose difficult prob-lems in matching engine operating characteristics toautomobile driving demands. But the most criticaldrawback of gas turbine powerplants is finally thatthey are unlikely to achieve competitive efficiencieswhen sized for small cars—those in the vicinity of2,000 lb. As these size classes become a larger frac-tion of the market, the prospects for automotive gasturbines grow dimmer.

Stirling-cycle engines are at much earlier stages ofdevelopment. High efficiency in small sizes is a morerealistic possibility for a Stirling engine than for a gasturbine, but the costs of Stirling engines are likely tobe even higher than those for gas turbines 42–and bothengines will probably always be more expensive tomanufacture than S1 engines. Like turbines, ceramiccomponents will be needed to achieve the best possi-ble efficiencies in Stirling-cycle powerplants–here themost immediate needs are probably in the heater headand preheater. Seals have also been a persistent blockto practical Stirling-cycle powerplants.

Both gas turbine and Stirling engines–because com-bustion is continuous–have intrinsic advantages inemissions control, and can burn a wide range of fuels.But intermittent-combustion engines (e.g., S1 and Cl)have thus far demonstrated levels of emissions con-trol adequate to meet regulations. Broad fuel toleranceis again not unique to gas turbine and Stirling engines.These advantages are probably not enough to over-come the drawbacks of such engines, at least over thenext 20 years.

Transmissions

Table 5A-1 lists a pair of developmental paths forautomatic transmissions. (Manual transmissions arenot explicitly included in the table; although more

42’’ Alternate Powerplants Revisited, ” Automotive Engineering,February 1980, p. 55.

American purchasers are now choosing manual trans-missions as small cars take a greater share of the mar-ket, automatics still predominate.) Geared automatictransmissions with lockup torque converters are al-ready available in some cars; these are straightforwardextensions of current technology, in contrast to contin-uously variable transmissions (CVTs). In principle, anengine on-off feature—in which the powerplant canbe automatically shut off when not needed–could beimplemented with either system (or with manual trans-missions). Placing the engine drive shaft parallel towheel axles would also yield a small improvement infuel economy–because crossed axis gears could bereplaced by more efficient parallel axis gears, orchains.

Geared automatic transmissions with either three orfour speeds and a lockup torque converter-or witha split power path, an alternate method for m minimiz-ing converter slip and the consequent losses—arealready on the market. A fourth gear ratio gives a bet-ter match between engine operating characteristicsand road load demands. The fourth speed, for exam-ple, may function as an “overdrive” to keep engineload and efficiency high at highway driving speeds.Neither development—bypassing the torque converterwhen possible, or adding a fourth speed to an auto-matic transmission—is new, but the added costs ofsuch designs are now more likely to be judged worth-while. Many manual transmissions incorporate fiverather than four speeds for similar reasons—the addedgear benefiting fuel economy, as well as performanceat low power-to-weight ratios.

Although the efficiencies of manual transmissionsare greater than for automatics—that is, less of thepower passing through the transmission is dissipated–the fuel economy achieved by many drivers may beas high or higher in cars equipped with an automatictransmission. By relying on the logic designed into thetransmission to chose the appropriate gear ratio forgiven conditions, wasteful driving habits–e.g., usinghigh engine speeds in intermediate gears–can oftenbe avoided.

Further improvements in the control systems forautomatic transmissions, as well as other changes suchas variable displacement hydraulic pumps, will helpto counterbalance their inherently lower efficiencies.In the past, automatic transmissions have dependedon hydromechanical control systems—just as engineshave. Hydromechanical control–although well devel-oped and effective— limits the number of parametersthat can be sensed, as well as the logic that can be

employed. In the past, automatic transmissions havegenerally decided when to shift by measuring enginespeed, road speed, and throttle position. By movingto fully electronic control systems, a greater number

146 • Increased Automobile Fuel Efficiency and Synthetic Fuel: Alternatives for Reducing Oil /reports

of engine parameters can be measured, and moresophisticated control algorithms implemented—ena-bling the transmission to be “smarter” in selectingamong the available speeds. Electronics might also beused with manual or semiautomatic transmissions tohelp the driver be “smarter.”

As pointed out above, increasing the number ofspeeds in an automatic or manual transmission—fromthree to four or five–can help fuel economy. Althoughtrucks often have many more speeds (for reasons be-yond fuel economy), mechanical complexity (in auto-matics) and the demands on the driver (for manualtransmissions)—as well as rapidly diminishing returnswhen still more speeds are added—will probably con-tinue to limit the number of discrete gear ratios inpassenger-car transmissions to four or five. if, how-ever, discrete gearing steps can be replaced by a step-Iess CVT, then the engine could operate at the speedand throttle opening (or fuel flow for a diesel) thatwould maximize its efficiency for any road-load de-mand—i.e., engine speed would be largely independ-ent of vehicle speed. 43 If otherwise practical, such atransmission could give markedly better fuel economythan other automatic transmissions–provided the CVTitself was reasonably efficient. Smooth, shiftless opera-tion is another potential advantage of CVTs.

Continuously variable speed ratios can be accom-plished in a variety of ways—e.g., the hydrostatic trans-missions sometimes used in farm and constructionequipment. A series hybrid electric vehicle—in whichthe engine drives a generator, with the wheelspowered by an electric motor, typically drawing frombatteries as well as the generator–in effect uses themotor-generator set as a CVT. Hydrostatic or electricCVTs are expensive and inefficient. CVTs used in pastapplications to passenger cars have generally been all-mechanical—e.g., based on friction drives, or belts.Typically, such designs have been limited in powercapacity and life by wear and other durability/reliabil-ity problems.

At present, the most promising CVT designs arethose based on chains or belts. Continuing develop-ment may well overcome or reduce the significanceof their drawbacks relative to the fuel savings possi-ble. These fuel economy improvements could be ofthe order of 10 percent compared with a conventionalautomatic transmission—again, depending on the effi-ciency of the CVT. Fuel economy better than that ofa properly driven car with a manual transmissionwould be more difficult to achieve. Although the CVT

43f3. C. chri~tenson, A. A. Frank, and N. H. Beachley, “The Fuel-

Saving Potential of Cars With Continuously Variable Transmissionsand an Optimal Control Algorithm, ” American Society of Mechani-cal Engineers Paper 75-WAIAut-20, 1975.

would permit the engine to operate more efficiently,poorer transmission efficiency would counterbalanceat least some of the savings. One reason that CVTsare expected to have lower efficiencies is the needfor a startup device, such as a torque converter, in ad-dition to the CVT mechanism itself. Given that the pro-duction costs of a CVT would also be higher thanthose of a manual design—in part because of the start-up device—CVTs appear most likely to find a placeas replacements for conventional automatic transmis-sions.

The third transmission technology listed in table5A-1, engine on-off, has been placed in this sectiononly for convenience—it could just as well appear inthe engine category. “Engine on-off” systems, bywhich the powerplant can be automatically shut offduring coasting or when stopped at signal lights or intraffic, are in principle easy to implement. Indeed,when current engines are equipped with electronicfuel injection the fuel flow is sometimes cut off whencoasting above a predetermined speed. For an engineon-off design to be practical (and safe), the enginemust restart quickly and reliably, and the operationof the system should not otherwise affect driveability—i.e., it should be operator-invisible, primarily a mat-ter of control system design. Engine on-off systems areunder development, and presumably will be imple-mented if the production costs prove reasonable com-pared with the expected fuel savings.

Vehicle Weight

The final group of technologies in table 5A-1–vehi-cle systems—includes several means of reducing pow-er demand, hence the fuel consumed in moving thecar. As discussed in the body of this chapter, the singlemost important means of reducing fuel consumptionis by reducing the weight of the vehicle; a 1 -percentdecrease in weight typically cuts fuel consumption by0.7 to 0.8 percent, provided engine size is reducedproportionately. Fuel consumption can also be les-sened by reducing air drag, frictional, and parasiticlosses–such as accessory demands.

The easiest way to decrease the weight of an auto-mobile is to make it smaller. In most newly designedcars, front-wheel drive is adopted to preserve interiorvolume, while considerable attention has been givento maximizing space utilization and removing un-needed weight. For cars of a given size, materials withhigher strength-to-weight ratios can be used wherecost effective, provided they meet requirements forcorrosion resistance and stiffness. Progress has alsobeen made by specifying less conservative margins ofsafety for structural design. Many of the steps takento reduce vehicle weight interact—i.e., taking weight


out of one part of the car, perhaps by replacing 5-mphbumpers with 2-mph bumpers, allows secondaryweight savings elsewhere in the body and chassis.

In the future, big gains will be harder to achieve.Most of the waste space has already been taken outof newly designed American cars. Overhangs for styl-ing purposes are being reduced or eliminated, doorthicknesses decreased, space utilization in passengercompartments and trunks more carefully planned.Considerable progress can still be made through care-ful detail design, but the easiest steps are being taken.In the future, tradeoffs between space for passengersand luggage and the weight of the vehicle will be moredifficult to manage.

The two basic approaches to reducing weight are:1 ) to use materials which provide comparable per-formance characteristics but weigh less; and 2) todesign each component and subsystem with minimumweight as a primary objective—the latter more impor-tant now than in the past, when the costs associatedwith extra testing and analysis were harder to justifythrough savings in materials and fuel. The first pathalso tends to raise the manufacturer’s costs becausesubstitute materials usually cost more.

Material characteristics most critical in automobilestructures are cost, strength, stiffness, and corrosionresistance. Costs—of the material itself, and of the fab-rication processes that the choice of material entails—are in the end the controlling factors, as for most massproduced products. Nonstructural parts carry differentdemands but often less opportunity for saving weight(e.g., upholstery and trim, typically already plastics).

Iron and steel have been the materials of choice forbuilding cars and trucks–as for other mechanical sys-tems—because of their combination of good mechani-cal properties and low cost. Iron castings have beenwidely used in engines and powertrain components,steel stampings and forgings in chassis members,bodies, and frames (now often unitized). Highlyloaded parts are generally made from heat treated al-loy steels–e.g., some internal engine components, aswell as gears, bearings, shafts. But elsewhere mild steelhas been chosen because it is cheap and easy to fabri-cate; it can be easily formed and spot welded, givesa good surface finish, and takes paint well. Greaterquantities of high-strength, low-alloy steels are nowbeing specified–particularly for bumpers and morecritical structural applications such as door guardbeams; some new cars contain 200 lb of high-strengthsteel, triple the amounts of a decade ago. 44 Thinner

44H. E. Chandler, “Useage Update: Light, Longer Lasting SheetSteels for Autos, ” Meta/ Progress, October 1981, p. 24.

body parts with good corrosion resistance can bemade from galvanized or aluminized sheet steel.

The strength-to-weight ratio of inexpensive, low-strength steels can also be equaled or exceeded byalloys of aluminum and magnesium, as well as by non-metallic materials such as reinforced plastics. Alumi-num usage, now 115 to 120 lb per car, is expectedto reach 200 lb per car by 1990—mostly in the formof castings .45 Aluminum can substitute for iron andsteel in engines and transmissions—cylinder heads aswell as simpler, less critical components such as hous-ings, covers, and brackets. Magnesium and reinforcedplastics are other candidates for some of these appli-cations (use of magnesium is currently limited by highcosts, but these may come down in the future). How-ever, aluminum sheet for body parts and structuralmembers has been limited not only by high costs butby difficulty in spot welding–a problem that new alloycompositions are helping overcome.

A variety of plastics and fiber-reinforced composites–ABS, glass-reinforced polyester sheet molding com-pound, reaction-injection molded polymers with orwithout reinforcement—are being specified for pro-duction parts; some have been used for years. Whilemore of these materials will be used in the future,GM’s Corvette–a high-priced specialty vehicle madein quantities small compared with most other domes-tic vehicles—remains the only mass-produced Ameri-can car with a glass-reinforced plastic body. Intro-duced nearly 30 years ago but never emulated, thisillustrates the continuing advantages of metals, particu-larly at high production levels.

In the past, plastics have generally been applied tononstructural parts. Polymer-matrix composites arenow candidates for some structural applications—one1981 car had a fiberglass rear spring weighing 8 lb,compared with 41 lb for the steel spring it replaced. 46

Examples of related applications, none yet in produc-tion, are driveshafts and wheels. Other types of com-posites—i,e., laminates consisting of two metal layers,probably steel, sandwiching a plastic such as polypro-pylene–may have potential as body materials. Thethickness of the laminate makes it more rigid in bend-ing for a given weight, and the plastic dampens noiseand vibration; such laminates, like many other com-posite materials, are now too costly for widespreaduse. 47

4SH E chandler, “A Look Ahead at Auto Materials and f% OCeSSW

in the 80’s,” Meta/ Progress, May 1980, p. 24.4GI bid,47H. S. Hsia, “Weight Reduction for Light Duty Vehicles, 1980

Summary Source Document” (draft), Department of Transporta-tion, Transportation Systems Center, March 1981, p. 8-31.


Improvements in steels and their applications stilloffer the greatest scope for near-term weight savingsin passenger cars—one reason is that the designer’sjob becomes more difficult when materials arechanged. But the manufacturing problems mentionedabove for aluminum as a body material—difficulty inspot welding, forming characteristics that call forchanges in die design—at least remain within therealm of conventional, mass production metalwork-ing techniques. Plastics and composites demand proc-essing quite different from that used for metals—andhigh volume production of structural parts made fromsuch materials is new, not only for the automakers,but for virtually all industries. Furthermore, unconven-tional materials may need additional engineering anal-ysis and testing–e.g., they are often susceptible to dif-ferent failure modes (such as environment-inducedembrittlement, or discoloring). In fact, the secondavenue for weight reduction is precisely an improve-ment in design methods.

With better methods for analyzing and controllingthe stress and deflection in the vehicle structure,weight can be reduced without sacrificing structuralintegrity. Through better understanding of the failuremodes of the materials used, as well as service load-ings, margins of safety can be reduced. Both analysisand testing are important to these objectives. Thegreatest strides have come from widespread adoptionby the automobile industry of finite-element methodsfor structural analysis, Not only can body and chassisstructures be designed with more precise control overstresses and deflections—eliminating unnecessary ma-terial—but finite-element techniques can also help re-duce the weights of engines and other powertraincomponents; section sizes of engine blocks can bedecreased, for example.

The weight reductions, hence fuel economy gains,that are possible through materials substitution arelimited primarily by costs–both material cost andmanufacturing cost. Graphite reinforcements for po-lymers perform better than glass, for example, but areconsiderably more expensive; at the highest strengthlevels, aluminum alloys, in addition to being expen-sive and difficult to form, cannot be welded. If thecosts justify the benefits in terms of fuel economy andother performance advantages—e.g., corrosion resist-ance—then automobile designers will choose new ma-terials. In some cases, costs will come down as pro-duction volume increases, but there will always be apoint of diminishing returns. Nonetheless, continuedattention to detail design with conventional materials—with which the automakers have engineering andproduction experience–and improved methods ofstructural analysis, can give substantial reductions in

weight, as table 5A-1 indicates. Even though downsiz-ing and weight reduction will have proceeded consid-erably by 1985, improvements will continue—oftenrather gradually, as manufacturers gain confidence in,and experience with, new materials and improved de-sign methods.

Safety poses a further constraint on the selection ofstructural materials for automobiles. The tradeoffs be-tween vehicle size and occupant safety are discussedin chapters 5 and 10. For a vehicle of a given size,the mechanical properties of the structural materialsare one of the factors on which passenger protectiondepends. Because the structure needs to be able toabsorb large amounts of energy in a collision, thematerials should be capable of extensive plastic defor-mation, or else able to absorb energy by some alter-native process such as microfracturing while beingcrushed. This may limit applications of higher strengthmaterials—both metals and nonmetals—because ca-pacity for plastic deformation is inversely proportionalto strength; it may also pose difficult design problemsfor some composite materials.

Aerodynamic Drag

A lighter automobile needs less power for accelera-tion and for constant speed travel and therefore con-sumes less fuel. A car with less aerodynamic dragburns less fuel at any given speed, but the powerneeded for acceleration is not directly affected.Because drag caused by air resistance is proportionalto frontal area and to speed squared, drag reductionhelps most at higher speeds–i.e., during highway driv-ing.

Smaller cars have less frontal area, hence less drag.But drag can also be reduced by making a car more“streamlined. ” This characteristic is quantified by thedrag coefficient-which has a value of 1.2 for a flatplate pushed through the air, but only 0.1 for a tear-drop shape. For complex geometries such as airplanesor automobiles, drag coefficients can be preciselydetermined only by experiment. Extensive–and ex-pensive—wind tunnel testing is the basic techniquefor minimizing the drag coefficient of an automobile.

Theoretical aspects of the aerodynamics of groundvehicles are poorly understood, particularly for shapesas complex as automobile bodies. Interactions be-tween the stationary roadway and the moving car area particular problem. Drag reductions are sensitive notonly to overall vehicle shape—e.g., the sloped front-ends now common on passenger cars—but to relative-ly subtle details–such as integration of the bumpersinto the front-end design, and the flow of air through


the radiator. Testing and experiment are required be-fore the final form can be chosen.

While typical cars of the early 1970’s had drag coef-ficients in the range of 0.50 to 0.60, many currentmodels have values closer to 0.45, or less; the 1982Pontiac 6000 has a claimed drag coefficient of 0.37. 48

Reductions to values of less than 0.35 are possible,but eventually limited by practical compromises in-volving the utility of the vehicle (passenger and lug-gage space can suffer, as well as accessibility for re-pairs), safety (a streamlined design may compromisevisibility for the driver), and manufacturing costs(curved side glass can cut drag but is more expensive).’Even so, by 1990 drag coefficients may average 0.35or less. 49

Nonetheless, as table 5A-1 indicates, improvementsin fuel economy in the years past 1985 from continu-ing reductions in aerodynamic drag will be relativelysmall. The reasons are, first, that considerable progresshas already been made, and more can be expectedbetween now and 1985–and the returns from dragreduction rapidly diminish (frontal areas are con-strained by the need to fit people into the car; no prac-tical vehicle could approach the lower limit drag co-efficient of the teardrop shape) —and, second, thatdrag reduction has the greatest benefits at high speeds,whereas most driving is done at lower speeds. In gen-eral, a 10-percent reduction in drag will yield an im-provement in driving-cycle fuel economy of perhaps2 percent. 50

Rolling Resistance

Even in the absence of air resistance, some fuelwould be burned in pushing a car at a constant speed.This rolling resistance depends on tire characteristics,

4BJ. Burton, “82 Pontiac 6000: A Step Further, ” Autoweek, Nov.30, 1981, p. 10.

49D. Scott, “Double LOOp cuts Wind Tunnel Size and Cost, ” Auto-

motive Engineering, May 1981, p. 69.SoAutomotive Fue/ Economy Program: Fitih Annual Report to the

Congress (Washington, D. C.: Department of Transportation, January1981), p. 119.

and on friction and drag in moving parts such as axlebearings. It also depends on the road surface (con-crete offers slightly less rolling resistance than asphalt).Most of the resistance is caused by deformation in thetires. Carcass design, tread pattern, and inflationpressure all affect resistance. Radial tires decrease re-sistance compared with bias-ply carcasses, with fuel-economy improvements of 2 to 5 percent possible; 51

more aggressive tread patterns—e.g., snow tires—in-crease resistance; higher inflation pressures decreaseresistance.

Improved lubricants and bearing designs can also cut resistance slightly, as can brakes with minimaldrag. However, more scope for fuel-economy im-provements through better lubricants exists elsewherein the vehicle—particularly in engines, but also intransmissions and rear axles—where more “slippery”oils, as well as design changes that minimize churn-ing and oil spray, can reduce viscous drag. Althoughdecreases in friction and rolling resistance benefit fueleconomy at low speeds almost as much as at high,many of the possible gains have already beenachieved, or are in sight—thus, further improvementsafter 1985 will be small (table 5A-1).

Accessories

Some of the power produced by the engine is used,not to move the car or to overcome the engine’s inter-nal friction, but in driving pumps, fans, and acces-sories. To produce this power, fuel must be burned.Among the specific parasitic losses that automobiledesigners strive to minimize are those associated withcooling fans, air-conditioning compressors, power-steering pumps, and electrical loads supplied by thealternator. Decreases are possible in many of these,as table 5A-1 indicates, though often at somewhatgreater cost. In some cases, downsizing the vehiclehelps to reduce or eliminate parasitic Iosses–e.g.,power steering may not be needed.

51 Ibid.

APPENDIX B.–OIL DISPLACEMENTPOTENTIAL OF ELECTRIC VEHICLES

Electric Vehicles and Electric Utilities petroleum-based electricity for vehicle recharging andthe fuel consumption of the car the EV replaces. Be-

The extent to which electric vehicle (EV) technology cause of the limited performance of EVs, they wouldcan contribute to the national goal of reducing oil im- most likely be substitutes for relatively fuel-efficientports will depend on the availability and use of non- small cars. Also, because of the limited range and


hauling capacity of EVs, it is assumed that they canreplace only 80 percent of the (10,000 miles per year)normal travel in a gasoline car. The remaining 2,000miles per year would have to be accomplished witha possibly rented gasoline-fueled car, which might beless fuel efficient than the small car that the EV re-placed. This latter complication was ignored, how-ever, so the results shown here are slightly more favor-able in terms of net oil displacement than might bethe case in practice.

Figure 59-1 shows the consequences of introduc-ing EVs in terms of either increased petroleum use,or net petroleum savings, for alternative assumptionsabout automotive fuel economy. For example, refer-ring to the figure, if the fuel economy of the car re-placed by an EV is 60 mpg (case A), then one couldsave as much as 133 gal per year or increase petrole-um consumption by 123 gal (of gasoline equivalent)per year depending on whether, respectively, all ornone of the recharge electricity is petroleum-based. *As long as the fraction of petroleum used for generat-ing recharge energy for EVs in this case is less thanabout 50 percent, the introduction of EVs will resultin net petroleum savings. In case B, where a car thatachieves 40 mpg is replaced by an EV, the fractionof petroleum used for generating recharge energymust be less than about 80 percent to result in netpetroleum savings.

Utilities plan their capacity and operations to en-sure that the maximum instantaneous demand on thesystem, typically occurring at midday, can be met. This

*Assumes that electric vehicle recharge energy is 0.4 kWh/mile.

Figure 5B-1.-

implies that peakloads are satisfied with generatingcapacity that is idle at other times. A utility will thusrespond to demand fluctuations by using the most effi-cient (“baseload” as well as “intermediate”) plantsas much as possible and progressively adding other“peaking” plants as loads increase. Baseload plants,which often cannot be adjusted rapidly (i.e., under2 hours) to respond to demand fluctuations, are eithernuclear, hydro, geothermal, or steam (oil, coal, or gas).Peaking plants can be operated for short-term re-sponse and are gas turbines fueled by oil or naturalgas and pumped-storage hydro. The ability of utilitiesto handle the additional load created by EVs will de-pend on such factors as total generation potential, theequipment and fuel mix, and the time pattern of de-mands. These characteristics generally vary by region(fig. 59-2) as illustrated in table 59-1,

Figure 59-3 shows a peak summer demand curveand equipment mix for an individual, representativeutility. Also shown are the likely changes in the loadprofile that would occur with the addition of EV loadsunder the following conditions: 1) recharging occursover 12 hours during the night when demands on thesystem are the smallest, 2) recharging occurs uniformlyduring the day, and 3) recharging occurs during 2hours at midday. As long as the additional EV loadoccurs either at night or evenly throughout the day,this load could be accommodated by increasing base-Ioad output. Recharging over 2 hours during the daywould be satisfied with peaking plants.

Assuming that the available oil-fueled baseloadcapacity used for recharging EVs is proportional to theamount of oil-fueled baseload in the system, figure

‘Relationship Between the Net Fuel Savings From the Use oand the Fuel Used for Electric Generation

f EVs

I

Ch. 5—Increased Automobile Fuel Efficiency Ž 151

Figure 5B-2.— Regional Electric Reliability Council Areas

Council -

SOURCE: Department of Energy, Energy Information Administration, April 1978

5B-4 can be used to determine the fuel efficiency thatwould be required of a small automobile if the overaIloil consumption of the small automobile is to beequivalent to an EV in each region. For example, atone extreme, the Texas region uses not oil in its base-Ioad, so an EV always consumes less oil, and at theother extreme, in the northeast a gasoline-fueled carwould have to get about 50 mpg if it were to consumean equivalent amount of oil as an EV. In terms of pre-mium fuel, * the extreme points are several hundredmpg in the midcontinent area and about 40 mpg in

the Texas region to achieve a fuel-use equivalence be-tween a small car and an EV.

The electricity requirements and oil/premium fuelsavings for an EV fleet which constitutes 20 percent*of the total vehicle fleet are shown in table 5B-2. Ascan be seen, as long as the EV fleet can be rechargedusing baseload capacity, regions should be able tomeet the additional load with existing available base-Ioad capacity. In general, the Northeast, West, andSoutheast regions would utilize the greatest absoluteamounts of oil-fueled baseload capacity if EVs were

*A 20-percent market penetration is considered to be the upper*Oil + natural gas = premium fuel. bound on EV use through 2010,

~ B -291 ‘J - E 2 - 11 : ; 1, 3


Table 5B-1.—Utility Capabilities by Region (contiguous United States) a

Percent of baseloadInstalled capacity Net capability Available baseload Available peaking that is fueled by:

Region b (x1O 3 MW) ( x 1 O3 MW) C capacity (x10 3 M W ) d capacity (x 1O 3 M W ) e

Oil GasECAR . . . . . . . . . . 84.9 79.1 19.6 1.3 6.7 0.2MAAC . . . . . . . . . . 44.0 40.6 7.1 2.2 35.3 0.0MAIN . . . . . . . . . . 43.1 39.9 7.6 0.6 12.7 1.1MARCA . . . . . . . . 25.4 24.4 5.1 0.8 2.5 0.9NPCC . . . . . . . . . . 50.9 49.8 11.4 2.3 60.4 0.0SERC . . . . . . . . . . 113.7 103.1 17.5 2.6 17.9 0.2SWPP . . . . . . . . . . 48.9 46.1 6.8 0.4 21.6 42.8ERCOT . . . . . . . . . 40.9 39.9 9.7 0,0 0.0 75.7WSCC . . . . . . . . . . 94.6 93.3 22.0 1.4 26.9 2.3

Total . . . . . . . . . 546.4 516.2 106.8 11.6 20,9 10.8%nless otherwlee indicated, data are taken from the 19M Summary, National Electric Reiiebility Council, Juiy 19S0.bSW att=h~ map in fia. 5B”2.

cNet cap~ility is calculated based on the ratio of Net Capability to Installed Capacity as reportad in the Electrlc paver A40rrtlr/Y, U.S. Department of l%erw, EmmyInformation Administration, August 19S0.

dCalculation9 eggume that the total avaiiable capacity is atlocated between baseload and peaking capacity according to the ratio of peaking to baseioad CaPacltY withinthe system. Total availabie capacity - (net capability) – (peakload).


Figure 5B=3.— Illustrative Load Profile With and Without Electric Vehicles

SOURCE: Office of Technology Assessmen\


Figure 5B-4.— Dependence of Net Oil (premium fuel)Consumption on Efficiency of Gasoline Car

Replaced and Fuel Used in Baseload ElectricGeneration

Petroleum savings accruing from the substitution ofEVs (as opposed to 60-mpg gasoline-fueled vehicles)for 20 percent of the total vehicle fleet would be ap-proximately 0.1 MMB/D of oil and 0.07 MMB/D of

Gasoline-fueled car consumesless oil (premium fuel) than EV

est amount of oil savings would occur in the Texas,Southwest, and midcontinent regions; substituting EVsfor small cars in the Northeast would actually increaseoil usage. The greatest premium fuel savings accru-ing from the substitution of small cars would occurin the East-Central, Southeast, and West regions. In-creased premium fuel use would occur in Texas, theNortheast, and the Southwest. In the future, however,both oil and premium fuel savings with EVs will in-crease as utilities switch away from the use of thesefuels for electric generation.

Analyses conducted at the national and regional lev-els cannot be used to assess the attractiveness of EVsfor individual cities or utilities. For example, individualutilities may experience significant increments to theirloading, and hence, require a change in baseload ca-pacity and/or mix of fuel use, depending on the timepattern of recharging assumed, the percentage ofmarket penetration, and the technical characteristicsof the battery and charging system (e. g., amperage,voltage, and efficiency profiles).

Table 5B.2.—Electricity Requirements and Oil Savings With an Electric Vehicle Fleet With 20= Percent Penetration

Fuel saved by replacingElectricity capacity required 20°A of small carsif 20°/0 penetration by EVs b

Fuel consumed with EVs h

as percent of available Case Af baseload capacity by fleet of PremiumTotal vehicles baseload (percent) that would be fueled by: small cars g Oil saved fuel saved

Region 1979 a (x 10 6) Case A c Case B d Case C e Oil (MW) Premium (MW) (MMB/DOE) (MMB/DOE) (MMB/DOE)

ECAR 19.9 0.15 0.07 0.87 194 200 0.191 0.027 0.027MAIN 9.2 0.19 0.09 1.14 474 474 0.088 0.011 0.010MAAC 10.9 0.21 0.11 1.26 203 220 0.105 0.005 0.005MARCA 5.8 0.16 0.08 0.99 21 29 0.056 0.009 0.008NPCC 14.7 0.19 0.09 1.13 1296 1296 0.141 –0.004 –0.004SERC 21.2 0.18 0.09 1.06 554 560 0.203 0.021 0.021SWPP 9.0 0.19 0.10 1.16 284 846 0.087 0.008 –0.003ERCOT 6.5 0.10 0.05 0.59 0 716 0.063 0.010 –0.005WSCC 22.6 0.15 0.07 0.90 887 962 0.217 0.017 0.015——

Total 119.8 0.16 0.08 0.98 3913 5303 1.151 0.104 0.074a~~r~)~ A~~O~O~j~e ~ear~~ IgSO, For eacfl state sewed by more than one council, vehicles are distributed among the WJiOnS according to the percentage Of the

State’s residential consumers served by each council as estimated by the State’s public utility commission.bEVs require 0.4 kWh/mile and are driven 8,000 miles Per Year.cRec~arg/~g occurs over 12 hours during the night (1 Year = 8,7W hours).dRecharg~ng occurs evenly throughout the day.ef+echarging occurs over 2 hours at middayfAssumes that th e fuel u5ed t. generate electricity for Evs is in the same percentage as used for baseload generation (See table 5B-1).gsmall automobile Gets ~ mpg and drives 10,000 miles per yearhEV replaces ~ percent of the miles driven by 20 percent of the cars, Negative sign indicates fuel use increaSes rather than decreases


Chapter 6

Synthetic Fuels

Page

Introduction . . . . . . . . . . . . . . . . . . . . . . . . 157Petroleum Refining . . . . . . . . . . . . . . . . . 157

Process Descriptions . . . . . . . . . . . . . . . . . 160Synfuels From Coal . . . . . . . . . . . . . . . . 161Shale Oil . . . . . . . . . . . . . . . . . . . . . . . . . 165Synthetic Fuels From Biomass . . . . . . . . 167

Cost of Synthetic Fuels . . . . . . . . . . . . . . . 168Uncertainties. . . . . . . . . . . . . . . . . . . . . . 168Investment Cost . . . . . . . . . . . . . . . . . . . 170Consumer Cost. . . . . . . . . . . . . . . . . . . . 173

Developing a Synfuels Industry . . . . . . . . 175Constraints . . . . . . . . . . . . . . . . . . . . . . . 176Development Scenarios . . . . . . . . . . . . . 177

TABLES

Table No. Page

40.

41.42.

43.

44.

45.

Analysis of Potential for UpgradingDomestic Refining Capacity . . . . . . . . 159Principal Synfuels Products . . . . . . . . . 160Average Cost Overruns forVarious Types of Large ConstructionProjects . . . . . . . . . . . . . . . . . . . . . . . . . 169Selected Synfuels Processes andProducts and Their Efficiencies . . . . . . 171Best Available Capital and operatingCost Estimates for Synfuel PlantsProducing 50,000 bbl/d OilEquivalent of Fuel to End Users . . . . . 172Estimates of Cost Escalation in FirstGeneration Synfuels Plants . . . . . . . . . 172

Table No. Page

46.

47.48.

494

50.

Consumer Cost of Various SyntheticTransportation Fuels . . . . . . . . . . . . . . 173Assumptions . . . . . . . . . . . . . . . . . . . . . 174Effect of Varying Financial Parametersand Assumptions on Synfuels’ Costs . 174Fossil Synthetic Fuels DevelopmentScenarios . . . . . . . . . . . . . . . . . . . . . . . 177Biomass Synthetic Fuels DevelopmentScenarios . . . . . . . . . . . . . . . . . . . . . . . 180

FIGURES

Figure No. Page

13. Schematic Diagrams of Processes

14.

15

16

for Producing Various SynfuelsFrom Coal . . . . . . . . . . . . . . . . . . . . . . 162Cost Growth in Pioneer EnergyProcess Plants . . . . . . . . . . . . . . . . . . . 169Consumer Cost of Selected SynfuelsWith Various After tax Rates ofReturn on Investment . . . . . . . . . . . . . 174Comparison of Fossil SyntheticFuel Production Estimates . . . . . . . . . . 178

17. Synthetic Fuel Development

D.

Scenarios . . . . . . . . . . . . . . . . . . . . . . . 180

BOX

Page

Definitions of Demonstration andCommercial-Scale Plants . . . . . . . . . . . 179

Chapter 6

Synthetic Fuels

INTRODUCTION

Synthetic fuels, or “synfuels,” in the broadestsense can include any fuels made by breakingcomplex compounds into simpler forms or bybuilding simple compounds into others morecomplex. Both of these types of processes are car-ried out extensively in many existing oil refineries.Current technical usage, however, tends to re-strict the term to liquid and gaseous fuels pro-duced from coal, oil shale, or biomass. This usagewill be followed in this report.

Synfuels production is a logical extension ofcurrent trends in oil refining. As sources of themost easily refined crude oils are being depleted,refiners are turning to heavier oils and tar sands.Oil shale and coal, as starting materials for liquidhydrocarbon production, are extreme cases ofthis trend to heavier feedstocks.

Although synfuels production involves severalprocesses not used in crude oil refining, manycurrent oil refining techniques will be applied atvarious stages of synfuels processing. In order toindicate the range of currently used hydrocarbonprocessing techniques and to provide definitionsof certain terms used later in describing some syn-fuels processes, a brief description of common-ly used oil refining processes is given below. Fol-lowing this are descriptions of coal, oil shale, andbiomass synfuels processes; an evaluation of syn-fuel economics; and a presentation of two plaus-ible development scenarios for a U.S. synfuelsproduction capacity.

Petroleum Refining

A petroleum refinery is normally designed toprocess a specific crude oil (or a limited selec-tion of crudes) and to produce a “slate” of prod-ucts appropriate to the markets being supplied.Refineries vary greatly in size and complexity. Atone extreme are small “topping” plants withproduct outputs essentially limited to the com-ponents of the crude being processed. At theother extreme are very large, complex refinerieswith extensive conversion and treating facilities

and a corresponding ability to produce a rangeof products specifically tailored to changing mar-ket needs.

Refining processes include:

●

●

●

Atmospheric Distillation. –The “crude unit”is the start of the refining process. Oil underslight pressure is heated in a furnace andboiled into a column containing trays orpacking which serve to separate the variouscomponents of the crude oil according totheir boiling temperatures. Distillation (“frac-tionation”) is carried out continuously overthe height of the column. At several pointsalong the column hydrocarbon streams ofspecific boiling ranges are withdrawn for fur-ther processing.Vacuum Distillation. –Some crude oil com-ponents have boiling points that are too high,or they are too heat-sensitive, to permit dis-tillation at atmospheric pressure. In suchcases the so-called “topped crude” (bottomsfrom the atmospheric column) is further dis-tilled in a column operating under a vacuum.This lowers the boiling temperature of thematerial and thereby allows distillation with-out excessive decomposition.Desulfurization. –Sulfur occurs in crude oilin various amounts, and in forms rangingfrom the simple compound hydrogen sulfideand mercaptans to complex ring com-pounds. The sulfur content of crude oil frac-tions increases with boiling point. Thus,although sulfur compounds in fractions withlow boiling points can readily be removedor rendered unobjectionable, removal be-comes progressively more difficult and ex-pensive with fractions of higher boilingpoints. With these materials, sulfur is re-moved by processing with hydrogen in thepresence of special catalysts at elevated tem-peratures and pressures. The “hydrofining,”“hydrodesulfurization,” “residuum hydro-treating, ” and “hydrodemetallation” proc-esses are examples. Nitrogen compounds

157


and other undesirable components are alsoremoved in many of these hydrotreatingprocesses.Therma/ Cracking Processes, –Prior to thedevelopment of fluid catalytic cracking (seebelow), the products of distillation that wereheavier than gasol ine were commonly“cracked” under high temperature and pres-sure to break down these large, heavy mole-cules into smaller, more volatile ones andthereby improve gasoline yields. Althoughthe original process is no longer applied forthis purpose, two other thermal crackingprocesses are being increasingly used. In vis-breaking, highly viscous residues from crudeoils are mildly cracked to produce fuel oilsof lower viscosity. In delayed coking, crudeunit residues are heated to high temperaturesin large drums and severely cracked to driveoff the remaining high-boiling materials forrecovery and further processing; the porousmass of coke left in the drums is used as asolid fuel or to produce electric furnace elec-trodes.Fluid Catalytic Cracking.—This process in itsvarious forms is one of the most widely usedof all refinery conversion techniques. It isalso undergoing constant development.Charge stocks (which can be a range of dis-tillates and heavier petroleum fractions) areentrained in a hot, moving catalyst and con-verted to lighter products, including high-octane gasoline. The catalyst is separatedand regenerated, while the reaction productsare separated into their various componentsby distillation.Hydrocracking. —This process converts awide range of hydrocarbons to lighter, clean-er, and more valuable products. By catalytic-ally adding hydrogen under very high pres-sure, the process increases the ratio of hydro-gen to carbon in the feed and produces low-boiling material. Under some conditionshydrocracking maybe competitive with fluidcatalytic cracking.Catalytic Reforming. –Reforming is a cata-lytic process that takes low-octane “straight-run” materials and raises the octane numberto approximately 100. Although severalchemical reactions take place, the predomi-

nant reaction is the removal of hydrogenfrom naphthenes (hydrogen-saturated ring-Iike compounds) and their conversion to aro-matics (benzene-ring compounds). In addi-tion to markedly increasing octane number,the process produces hydrogen that can beused in desulfurization units.Isomerization, Catalytic Polymerization, andAlkylation. -These are specialized processesthat increase refinery yields of high-octanegasoline blending components from selectedstraight-chain liquids and certain refinerygases.

Historically, the U.S. refining industry has dealtprimarily with light, low-sulfur crudes. Usingprocesses described above, the industry achieveda balance between refinery output and markets.Adjustments have been made to meet the in-creasing demand for lead-free gasolines and tothe mandated reduction of lead in other gaso-lines. The heavy residual fuels, considerably high-er in sulfur content than treated distillate fuels,have continued to find a market as ships’ boilersand as fuels for utility plants that have not con-verted to coal. (In the latter market, it has some-times been necessary to blend in desulfurized fueloils to meet maximum fuel sulfur specifications.)In addition, large volumes of residual fuel oilshave continued to be imported, largely from Ven-ezuela and the Caribbean.

Now, however, the picture is changing. Dueto the limited availability of light crude oils, refin-eries are being forced to run increasing volumesof heavy crudes that are higher in sulfur and othercontaminants. With traditional processing meth-ods, these crudes produce fewer light productsand more heavy fuel oils of high sulfur content.On the other hand, fuel switching and conserva-tion in stationary uses will shift market demandincreasingly toward transportation fuels—gaso-Iine, diesel, and jet–plus petrochemical feed-stock.

Refiners are responding to this situation bymaking major additions to processing facilities.Although they differ in detail, the additions areintended to reduce greatly the production ofheavy fuel oil and to maximize the conversionand recovery of light liquids. For a typical major

Ch. 6—Synthetic Fuels . 159

refinery, the additions could include: 1 ) vacuumdistillation facilities, 2) high-severity hydroproc-essing, such as residuum desulfurization, togetherwith hydrogen manufacturing capacity, 3) de-layed coking, along with processes to recover andtreat the high-boiling vapor fractions driven off,and 4) perhaps visbreaking, catalytic cracker ex-pansions, and other modifications to accommo-date the changed product slate. It should also benoted that none of these additions increases thecrude-processing capacity of a refinery; theymerely adapt it to changed supply and marketingconditions.

Purvin and Gurtz1 have estimated the costs ofupgrading domestic refining capacity to makesuch changes. Their results are shown in table40. Although, as indicated in note d of the table,the investments shown do not include all applic-able costs, upgrading existing refineries is, in mostcases, less expensive than building synfuels plantsto produce the same products; and there are reg-ular reports that investments are being made inoil refineries to upgrade residual oil and changethe product slate. *2

I Purvin & Gertz, Inc., “An Analysis of Potential for UpgradingDomestic Refining Capacity, ” prepared for American Gas Associa-tion, Arlington, Va., undated.

*Another issue related to refining and oil consumption is the lowyield of lubrication and specialty oils from certain types of crudeoils (paraffinic crudes) and the redefining or reuse of these oils. There

For a discussion of other issues related to oilrefineries, the reader is referred to a Congres-sional Research Service report on “U.S. Refin-eries: A Background Study. ”3

appears to be no technical problem with increasing the yield oflubrication and specialty oils from the paraffinic crudes (0;/ andGas journal, “Gulf’s Port Arthur Refinery Due More Upgrading,”Sept. 8, 1980, p. 36.) or the redefining of lubrication oils. However,heat transfer, hydraulic, capacitor, and transformer fluids oftenbecome contaminated with PCBS (polychlorinated biphenyls)leached from certain plastics such as electrical insulating materials.Because of the health hazard, EPA regulations limit the allowablelevel of PCBS in enclosed systems to 50 ppm (parts per million).The contaminated oils pose a waste disposal problem and coulddamage refinery equipment (through the formation of corrosivehydrogen chloride and possible catalyst poisoning) if rerefinedwithout treatment. Recently, however, two processes (Chernica/and Engineering News, “Goodyear Develops PCB RemovalMethod, ” Sept. 1, 1980, p. 9; Chemical and Engineering News,“More PCB Destruction Methods Developed, ” Sept. 22, 1980, p.6.) have been announced that enable the removal of most of thePCBS, thereby enabling reuse directly or redefining if necessary; andone of these processes has been demonstrated with a prototypecommercial unit. Consequently, there do not appear to be signifi-cant technical problems with decontamination and reuse of PC B-contaminated oils. Due to the limits of this study, however, OTAwas unable to perform an economic analysis of oil production fromparaffinic crudes, redefining of lubrication oils, or decontaminationof specialty oils.

ZFor example, 0;/ and Gas ]ourna( Aug. 25, 1980, p. 69; Oil andGas )ournal, Sept. 8, 1980, p. 36; Oil and Gas Journa/, Nov. 10,1980, p. 150; Oi/ and Gas journal Jan. 19, 1981, p. 85.

3Congressional Research Service, “U.S. Refineries: A BackgroundStudy, ’’prepared at the request of the Subcommittee on Energy andPower of the House Committee on Interstate and Foreign Com-merce, U.S. House of Representatives, July 1980.

Table 40.—Analysis of Potential for Upgrading Domestic Refining Capacity

Topping refineries Total U.S. refineriesCase 1a a Case 1b b Case 2 C

Total investment d . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . $2.3 billionReduction in total U.S. residual fuel production, bbl/d . 217,000-301,000

Percent total pool . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13-18Increase in motor gasoline production, bbl/d . . . . . . . . . 134,000-200,000

Percent total pool . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-3Increase in diesel/No. 2 fuel production, bbl/d . . . . . . . . 105,000-135,000

Percent total pool . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.5-4.5Increase in low-Btu gas, MM Btu/D . . . . . . . . . . . . . . . . . . —Implementation period, years . . . . . . . . . . . . . . . . . . . . . . .Investment per unit capacity . . . . . . . . . . . . . . . . . . . . . . . $6,800-9,600

per bbl/d

$4.7 billion418,000-501,000

25-30167,000-234,000

2.5-3.5150,000-180,000

5-6233

$11,300-14,000per bbl/d

$18.0 billion1,587,000-1,670,000

95-1oo467,000-534,000

7-8540,000-600,000

18-201,3204-1o

$15,800-17,900per bbl/d

avacuum distillation, catalytic cradking, vlsbreakirw.bvacuum distillation, catalytic cracking, coking PIUS gasification.Ccase I b PIUS coking and gasification and downstream upgrading at remainin9 us. refineries.dFirst.quafier 19&3 investment, No provision for escalation, contingency or interest during construction.

SOURCE: Purvin & Gertz, Inc., “An Analysis of Potential for Upgrading Domestic Refining Capacity,” prepared for American Gas Association, 19S0.


PROCESS DESCRIPTIONS

A variety of synthetic fuels processes are cur-rently being planned or are under development.Those considered here involve the chemical syn-thesis of liquid or gaseous fuels from solid materi-als. As mentioned above, the impetus for synthe-sizing fluid fuels is to provide fuels that can eas-ily be transported, stored, and handled so as tofacilitate their substitution for imported oil and,to a lesser extent, imported natural gas.

The major products of various synfuels proc-esses are summarized in table 41. Depending onthe processes chosen, the products of synfuelsfrom coal include methanol (a high-octane gaso-line substitute) and most of the fuels derived fromoil* and natural gas. The principal products fromupgrading and refining shale oil are similar tothose obtained from conventional crude-oil refin-ing. The principal biomass synfuels are eithermethanol or a low- to medium-energy fuel gas.Smaller amounts of ethanol (an octane-boostingadditive to gasoline or a high-octane substitute

*As with natural crude oil, however, refining to produce largegasoline fractions usually requires more refining energy and expensethan producing less refined products such as fuel oil.

for gasoline) and biogas can also be produced.Each of these fuels can be synthesized further intoany of the other products, but these are the mosteasily produced from each source and thus prob-ably the most economic.

In the following section, the technologies forproducing synfuels from coal, oil shale, and bio-mass are briefly described. Indirect and directcoal liquefaction and coal gasification are pre-sented first. Shale oil processes are described sec-ond, followed by various biomass synfuels. Hy-drogen and acetylene production are not in-cluded because a preliminary analysis indicatedthey are likely to be more expensive and less con-venient transportation fuels than are the syntheticIiquids.4

4For a more detailed description o f v a r i o u s pf’OCeSSeS, see:

Engineering Societies Commission on Energy, Inc., “Coal Conver-sion Comparison, ” July 1979, Washington, D. C.; An Assessmentof Oil Shale Technologies (Washington, D, C.: U.S. Congress, Of-fice of Technology Assessment, June 1980), OTA-M-1 18; EnergyFrom Bio/ogica/ Processes (Washington, D, C.: U.S. Congress, Of-fice of Technology Assessment, July 1980), vol. 1, OTA-E-124; andEnergy From Biological Processes, Volume I/–Technical and En-vironmenta/ Ana/yses (Washington, D. C.: U.S. Congress, Office of

,Technology Assessment, September 1980), OTA-E-1 28.

Table 41 .—Principal Synfuels Products

Process Fuel production

Oil shale Gasoline, diesel and jet fuel,fuel oil, liquefiedpetroleum gases (LPG)

Fischer-Tropsch Gasoline, synthetic naturalgas (SNG), diesel fuel, andLPG

Coal to methanol, Mobil Gasoline and LPGmethanol to gasoline(MTG)

Coal to methanol Methanol

Wood or plant herbage tomethanol

Direct coal liquefaction

Grain or sugar to ethanol

SNGCoal to medium- or

low-energy gasWood or plant herbate

gasificationAnaerobic digestion

Methanol

Gasoline blending stock, fueloil or jet fuel, and LPG

Ethanol

SNGMedium- or low-energy fuel

gasMedium- or low-energy fuel

gasBiogas (carbon dioxide and

methane] and SNG

Comments

Shale oil is the synfuel most nearly like natural crude.

Process details can be modified to produce principallygasoline, but at lower efficiency.

LPG can be further processed to gasoline. Some processeswould also produced considerable SNG.

Depending on gasifier, SNG may be a byproduct. Methanolmost useful as high-octane gasoline substitute or gasturbine fuel, but can also be used as gasoline octanebooster (with cosolvents), boiler fuel, process heat fuel,and diesel fuel supplement. Methanol can also beconverted to gasoline via the Mobil MTG process.

Product same as above.

Depending on extent of refining, product can be 90 volumepercent gasoline.

Product most useful as octane-boosting additive togasoline, but can serve same uses as methanol.

Product is essentially indistinguishable from natural gas.Most common product likely to be close to synthesis gas.

Fuel gas likely to be synthesized at place where it is used.

Most products likely to be used onsite where produced.


Ch. 6—Synthetic Fuels ● 161

Synfuels From Coal

Liquid and gaseous fuels can be synthesizedby chemically combining coal with varyingamounts of hydrogen and oxygen, * as describedbelow. The coal liquefaction processes are gen-erally categorized according to whether liquidsare produced from the products of coal gasifica-tion (indirect processes) or by reacting hydrogenwith solid coal (direct processes). The fuel gases

*Some liquid and gaseous fuel can be obtained simply by heatingcoal, due to coal’s small natural hydrogen content, but the yieldis low.

from coal considered here are medium-Btu gasand a synthetic natural gas (SNG or high-Btu gas).Each of these three categories is consideredbelow and shown schematically in figure 13.

Indirect Liquefaction

The first step in the indirect liquefaction proc-esses is to produce a synthesis gas consisting ofcarbon monoxide and hydrogen and smallerquantities of various other compounds by react-ing coal with oxygen and steam in a reaction ves-sel called a gasifier. The liquid fuels are produced

Photo credit: Fluor Corp

Synthesis gas is converted to liquid hydrocarbons inFischer-Tropsch type reactors


Figure 13.—Schematic Diagrams of Processes for Producing Various Synfuels From Coal


Ch. 6—Synthetic Fuels Ž 163

by cleaning the gas, adjusting the ratio of carbonmonoxide to hydrogen in the gas, and pressuriz-ing it in the presence of a catalyst. Dependingon the catalyst, the principal product can begasoline (as in the Fischer-Tropsch process) ormethanol. The methanol can be used as a fuelor further reacted in the Mobil methanol-to-gasoline (MTG) process (with a zeolite catalyst)to produce Mobil MTG gasoline. The composi-tion of the gasoline and the quantities of otherproducts produced in the Fischer-Tropsch proc-ess can also be adjusted by varying the temper-ature and pressure to which the synthesis gas issubjected when liquefied.

With commercially available gasifiers, part ofthe synthesis gas is methane, which can be puri-fied and sold as a byproduct of the methanol orgasoline synthesis. * However, the presence ofmethane in the synthesis gas increases the energyneeded to produce the liquid fuels, because itmust be pressurized together with the synthesisgas but does not react to form liquid products.Alternatively, rather than recycling purge gas(containing increasing concentrations of meth-ane) to the methanol synthesis unit, it can be sentto a methane synthesis unit and its carbon mon-oxide and hydrogen content converted to SNG.With “second generation” gasifiers (see below),little methane would be produced and the metha-nol or gasoline synthesis would result in relativelyfew byproducts.

There are three large-scale gasifiers with com-mercially proven operation: Lurgi, Koppers-Totzek, and Winkler. Contrary to some reportsin the literature, all of these gasifiers can utilizea wide range of both Eastern and Westerncoals, * * although Lurgi has not been commercial-

*For example, the synthesis gas might typically contain 13 per-cent methane. Following methanol synthesis, the exiting gases mightcontain 60 percent methane, which is sufficiently concentrated foreconomic recovery.

* ● For example, Sharman (R. B. Sharman, “The British Gas/LurgiSlagging Gasifier–What It Can Do,” presented at Coal Technology’80, Houston, Tex., Nov. 18-20, 1980) states: “h has been claimedthat the fixed bed gasifiers do not work well with swelling coals.Statements such as this can still be seen in the literature and arenot true. In postwar years Lurgi has given much attention to theproblem of stirrer design which has much benefited the WestfieldSlagging Gasifier, Substantial quantities of strongly caking and swell-ing coals such as Pittsburgh 8 and Ohio 9, as well as the equivalentstrongly caking British coals have been gasified. No appreciableperformance difference has been noted between weakly cakingand strongly caking high volatile bituminous coals. ”

Iy proven with Eastern coals. In all cases, thephysical properties of the feed coal will influencethe exact design and operating conditionschosen * for a gasifier. For example, the coalswelling index, ease of pulverization (friability),and water content are particularly important pa-rameters to the operation of Lurgi gasifiers, andthe Koppers-Totzek gasifier requires that the ashin the coal melt for proper operation, as do theShell and Texaco “second generation” designs.,

it is expected that the developing pressurized,entrained-flow Texaco and Shell gasifiers will besuperior to existing commercial gasifiers in theirability to handle strongly caking Eastern coalswith a rapid throughput. This is achieved by rapidreaction at high temperatures (above the ashmelting point). These temperatures, however, areachieved at the cost of reduced thermal efficiencyand increased carbon dioxide production.

The Fischer-Tropsch process is commercial inSouth Africa, using a Lurgi gasifier, but the UnitedStates lacks the operating experience of SouthAfrica and it is unclear whether this will poseproblems for commercial operation of this proc-ess in the United States. The methanol synthesisfrom synthesis gas is commercial in the UnitedStates, but a risk is involved with putting togethera modern coal gasifier with the methanol syn-thesis, since these units have not previously beenoperated together. Somewhat more risk is in-volved with the Mobil MTG process, since it hasonly been demonstrated at a pilot plant level.Nevertheless, since the Mobil MTG process in-volves only fluid streams* * the process can prob-ably be brought to commercial-scale operationwith little technical difficulty.

Direct Liquefaction

The direct liquefaction processes produce a liq-uid hydrocarbon by reacting hydrogen directlywith coal, rather than from a coal-derived synthe-sis gas. However, the hydrogen probably will beproduced by reacting part of the coal with steamto produce a hydrogen-rich synthesis gas, sothese processes do not eliminate the need for coal

*Including steam and oxygen requirements.**The physical behavior of fluids is fairly well understood, and

processes involving only fluid streams can be scaled up much morerapidly with minimum risk than processes involving solids.


gasification. The major differences between theprocesses are the methods used to transfer thehydrogen to the coal, while maximizing catalystlife and avoiding the flow problems associatedwith bringing solid coal into contact with a solidcatalyst, but the hydrocarbon products are like-ly to be quite similar. The three major direct liq-uefaction processes are described briefly below,followed by a discussion of the liquid product andthe state of the technologies’ development.

The solvent-refined coal (SRC I) process wasoriginally developed to convert high-sulfur, high-ash coals into low-sulfur and low-ash solid fuels.Modifications in the process resulted in SRC II,which produces primarily a liquid product. Thecoal is slurried with part of the liquid hydrocar-bon product and reacted with hydrogen at about850° F and a pressure of 2,000 per square inch(psi). 5

AS it now stands, however, feed coal forthis process is limited to coals containing pyriticminerals which act as catalysts for the chemicalreactions.

The H-coal process involves slurrying the feedcoal with part of the product hydrocarbon andreacting it with hydrogen at about 650° to 700°F and about 3,000 psi pressure in the presenceof a cobalt molybdenum catalyst.6 A novel aspectof this process is the so-called “ebullated” bedreactor, in which the slurry’s upward flowthrough the reactor maintains the catalyst parti-cles in a fluidized state. This enables contact be-tween the coal, hydrogen, and catalyst with a rel-atively small risk of clogging.

The third major direct liquefaction method isthe EXXON Donor Solvent (EDS) process. In thisprocess, hydrogen is chemically added to a sol-vent in the presence of a catalyst. The solvent isthen circulated to the coal at about 800° F and1,500 to 2,000 psi pressure.7 The solvent then,in chemical jargon, chemically donates thehydrogen atoms to the coal; and the solvent isrecycled for further addition of hydrogen. Thisprocess circumvents the problems of rapid cat-alyst deactivation and excessive hydrogen con-sumption.

5Erlgineering societies Commission on Energy, Inc.) oP. cit.

Wameron Engineering, Inc., “Synthetic Fuels Data Handbook,”2d cd., compiled by G. L. Baughman, 1978.

‘Ibid.

In all three processes, the product is removedby distilling it from the slurry, so there is no resid-ual oil* fraction in these “syncrudes. ” Becauseof the chemical structure of coal, the product ishigh in aromatic content. The initial product isunstable and requires further treatment to pro-duce a stable fuel. Refining** the “syncrude”consists of further hydrogenation or coking (toincrease hydrogen content and remove impuri-ties), cracking, and reforming; and current indica-tions are that the most economically attractiveproduct slate consists of gasoline blending stockand fuel oil,8 but it is possible, with somewhathigher processing costs, to produce products thatvary from 27 percent gasoline and 61 percent jetfuel up to 91 percent gasoline and no jet fuel.9

The gasoline blending stock is high in aromat-ics, which makes it suitable for blending withlower octane gasoline to produce a high-octanegasoline. Indications are that the jet fuel can bemade to meet all of the refinery specifications forpetroleum-derived jet fuel.10 However, since themethods used to characterize crude oils and theproducts of oil refining do not uniquely determinetheir chemical composition, the refined productsfrom syncrudes will have to be tested in variousend uses to determine their compatibility withexisting uses. Because of the chemistry involved,these syncrudes appear economically less suita-ble for the production of diesel fuel.***

*ResiduaI oil is the fraction that does not vaporize under distilla-tion conditions. Since this syncrude is itself the byproduct of distilla-tion, all of the fractions vaporize under distillation conditions.

● *At present, it is not clear whether existing refineries will bemodified to accept coal syncrudes or refineries dedicated to thisfeedstock will be built. Local economics may dictate a combina-tion of these two strategies. Refining difficulty is sometimes com-pared to that of refining sour Middle East crude when no high-sulfurresidual fuel oil product is produced (i.e., refining completely tomiddle distillates and gasoline) (see footnote 8). This is moderate-ly difficult but well within current technical capabilities. One of theprincipal differences between refining syncrudes and natural crudeoil, however, is the need to deal with different types of metallicimpurities in the feedstock.

BUC)p, Inc., and System Development Corp., “Crude oil VS. Coal

Oil Processing Compassion Study,” DOE/ET/031 17, TR-80/009-001,November 1979.

gChevron, “Refining and Upgrading of Synfuels From Coal andOil Shales by Ad~anced Catalytic Processes, Third Interim Report:Processing of SRC II Syncrude,” FE-21 35-47, under DOE contractNo. EF-76-C-01-2315, Apr. 30, 1981.

‘oIbid.* * *Th e product is hydrogenated and cracked to form saturated,

single-ring compounds, and to saturated diolefins (which wouldtend to form char at high temperatures). The reforming step pro-


None of the direct liquefaction processes hasbeen tested in a commercial-scale plant. All in-volve the handling of coal slurries, which arehighly abrasive and have flow properties that can-not be predicted adequately with existing theoriesand experience. Consequently, engineers cannotpredict accurately the design requirements of acommercial-scale plant and the scale-up must gothrough several steps with probable process de-sign changes at each step. As a result, the directliquefaction processes are not likely to make asignificant contribution to synfuels production be-fore the 1990’s. At this stage there would be asubstantial risk in attempting to commercializethe direct liquefaction processes without addi-tional testing and demonstration.

GasificationThe same type of gasifiers used for the liquefac-

tion processes can be used for the production ofsynthetic fuel gases. The first step is the produc-tion of a synthesis gas (300 to 350 Btu/SCF). *The synthesis gas can be used as a boiler fuel orfor process heat with minor modifications in end-use equipment, and it also can be used as achemical feedstock. Because of its relatively lowenergy density and consequent high transportcosts, synthesis gas probably will not be trans-ported (in pipelines) more than 100 to 200 miles.There is very little technical risk in this process,however, since commercial gasifiers could beused.

The synthesis gas can also be used to synthesizemethane (the principal component of natural gashaving a heat content of about 1,000 Btu/SCF).This substitute or synthetic natural gas (SNG) canbe fed directly into existing natural gas pipelinesand is essentially identical to natural gas. Thereis some technical risk with this process, since themethane synthesis has not been demonstrated ata commercial scale. However, since it only in-volves fluid streams, it probably can be scaled

duces aromatics from the saturated rings. The rings can also bebroken to form paraffins, but the resultant molecules and other par-affins in the “oil” are too small to have a high cetane rating (thecetane rating of one such diesel was 39 (see footnote 9), whilepetroleum diesels generally have a centane of 45 or more (E. M.Shelton, “Diesel Fuel Oils, 1980, ” DOE/BETC/PPS-80/5, 1980)).Polymerization of the short chains into longer ones to produce ahigh-cetane diesel fuel is probably too expensive.

**Assuming the gas consists primarily of carbon monoxide andhydrogen.

up to commercial-scale operations without seri-ous technical difficulties.

As mentioned under “Indirect Liquefaction, ”there are several commercial gasifiers capable ofproducing the synthesis gas. The principal tech-nical problems in commercial SNG projects arelikely to center around integration of the gasifierand methane synthesis process.

Shale Oil

Oil shale consists of a porous sandstone thatis embedded with a heavy hydrocarbon (knownas kerogen). Because the kerogen already con-tains hydrogen, a liquid shale oil can be producedfrom the oil shale simply by heating the shale tobreak (crack) the kerogen down into smaller mol-ecules. This can be accomplished either with asurface reactor, a modified in situ process, or aso-called true in situ process.

In the surface retorting method, oil shale ismined and placed in a metal reactor where it isheated to produce the oil. In the “modified insitu” process, an underground cavern is exca-vated and an explosive charge detonated to fillthe cavern with broken shale “rubble.” Part ofthe shale is ignited to produce the heat neededto crack the kerogen. Liquid shale oil flows to thebottom of the cavern and is pumped to the sur-face. In the “true in situ” process, holes are boredinto the shale and explosive charges ignited ina particular sequence to break up the shale. The“rubble” is then ignited underground, produc-ing the heat needed to convert the kerogens toshale oil.

The surface retort method is best suited to thickshale seams near the surface. The modified in situis used where there are thick shale seams deepunderground. And the true in situ method isbest suited to thin shale seams near the surface.The surface retorting method requires the min-ing and disposal of larger volumes of shale thanthe modified in situ method and the true in situmethod requires only negligible mining. It is moredifficult, using the latter two processes, however,to achieve high oil yields of a relatively uniformquality, primarily because of difficulties relatedto controlling the underground combustion and


Photo credit: Department of Energy

Synthane pilot plant near Pittsburgh, Pa., converts coal to synthetic natural gas


ensuring that the resultant heat is efficiently trans-ferred to the shale. It is likely, however, that theseproblems can be overcome with further develop-ment work.

The shale oil must be hydrogenated under con-ditions similar to coal hydrogenation (800° F,2,000 psi)11 to remove its tightly bound nitrogen,which, if present, would poison refinery catalysts.

The resultant upgraded shale oil is often com-pared to Wyoming sweet crude oil in terms ofits refining characteristics and is more easily re-fined than many types of higher sulfur crude oilscurrently being refined in the United States. Refin-ing shale oil naturally produces a high fractionof diesel fuel, jet fuel, and other middle distillates.The products, however, are not identical to thefuels from conventional crude oil, so they mustbe tested for the various end uses.

Shale oil production is currently moving tocommercial-scale operation, and commercial fa-cilities are likely to be in operation by the midto late 1980’s. Because of completed and ongo-ing development work, the risks associated withmoving to commercial-scale operation at this timeare probably manageable, although risks are nev-er negligible when commercializing processes forhandling solid feedstocks.

Synthetic Fuels From Biomass

The major sources of biomass energy are woodand plant herbage, from which both liquid andgaseous fuels can be synthesized. These synthesesand the production of some other synfuels fromless abundant biomass sources are describedbriefly below.

Liquid Fuels

The two liquid fuels from biomass consideredhere are methanol (“wood alcohol”) and ethanol(“grain alcohol”). Other liquid fuels from biomasssuch as oil-bearing crops must be considered asspeculative at this time.12

‘ ‘Chevron, “Refining and Upgrading of Synfuels From Coal andOil Shales by Advanced Catalytic Processes, First Interim Report:Paraho Shale Oil, ’’Report HCP/T23 15-25 UC90D, Department ofEnergy, July 1978,

12Errergy From Biological processes, of). cit.

Methanol can be synthesized from wood andplant herbage in essentially the same way as itis produced from coal. One partially oxidizes orsimply heats (pyrolyzes) the biomass to producea synthesis gas. The gas is cleaned, the ratio ofcarbon monoxide to hydrogen adjusted, and theresultant gas pressurized in the presence of a cat-alyst to form methanol. As with the indirect coalprocesses, the synthesis gas could also be con-verted to a Fischer-Tropsch gasoline or the meth-anol converted to Mobil MTG gasoline.

Methanol probably can be produced fromwood with existing technology, but methanol-from-grass processes need to be demonstrated.Several biomass gasifiers are currently under de-velopment to improve efficiency and reliabilityand reduce tar and oil formation. Particularly no-table are pyrolysis gasifiers which could signifi-cantly increase the yield of methanol per ton ofbiomass feedstock. Also mass production of small(5 million to 10 million gal/y r), prefabricatedmethanol plants may reduce costs significantly.With adequate development support, advancedgasifiers and possibly prefabricated methanolplants could be commercially available by themid to late 1980’s.

Ethanol production from grains and sugar cropsis commercial technology in the United States.The starch fractions of the grains are reduced tosugar or the sugar in sugar crops is used direct-ly. The sugar is then fermented to ethanol andthe ethanol removed from the fermentation brothby distillation.

The sugar used for ethanol fermentation canalso be derived from the cellulosic fractions ofwood and plant herbage. Commercial processesfor doing this use acid hydrolysis technology, butare considerably more expensive than grain-based processes. Several processes using enzy-matic hydrolysis and advanced pretreatments ofwood and plant herbage are currently under de-velopment and could produce processes whichsynthesize ethanol at costs comparable to thoseof ethanol derived from grain, but there are stillsignificant economic uncertainties.13

131bid

168 Ž Increased Automobile Fuel Efficiency and synthetic Fuels: Alternatives for Reducing Oil Imports

Fuel Gases

By 2000, the principal fuel gases from biomassare likely to be a low-energy gas from airblowngasifiers and biogas from manure. Other sourcesmay include methane (SNG) from the anaerobicdigestion of municipal solid waste and possiblykelp.

A relatively low-energy fuel gas (about 200 Btu/SCF) can be produced by partially burning woodor plant herbage with air in an airblown gasifier.The resultant gas can be used to fuel retrofittedoil- or gas-fired boilers or for process heat needs.Because its low energy content economically pro-hibits long-distance transportation of the gas,most users will operate the gasifier at the placewhere the fuel gas is used. Several airblown bio-mass gasifiers are under development, and com-mercial units could be available within 5 years.

Biogas (a mixture of carbon dioxide and meth-ane) is produced when animal manure or sometypes of plant matter are exposed to the appro-priate bacteria in an anaerobic digester (a tanksealed from the air). Some of this gas (e.g., fromthe manure produced at large feedlots) may bepurified, by removing the carbon dioxide, andintroduced into natural gas pipelines, but mostof it is likely to be used to generate electricity andprovide heat at farms where manure is produced.The total quantity of electricity produced this waywould be small and, to an increasing extent,

would be used to displace nuclear- and coal-gen-erated electricity. A part of the waste heat fromthe electric generation can be used for hot waterand space heating in buildings on the farm, how-ever.14 A small part of the biogas (perhaps 15 per-cent corresponding to the amount occurring onlarge feedlots) could be purified to SNG and in-troduced into natural gas pipelines.

Biogas can also be produced by anaerobic di-gestion of municipal solid waste in landfills andkelps. Any gas so produced is likely to be purifiedand introduced into natural gas pipelines.

Manure digesters for cattle manure are com-mercially available. Digesters utilizing othermanures require additional development, butcould be commercially available within 5 years.The technology for anaerobic digestion of munici-pal solid waste was not analyzed, but one systemis being demonstrated in Florida.15 In addition,if ocean kelp farms prove to be technically andeconomically feasible, there may be a small con-tribution by 2000 from the anaerobic digestionof kelp to produce methane (SNG),16 but thissource should be considered speculative at pres-ent.

14TRW, “Achievin g a production Goal of 1 Million B/D of coal

Liquids by 1990,” March 1980.’51. E., Associates, “Biological Production of Gas,” contractor re-

port to OTA, April 1979, available in Energy From Biologica/Proc-esses, Vol ///: Appendixes, Part C, September 1980,

16Energy From Biological prOCeSSeS, OP. cit.

COST OF SYNTHETIC FUELS

There is a great deal of uncertainty in estimating charges can vary by more than a factor of 2. Incosts for synthetic fuels plants. A number of fac- many cases, differences in product cost estimatestors, which can be predicted with varying degrees can be explained solely on the basis of these dif-of accuracy, contribute to this uncertainty. Some ferences. For most of the biomass fuels, the costof the more important are considered below, fol- of the biomass feedstock is also highly variable,lowed by estimates for the costs of various syn- and this has a strong influence on the productfuels. cost.

Uncertainties The cost of synfuels projects, and particularlythe very large fossil fuel ones, is also affected by:

For most of the synfuels, fixed charges are a 1) construction delays, 2) real construction costlarge part of the product costs. Depending on increases (corrected for general inflation) duringassumptions about financing, interest rates, and construction, and 3) delays in reaching full pro-the required rate of return on investment, these duction capacity after construction is completed,


due to technical difficulties. These factors are usu-ally not included in cost estimates, but they arelikely to affect the product cost.

Another factor that should be considered is thestate of development of the technology on whichthe investment and operating cost estimates arebased, As technology development proceeds,problems are discovered and solved at a cost, andthe engineer’s original concept of the plant isgradually replaced with a closer and closer ap-proximation of how the plant actually will look.Consequently, calculations based on less devel-oped technologies are less accurate. This usual-Iy means that early estimates understate the truecosts by larger margins than those based on moredeveloped and well-defined technologies. Thisis particularly true of processes using solid feed-stocks because of the inherent difficulty with scal-ing-up process streams involving solids. Figure 14illustrates cost escalations that can occur, by sum-

marizing the increases in cost estimates for vari-ous energy projects as technology developmentproceeded. Table 42 also illustrates this point byshowing average cost overruns that have oc-curred in various types of large constructionprojects.

It should be noted, however, that the periodof time in which most of the project evaluations

Table 42.-Average Cost Overruns for VariousTypes of Large Construction Projects

Actual cost dividedSystem type by estimated cost

Weapons systems. . . . . . . . . . . . . . . 1.40-1.89Public works . . . . . . . . . . . . . . . . . . . 1.26-2.14Major construction . . . . . . . . . . . . . . 2.18Energy process plants . . . . . . . . . . . 2.53SOURCES: Rand Corp., “A Review of Cost Estimates In New Technologies: lmpli-

cationa for Energy Process Piants,” prepared for U.S. Departmentof Energy, July 1979; Hufschmidt and Gerin, “Systematic Errors InCost Estimates for Public Investment Projects,” in The Ana/ys/s ofPub//c Output, Columbia University Press, 1970; and R. Perry, et al.,“Systems Acqulsitlon Strategies,” Rand Corp., 1970.

Figure 14.—Cost Growth in Pioneer Energy Process Plants (constant dollars)

Initial Preliminary Budget Definitive Actual Add-on

Type of estimate

SOURCE: “A Review of Cost Estimation in New Technologies: Implications for Energy Process Plants,” Rand Corp. R-2481- DOE, July 1979,


in table 42 were made had high escalation ratesfor capital investment relative to general econom-ic inflation. Historically this has not been the case;and if future inflation in plant construction morenearly follows general inflation, the expected syn-fuels investment increases from preliminary de-sign to actual construction would be lower thanis indicated in figure 14 and table 42.

When judging the future costs and economiccompetitiveness of synfuels, one must thereforealso consider the long-term inflation in construc-tion costs. To a very large extent, the ability toproduce synfuels at costs below those of petrol-eum products will depend on the relative infla-tion rates of crude oil prices and constructioncosts.

Although economies of scale are important forsynfuels plants, there are also certain disecon-omies of scale—factors which tend to increaseconstruction costs (per unit of plant capacity) forvery large facilities as compared with small ones.First, the logistics of coordinating constructionworkers and the timely delivery of constructionmaterials become increasingly difficult as the con-struction project increases in size and complex-ity. Second, construction labor costs are higherin large projects due to overtime, travel, and sub-sistence payments. Third, as synfuels plants in-crease in size, more and more of the equipmentmust be field-erected rather than prefabricatedin a factory. This can increase the cost of equip-ment, although in some cases components maybe “mass-produced” on site, thereby equalingthe cost savings due to prefabrications. Some ofthese problems causing diseconomies of scalecan be aggravated if a large number of synfuelsprojects are undertaken simultaneously.

Many of the above factors would tend to in-crease costs, but once several full-scale plantshave been built, the experience gained may helpreduce production costs for future generationsof plants. Delays in reaching full production ca-pacity can be minimized, and process innova-tions that reduce costs can be introduced. in ad-dition, very large plants that fully utilize the avail-able economies of scale can be built with confi-dence. Consequently, the first generation unitsproduced are likely to be the most expensive, ifadjustments are made for inflation in construc-tion and operating costs.

An example of this can be found in the chem-ical industry, where capital productivity (outputper unit capital investment) for the entire industryhas increased by about 1.4 percent per year since1949.17 In some sectors, such as methanol syn-thesis, productivity has increased by more than4 percent per year for over 20 years.18 Much ofthis improvement is attributable to increasedplant size and the resultant economies of scale:Because the proposed synfuels plants are alreadyrelatively large, cost decreases for synfuels pIantsmay not be as large and consistent as those ex-perienced in the chemical industry in recentyears; however, because of the newness of theindustry, some decreases are expected.

Investment Cost

For purposes of cost calculations, previous OTAestimates 19 20 were used for oil shale and biomasssynfuels (adjusted, in the case of oil shale, to 1980dollars) and the best available cost estimates inthe public literature were used for coal-derivedsynfuels. These latter estimates were compared21

to the results of an earlier Engineering Societies’Commission on Energy (ESCOE) study22 of coal-derived synfuels, which used preliminary engi-neering data. Since the best available cost esti-mates correspond roughly to definitive engineer-ing estimates, the ESCOE numbers were in-creased by 50 percent, the amount by which en-gineering estimates typically increase when go-ing from preliminary to definitive estimates.When these adjustments were made and thecosts expressed in 1980 dollars, the two sourcesof cost estimates for coal-based synfuels producedroughly comparable results. *s

Table 43 shows the processes and productslates used for the cost calculations. As describedabove, a variety of alternative product slates arepossible, but these were chosen to emphasize theproduction of transportation fuels. Table 44 givesthe best available investment and operating costs

“E. J. Bentz & Associates, Inc., “Selected Technical and EconomicComparisons of Synfuel Options, ” contractor report to OTA, 1981.

lalbid.lgAn Assessment of Oil shale Technologies, op. cit.2oEnergy From Biological /%XeSSt%, OP. cit.ZI E. J. Bentz & Associates, Inc., op. cit.zzEnglneering societies Commission on Energy, InC., OP. cit.Z3E. J. Bentz & Associates, Inc., Op. cit.


Table 43.—Selected Synfuels Processes and Products and Their Efficiencies

Energy efficiency (percent)

Fuel products Tranportation fuel productsFuel products (percent of input coal (percent of input coal

Process (percent of output) and external power) and external power)

Oil shale Gasoline (19) b N.A. C N.A. C

Jet fuel (22)Diesel fuel (59)

Methanol/synthetic natural gas (SNG) from coal Methanol (48) d 65 33SNG (49)Other (3)

Methanol from coal Methanol (lOO) ef 55 g 55 g

Coal to methanol/SNG, Mobil methanol Gasoline (40) d 63 27to gasoline SNG (52)

Other (8)

Coal to methanol, Mobil methanol to gasoline Gasoline (87) d 47 41Other (13)

Fischer-Tropsch/SNG from coal Gasoline (33) d 56 17SNG (65)Other (2)

Direct coal liquefaction Gasoline (33) fh 57 47Jet fuel (49)Other (18)

SNG from coal SNG (lOO) f 59 0

Methanol from wood Methanol (lOO) i 47 f 47 h

Ethanol from grain Ethanol (lOO) i N.A. C N.A. C

“Higher heating value of products divided by higher heating value Of the coal Plus impor’ted ener9Y.

bFf. F. Sullivan, et al., “Refinhrg and Upgrading of Synfuels From Coal and Oil Shales by Actvanced Catalytic Processes, ” first interim report, prepared by Chevron forDepartment of Energy, April 1978, NTIS No FE-2315-25.

cNot applicabledMM Schreiner, ,, Research Guidance Studies to Assess Gasoline From Coal to Methanol-to-Gasoline and Sasol-Type Fischer. Tropsch Technologies,” prepared by Mobil

R&D Co. for the Department of Energy, August 1978, NTIS No. FE-2447-13.eDHR, InC,, ,,phase 1 Methanol use options study,” prepared for the Deparfrnent of Energy under contract No, DE-AC01-79PE-70027, Dec. 23, 1980,

‘K. A. Rogers, “Coal Conversion Comparisons,” Engineering Societies Commission on Energy, Washington, DC., prepared for the Department of Energy, July 1979,No, FE-2488-51.

gsullivant and Frumking (footnote h) give 57 percent, DHR (footnote e) gives 52 percent.hR, F, Sullivan and H. A. Frumkin, “Refining and Upgrading of Synfuels by Advanced Catalytic Processes,” third interim report, prepared by Chevron for the Department

of Energy, Apr. 30, 1980, NTIS, No. FE-2315,47 Products shown are for H-Coal. It is assumed that same product slate results from refined EDS liquids.iOTA, Energy from Sio/oQ/ca/ processes, Vo/ume //, September lg~, GpO stock No. 052-003-00782-7.

SOURCE Office of Technology Assessment

(excluding coal costs) in 1980 dollars for thevarious processes, with all results normalized tothe production of 50,000 barrels per day (bbl/d)oil equivalent of product to the end user (i.e., in-cluding refining losses). Only a generic direct liq-uefaction process is included because currentestimation errors appear likely to be greater thanany differences between the various direct lique-faction processes.

Based on the history of cost escalation in theconstruction of chemical plants, one can be near-ly certain that final costs of the first generationof these synfuels plants will exceed those shownin table 44 (with the exception of ethanol whichis already commercial), Using a methodology de-veloped to estimate this cost escalation, Rand

Corp. has examined several synfuels processesand derived cost growth factors, or estimates ofhow much the capital investment in the synfuelsplant is likely to exceed the best available engi-neering estimates. Some of the results of the Randstudy are shown in table 45. Also shown is theexpected performance of each plant if it werebuilt today, expressed as the percentage of de-signed fuel production that the plant is likely toachieve.

The figures reflect Rand’s judgment that directliquefaction processes require further develop-ment before construction of a commercial-scaleplant should be attempted; but the calculationsalso indicate that even the first generation of near-commercial processes are likely to be more ex-


Table 44.—Best Available Capital and Operating Cost Estimates for SynfuelsPlants Producing 50,000 bbl/d Oil Equivalent of Fuel to End Users

Annual operating costsCapital investment (exclusive of coal costs)

Process (billion 1980 dollars) (million 1980 dollars)

Oil shale a. ... , . . . . . . . . . . . . . . . . . . . . . . . . $2,2 $250Methanol/SNG from coal b. . . . . . . . . . . . . . . 2.1 150Methanol from coal c . . . . . . . . . . . . . . . . . . . 2.8 200Coal to methanol/SNG, Mobil methanol

to gasoline b. . . . . . . . . . . . . . . . . . . . . . . . . 2.4 170Coal to methanol, Mobil methanol

to gasoline b. . . . . . . . . . . . . . . . . . . . . . . . . 3.3 230Fischer-Tropsch/SNG from coal b. . . . . . . . . 2.5 190Direct coal Iiquefaction d . . . . . . . . . . . . . . . . 3.0 250SNG e . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2..2 150Methanol from wood f. . . . . . . . . . . . . . . . . . . 2.9 610

(wood at $30/dry ton)

(wood at $45/dry ton)860

Ethanol f . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.8 ($3/bu. corn)1,112

($4.50/bu. corn)aofflce of Technology Assessment, An Aggg~ment of 0// Shale Techno/og/es, June 1980, $1.7 billion Investment in 1979 dollars

becomes $1.9 bllllon In 1980 dollars for 50,000 bbl/d of ahale oH. Aasuming 88 percent refining efficiency, one needs 57,000bbl/d of shale 011 to produce 50,000 bbl/d 011 equivalent of products, at an investment of $1.9 billion/O.88 -$2.2 billion.

b~rived from R. M. Wham, et al., “Liquefaction Technology Assessment-Phase 1: Indirect Liquefaction of cod to Mettranoland Gasollne Using Available Technology,” Oak Ridge National Laboratory, ORNL-5884, February 1981.

cFrom DHR, Inc., “Phase I Methanol Use Optlona Study,” prepared for the Department of Energy under contract No.DE-AC01-79PE-70027, Dec. 23, 1980, one finds that the ratio of investment cost of a methanol to a Mobil methanol-to-gasollneplant is 0.85. Assuming this ratio and the value for a methanol-to-gasoline plant from footnote b, one arrives at the invest-ment cost shown. The operating cost was increased in proportion to investment coat. Thia adjustment is necessary to putthe costs on a common baais. These vaiuea are 50 percent more than the estlmatea given by DHR (reference above) andBadger (Badger Plants, Inc., “Conceptual Design of a coal to Methanol Commercial Plant,” prepared for the Department ofEnergy, February 1978, NTIS No. FE-2416-24). In order to compare Badger with this eatimate, It was necessary to scale downthe Badger plant (ualng a 0.7 scaling factor) and inflate the result to 1980 dollars (increase by 39 percent from 1977).

d~xon Research and Engineering ~., “EDS Coal Liquefaction Process Development, Phaae V,” prepared for the Departmentof Energy under cooperative agreement DE-FCO1-77ETIOO89, March 1981. Investment and operating cost assumes an energyefficiency of 82.5 percent for the refining procesa. Refinery Investment of up to S700 million Is not included in the capitalinvestment.

‘Rand Corp., “Cost and Performance Expectations for Pioneer Synthetic Fuels Plants,” report No. R-2571-DOE, 1981.f Office of Technology Assessment, Energy From Biological Processes, Vo lume //, SePternbSr 1980, Gpo stock No.052-003-00782-7; i.e., 40 million gatlons per year methanol plant costing S88 million, 50 million gallons per year ethanol plantcosting $75 million,


Table 45.—Estimates of Cost Escalation in First Generation Synfuels Plants

Cost growth factorderived by Rand a Expected performance for 90

for 90 percent Revised investment cost percent confidence interval b

Process confidence interval b (billion 1980 dollars) (percent of plant design)

Oil shale . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.04-1.39 2.3-3.1 57-85Coal to methanol to gasoline (Mobil, no

SNG byproduct). ., . . . . . . . . . . . . . . . . . . . 1.06-1.43 3.5-4.7 65-93Direct coal liquefaction (H-Coal) . . . . . . . . . 1.52-2.38a 4.0-6 .3a 25-53SNG . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0.95-1.23 2.1-2.7 69-97aEaaed on Rand @@ in which best estimate for H-Coal is $2.2 billion for 50,000 bbl/d of Product syncrude. With 82.5 Percent refining efficiency, this becomes $2.7billion for 50,000 bblld of product to end user.

bln other words, go percent probability that actual cost growth factor or Performance wiit fall in the inte~al.

SOURCE: Office of Technology Assessment; adapted from Rand Corp., “Cost and Performance Expectations for Pioneer Synthetic Fuels Plants,” R-2571-DOE, 1981.


pensive and less reliable than the best conven-tional engineering estimates would indicate. Thisanalysis indicates that it is quite reasonable to ex-pect first generation coal liquefaction plants ofthis size to cost $3 billion to $5 billion or moreeach in 1980 dollars.

Consumer Cost

The consumer costs of the various synfuels areshown in table 46. These consumer costs arebased on the estimates in table 44 and the eco-nomic assumptions listed in table 47. The effecton the calculated product cost of varying someof the economic assumptions is then shown intable 48.

With these economic assumptions, deliveredliquid fossil synfuels costs (1980 dollars) rangefrom $1.25 to $1.85 per gallon of gasoline equiva-

lent (gge) for 100 percent equity financing and$0.80 to $1.25/gge with 75 percent debt financ-ing at 5 percent real interest (i. e., relative to in-flation), In 1981 dollars, these estimates become$1.40 to 2.10/gge and $0.90 to $1.40/gge, re-spectively. This compares with a reference costof gasoline from $32/bbl crude oil of $1.20/gal(plus $0.1 7/gal taxes).

Extreme caution should be exercised when in-terpreting these figures, however. They representthe best current estimates of what fossil synfuelswill cost after technical uncertainties have beenresolved through commercial demonstration.They do not include any significant cost increasesthat may occur from design changes, hyperinfla-tion in construction costs, or construction delays.They most likely represent a lower limit for thesynfuels costs.

Table 46.—Consumer Cost of Various Synthetic Transportation Fuels

1000/0 equity financing, IO% real return on investment25°/0 equity/75% debt financing,10% real return on investment

Plant or refinerygate product cost Delivered consumer cost of fuel a

1960 dollars 1960 dollarsper barrel per gallon

oil equivalent gasoline equivalentProcess (5.9 MMBtu/B) (125 k Btu/gal) 1980 $/MMBtu

Plant or refinerygate product cost Delivered consumer cost of fuel a

1980 dollars 1960 dollarsper barrel per gallon

oil equivalent gasoline equivalent(5.9 MMBtu/B) (125 k Btu/gal) 1980 $/MMBtu

Reference cost of gasolinefrom $32/bbl crude oil . . 47 1.20 9.50 1.20 9.50

Oil shale. . . . . . . . . . . . . . . . 52 b

1.30 10.40 3 3b 0.90 7.20Methanol/SNG from coal . . 43 1.30 c d 10.60 c 25 0.95 C d 7.50 c

Methanol from coal . . . . . . 58 1.60 d

13.00 33 1.10 d 8.60Coal to methanol/SNG,

Mobil methanol togasoline . . . . . . . . . . . . . . 49 e 1.25C d 10.00 c 29 e 0.80 C d 6.40 C

Coal to methanol, Mobilmethanol to gasoline . . . 67 e 1.60 d 12.90 38 e 1.00 d

8.10Fischer-Tropsch/SNG

from coal . . . . . . . . . . . . . 52 e 1 .30 c 10.40 C 3o e 0.85 C 6.70 C

Direct coal liquefaction . . . 77 f

1.85 14.60 5 1 f

1.25 10.20SNG from coal . . . . . . . . . . 1.15 g 9.10 g 0.75 g 5.90 g

Methanol from wood . . . . . 68 (79) h 1.65 (2.05) h 14.70 (16.60)h 49 (60) h 1.45 (1.70) h 11.60 (13.40) h

Ethanol from grain . . . . . . . 71 (87)’ 1.60 (2.15)’ 14.50 (17.10)’ 60 (75)’ 1.55 (1.90)’ 12.60 (15.10)’aA9~uming ~.~pfryaical gatlon delivery charge and mark-uP; fttei tties not inciuded.blnclude~ $e/bbl refining ~o~t. Derived from R. F. Sulllvm, et al., “Refining and IJpgr~ing of Syfrfueis From (hal and 011 Shaies by Advanced htdytiC processes,”

first interim report by Chevron Research Corp. to the Department of Energy, April 1978, by increasing cost of S4.541!bbl by 22 percent to reflect 19S0 dollars and ad-justing for 88 percent refinery efficiency.

cAssumes copr~uct SNG selling for same pdce per MMBtu at the Plant 9ate ss the iiquid Pr~uct.dAlthough the plant or refiney gate cost of methanol iS lower than MTG gasoline, the deilvered consumer CO$t Of methanoi is I’rlgher due to the higher cost of dellvedn9

a given amount of energy in the form of methanol aa compared with gasoline, because of the lower energy content per gallon of the former.eAii necessa~ refining is included in the COflvW8iOfl Plant.f Inciudes $l~bbi refining cost from R F. suliiv~ and H. A. F~mkin, “Refining ~d Upgrading of Synfuels From cOd and Oil Shales by Advanced cddytiC prOC9SSeS,

Third Interim Report,” report to the Department of Energy, Apr. 30, 19S0, NTIS No. FE-2315-47. Refining coats fOr EDS and H-Coal are assumed to be the same asSRC Il. Note, however, that reflnlng costs drop to $10/bbl for production of heating oil and gasollne and Increase to S16.Wbbl for production of easo)he CW+.

g$l.@IMMBtu delivery charge and markup, Wh[ch corresponds to the 19s0 difference between the welihead and residential price of natural gas. Energy InformationAdministration, U.S. Department of Energy, “19S0 Annual Report to Congress,” vol. 2, DOE/EIA-0173 (S0)/2, pp. 117 and 119.

hAss ume s ~dy ton wood. Number in parentheses corresponds to S@dv ton wood.i Assumes $3/bu corn Number In parentheses corresponds to ti.~lbu. corn.



Table 47.—Assumptions Figure 15.—Consumer Cost of Selected SynfuelsWith Various Aftertax Rates of Return on Investment

1, Project life—25 years following 5-year construction periodfor fossil synfuels and 2-year construction period forbiomass synfuels.

2.10 year straight-line depreciation.3. Local taxes and property insurance as in K. K. Rogers,

“coal Conversion Comparisons,” ESCOE, prepared for U.S.Department of Energy under contract No. EF-77-C-01-2468,July 1979.

4.10 percent real rate of return on equity investment with:1) 100 percent equity financing, and 2) 75 percent debt/25percent equity financing with 5 percent real interest rate.

5.90 percent capacity or “onstream factor.”6. Coal costs $30/ton delivered to synfuels plant (1980 dollars).7.46 percent Federal and 9 percent State tax.8. Working capital = 10 percent of capital investment.SOURCE: Office of Technology Assessment.

Table 48.—Effect of Varying Financial Parametersand Assumptions on Synfuels’ Costs

Change Effect on synfuels cost

Plant operates at a 50 percenton stream factor rather than90 percent . . . . . . . . . . . . . . . . . Increase 60-70%

Increase capital investment by50 percent . . . . . . . . . . . . . . . . . Increase 15-35%

8-year construction rather than5-year . . . . . . . . . . . . . . . . . . . . . Increase 5-20%

Increase coal price by $15/ton . . Increase $5-7/bblSOURCE: Office of Technology Assessment.

Because cost overruns and poor plant perform-ance lower the return on investment, investorsin the first round of synfuels plants are likely torequire a high calculated rate of return on invest-ment to ensure against these eventualities. Putanother way, anticipated cost increases (table 45)would lead investors to require higher productprices than those in table 46 before investing insynfuels.

The effects on product costs of various ratesof return on investment are shown in figure 15for selected processes, and the effect of changesin various other economic parameters is shownin table 48. As can be seen, product costs couldvary by more than a factor of 2 depending on thetechnical, economic, and financial conditionsthat pertain.

Nevertheless, it can be concluded that factorswhich reduce the equity investment and the re-quired return on that investment and those whichhelp to ensure reliable plant performance are the

0 5 10 15 20 25Real return on investment (o/o)

a5% real interest rate on debt.


most significant in holding synfuels costs down.These factors do not always act unambiguously,however. Inflation during construction, for exam-ple, increases cost overruns, but inflation afterconstruction increases the real (deflated) returnon investment. The net effect is that the synfuelscost more than expected when plant productionfirst starts, but continued inflation causes theprices of competing fuels to rise and consequentlyallows synfuels prices—and returns on investment—to rise as well. Similarly, easing of environmen-tal control requirements can reduce the time andinvestment required to construct a plant, but in-adequate controls or knowledge of the environ-mental impacts may lead to costly retrofits whichmay perform less reliably than alternative, lesspolluting plant designs.

Another important factor influencing the costof some synthetic transportation fuels is the priceof coproduct SNG. In table 46 it was assumedthat any coproduct SNG would sell for the sameprice per million Btu (MMBtu) at the plant gateas the liquid fuel products, or from $4 to $9/


MMBtu, which compares with current well headprices of up to $9/MM Btu24 for some decontrolledgas. (These prices are averaged with much largerquantities of cheaper gas, so average consumerprices are currently about $3 to $5/MMBtu.)However, the highest wellhead prices may notbe sustainable in the future as their “cushion”of cheaper gas gets smaller, causing averge con-sumer prices to rise. This is because consumerprices are limited by competition between gasand competing fuels—e.g., residual oil—andprobably cannot go much higher without caus-ing many industrial gas users to switch fuels. Largequantities of unconventional natural gas mightbe produced at well head prices of about $10 to$1 l/MMBtu,25 so SNG coproduct prices are un-likely to exceed this latter value in the next twodecades. If the SNG coproduct can be sold foronly $4/MMBtu or less, synfuels plants that donot produce significant quantities of SNG willlikely be favored. However, for the single-prod-uct indirect liquefaction processes, advancedhigh-temperature gasifiers, rather than the com-mercially proven Lurgi gasifier assumed for theseestimates, may be used. This adds some addition-al uncertainty to product costs.

Despite the inability to make reliable absolutecost estimates, some comparisons based on tech-nical arguments are possible. First, oil shale prob-ably will be one of the lower cost synfuelsbecause of the relative technical simplicity of theprocess: one simply heats the shale to producea liquid syncrude which is then hydrogenated toproduce a high-quality substitute for naturalcrude oil. However, handling the large volumesof shale may be more difficult than anticipated;and, since the high-quality shale resources arelocated in a single region and there is only a

ZAProcesS GaS Consumer Group, Process Gas Consumers Repofi,Washington, D. C., June 1981.

25J. F. Bookout, Chairman, Committee on Unconventional GasSources, “Unconventional Gas Sources,” National Petroleum Coun-cil, December 1980.

limited ability to disperse plants as an en-vironmental measure, large production volumescould necessitate particularly stringent andtherefore expensive pollution control equipmentor increase waste disposal costs.

Second, regarding the indirect transportationliquids from coal, the relative consumer cost (costper miles driven) of methanol v. synthetic gaso-line will depend critically on automotive technol-ogy. Although methanol plants are somewhat lesscomplex than coal-to-gasoline plants, the cost dif-ference is overcome by the higher cost of termi-nalling and transporting methanol to a service sta-tion, due to the latter’s lower energy content pergallon. With specially designed engines, how-ever, the methanol could be used with about 10to 20 percent higher efficiency than gasoline. Thiswould reduce the apparent cost of methanol,making it slightly less expensive (cost per mile)than synthetic gasoline. * Successful developmentof direct injected stratified-charge engines wouldeliminate this advantage, while successful devel-opment of advanced techniques for using metha-nol as an engine fuel could increase methanol’sadvantage. * *

This analysis shows that there is much uncer-tainty in these types of cost estimates, and theyshould be treated with due skepticism. The esti-mates are useful as a general indication of thelikely cost of synfuels, but these and any othercost estimates available at this time are inade-quate to serve as a principal basis for policy deci-sions that require accurate cost predictions withconsequences 10 to 20 years in the future.

*lf gasoline has a $0.10/gge advantage in delivered fuel pricefor synfuels costing $1 .50/gge, methanol would have an overall$0.20/gge advantage when used with a specially designed engine.This could pay for the added cost of a methanol engine in 2 to 4years (assuming 250/gge consumed per year).

**For example, engine waste heat can be used to decomposethe methanol into carbon monoxide and hydrogen before the fuelis burned. The carbon monoxide/hydrogen mixture contains 20 per-cent more energy than the methanol from which it came, with theenergy difference coming from what would otherwise be waste heat.

DEVELOPING A SYNFUELS INDUSTRY

Development of a U.S. synfuels industry can and proven and commercial-scale operation es-be roughly divided into three general stages. Dur- tablished. The second phase consists of expand-ing the first phase, processes will be developed ing the industrial capability to build synfuels


plants. in the third stage, ynfuels production isbrought to a level sufficient for domestic needsand possibly export. Current indications are thatthe first two stages could take 7 to 10 years each.Some of the constraints on this development areconsidered next, followed by a description of twosynfuels development scenarios.

Constraints

A number of factors could constrain the rateat which a synfuels industry develops. Thosementioned most often include:

●

●

●

Other Construction Projects. –Constructionof, for example, a $20 billion to $25 billionAlaskan natural gas pipeline or a $335 billionSaudi Arabian refinery and petrochemical in-dustry, if carried out, would use the sameinternational construction companies, tech-nically skilled labor, and internationally mar-keted equipment as will be required for U.S.synfuel plant construction.26

Equipment. –Building enough plants to pro-duce 3 million barrels per day (MMB/D) offossil synfuels by 2000 will require significantfractions of the current U.S. capacity for pro-ducing pumps, heat exchangers, compres-sors and turbines, pressure vessels and reac-tors, alloy and stainless steel valves, drag-lines, air separation (oxygen) equipment, anddistillation towers.27 28 29 30

critical Materials. -Materials critical to thesynfuels program include cobalt, nickel,molybdenum, and chromium. Two inde-pendent analyses concluded that only chro-mium is a potential constraint.31 32 (Current-ly, 90 percent of the chromium used in theUnited States is imported.) However, devel-opment of 3 MMB/D of fossil synfuels pro-duction capacity by 2000 would require only

Zbgusiness Week, Sept. 29, 1980, P. 83.ZzjbidozBBechtel International, Inc., “Production of Synthetic Liquids

From Coal: 1980-2000, A Preliminary Study of Potential impedi-ments,” final report, December 1979.

Z9TRW, op. cit.JoMechanical Technology, Inc., “An Assessment of Commercial

Coal Liquefaction Processes Equipment Performance and Supply,”January 1980.

31 [bid.

JzBechtel International, InC., op. cit.

●

●

●

●

●

●

7 percent of current U.S. chromium con-sumption .33

Technological Uncertainties. -The proposedsynfuels processes must be demonstratedand shown to be economic on a commer-cial scale before large numbers of plants canbe built.Transportation. — If large quantities of coalare to be transported, rail lines, docks, andother facilities will have to be upgraded.34

New pipelines for syncrudes and productswill have to be built.Manpower.–A significant increase in thenumber of chemical engineers and projectmanagers will be needed. For example,achieving 3 MMB/D of fossil synfuels capaci-ty by 2000 will require 1,300 new chemicalengineers by 1986, representing a 35-percentincrease in the process engineering workforce in the United States.35 More of othertypes of engineers, pipefitters, welders, elec-tricians, carpenters, ironworkers, and otherswill also be needed.Environment, Health, and Safety. -Delays inissuing permits; uncertainty about standards,needed controls, and equipment perform-ance; and court challenges can cause delaysduring planning and construction (see ch.10). Conflicts over water availability couldfurther delay projects, particularly in theWest (see ch. 11).Siting. –Some synfuels plants will be built inremote areas that lack the needed technicaland social infrastructure for plant construc-tion. Such siting factors could, for example,increase construction time and cost.Financial Concerns. –Most large synfuelsprojects require capital investments that arelarge relative to the total capital stock of thecompany developing the project. Conse-quently, most investors will be extremelycautious with these large investments andbanks may be reluctant to loan the capitalwithout extensive guarantees.

JJlbid.JdThe Direct Use of Coal: Prospects and Problems of Production

and Combustion, OTA-E-86 (Washington, D. C.: U.S. Congress, Of-fice of Technology Assessment, April 1979).

JsBechtel International, Inc., Op. cit.


None of these factors can be identified as anoverriding constraint for coal-derived synfuels,although the need for commercial demonstrationand the availability of experienced engineers andproject managers appear to be the most impor-tant. There is still disagreement about how im-portant individual factors like equipment avail-ability actually will be in practice. However, themore rapidly a synfuels industry develops, themore likely development will cause significant in-flation in secondary sectors, supply disruptions,and other externalities and controversies. But theexact response of each of the factors in synfuelsdevelopment is not known. For oil shale, on theother hand, the factor (other than commercialdemonstration) that is most likely to limit the rateof growth is the rate at which communities in theoil shale regions can develop the social infrastruc-ture needed to accommodate the large influx ofpeople to the region.36

The major impacts of developing a large syn-fuels industry are discussed in chapters 8 through10, while two plausible synfuels developmentscenarios are described below.

Development Scenarios

Based on previous OTA reports37 38 estimatesof the importance of the various constraints dis-cussed above, and interviews with Governmentand industry officials, two development scenarioswere constructed for synthetic fuels production.It should be emphasized that these are not pro-jections, but rather plausible development sce-narios under different sets of conditions. Fossilsynthetics are considered first, followed by bio-mass synfuels; and the two are combined in thefinal section.

Fossil Synfuels

The two scenarios for fossil synthetics areshown in table 49 and compared with other esti-mates in figure 16. It can be seen that OTA sce-— . — — .

jbAn Assessment of Oil shale Technologies, Op. cit.

Jzlbici.j8Energy From 6io/ogica/ f%m?Sses, Op. C It.

Table 49.—Fossil Synthetic FuelsDevelopment Scenarios (MMB /DOE )

Year

Fuel 1980 1985 1990 1995 2000

Low estimateShale oil . . . . . . . . . . . . . — O 0.2 0.4 0.5Coal liquids . . . . . . . . . . — — 0.1 0.3 0.8Coal gases . . . . . . . . . . . — 0.09 0.1 0.3 0,8

Total . . . . . . . . . . . . . . — 0.1 0.4 1.0 2.1

High estimateShale oil . . . . . . . . . . . . . — O 0.4 0.9 0.9Coal liquids . . . . . . . . . . — — 0.2 0.7 2.4Coal gases . . . . . . . . . . . — 0.09 0.2 0.7 2.4

Total . . . . . . . . . . . . . . — 0.1 0.8 2.3 5.7SOURCE: Office of Technology Assessment

narios are reasonably consistent with the otherprojections, given the rather speculative natureof this type of estimate.

In both scenarios, the 1985 production of fossilsynthetic fuels consists solely of coal gasificationplants, the only fossil synfuels projects that aresufficiently advanced to be producing by thatdate. For the high estimate it is assumed that eightoil shale, four coal indirect liquefaction, and threeadditional coal gasification plants have been builtand are operating by 1988-90. If no major tech-nical problems have been uncovered, a secondround of construction could proceed at this time.

Assuming that eight additional 50,000-bbl/dplants are under construction by 1988 and thatconstruction starts on eight more plants in 1988and the number of starts increases by 10 percentper year thereafter, one would obtain the quan-tities of synfuels shown for the high estimate. Ten-percent annual growth in construction starts waschosen as a high but probably manageable rateof increase once the processes are proven.

Oil shale is assumed to be limited to 0.9 MMB/D because of environmental constraints and, pos-sibly, political decisions related to water availabil-ity. Some industry experts believe that neither ofthese constraints would materialize because, atthis level of production, it would be feasible tobuild aqueducts to transport water to the region,and additional control technology could limit


Figure 16.—Comparison of Fossil Synthetic Fuel Production Estimates

plant emissions to an acceptable level. If this isdone and salt leaching into the Colorado Riverdoes not materialize as a constraint, perhapsmore of the available capital, equipment, andlabor would go to oil shale and less to the alter-natives.

The low estimate was derived by assuming thatproject delays and poor performance of the firstround of plants limit the output by 1990 to about0.4 MMB/D. These initial problems limit invest-ment in new plants between 1988-95 to about

the level assumed during the 1981-88 period, butthe second round of plants performs satisfactorial-Iy. This would add 0.6 MMB/D, assuming that thefirst round operated at 60 percent of capacity,on the average, while the second round operatedat 90 percent of capacity (i.e., at full capacity 90percent of the time). Following the second round,new construction starts increase as in the highestimate.

In both estimates, it is assumed that about halfof the coal synfuels are gases and half are liquids.


This could occur through a combination of plantsthat produce only liquids, only gases, or liquid/gascoproducts. Depending on markets for the fuels,the available resources (capital, engineering firms,equipment, etc.) could be used to construct facil-ities for producing more synthetic liquids and lesssynthetic gas without affecting the synfuels totalsignificantly.

When interpreting the development scenarios,however, it should be emphasized that there isno guarantee that even the low estimate will beachieved. Actual development will depend critic-ally on decisions made by potential investorswithin the next 2 years. I n addition to businesses’estimates of future oil prices, these decisions arelikely to be strongly influenced by availability ofFederal support for commercialization, in whichcommercial-scale process units are tested andproven. Unless several more commercial projectsthan industry has currently announced are in-itiated in the next year or two, it is unlikely thateven the low estimate for 1990 can be achieved.

Biomass Synfuels

Estimating the quantities of synfuels from bio-mass is difficult because of

Box D.-Definitions ofand Commercial-Scale

the lack of data on

DemonstrationPlants

. After laboratory experiments and bench-scaletesting show a process to be promising, ademonstration plant may be built to futher testand “demonstrate” the process, This plant is notintended to be a moneymaker and generally hasa capacity of several hundred to a few thousandbarrels per day. The next step may be variousstages of scale up to commercial scale, in whichcommercial-scale process units are used andproven, although the plant output is less thanwould be the case for a commercial operation.For synfuels, the typical output of a commercialunit may be about 10,000 bbl/d. A commercialplant would then consist of several units oper-ating in parallel with common coal or shalehandling and product storage and terminal fa-cilities.

the number of potential users, technical uncer-tainties, and uncertainties about future croplandneeds for food production and the extent towhich good forest management will actually bepracticed. OTA has estimated that from 6 to 17quadrillion Btu per year (Quads/yr) of biomasscould be available to be used for energy by 2000,depending on these and other factors.39g At thelower limit, most of the biomass would be usedfor direct combustion applications, but therewould be small amounts of methanol, biogas, eth-anol from grain, and gasification as well.

Assuming that 5 Quads/yr of wood and plantherbage, over and above the lower figure for bio-energy, is used for energy by 2000 and that 1Quad/yr of this is used for direct combustion,then about 4 Quads/yr would be converted tosynfuels. If half of this biomass is used in airblowngasifiers for a low-Btu gas and half for methanolsynthesis (60 percent efficiency), this would resultin 0.9 million barrels per day oil equivalent(MMB/DOE) of low-Btu gas and 0.6 MMB/DOEof methanol (19 billion gal/yr). *

The 0.9 MMB/DOE of synthetic gas is about 5percent of the energy consumption in the resi-dential/commercial and industrial sectors, or 9percent of total industrial energy consumption.Depending on the actual number of small energyusers located near biomass supplies, this figuremay be conservative for the market penetrationof airblown gasifiers. Furthermore, the estimatedquantity of methanol is contingent on: 1) develop-ment of advanced gasifiers and, possibly, prefab-ricated methanol plants that reduce costs to thepoint of being competitive with coal-derivedmethanol and 2) market penetration of coal-derived methanol so that the supply infrastruc-ture and end-use markets for methanol are readilyavailable. OTA’s analysis indicates that both as-sumptions are plausible.

In addition to these synfuels, about 0.08 to 0.16MMB/DOE (2 billion to 4 billion gal/yr) of etha-nol* * could be produced from grain and sugar

JglbidO*If advanced biomass gasifiers methanol can be produced with

an overall efficiency of 70 percent for converting biomass to meth-anol, this figure will be raised to 0.7 MMB/DOE or 22 billion gal/yr.

**Caution should be exercised when interpreting the ethano( {ev-els, however, since achieving this level will depend on a complexbalance of various forces, including Government subsidies, marketdemand for gasohol, and gasohol’s inflationary impact on foodprices.


crops, and perhaps 0.1 MMB/DOE of biogas andSNG from anaerobic digestion. * Taking thesecontributions together with the other contribu-tions from biomass synfuels results in the high andlow estimates given in table 50.

Summary

Combining the contributions from fossil andbiomass synfuels results in the two development

*The total potential from manure is about 0.14 MMB/DOE, butthe net quantitity that may be used to replace oil and natural gasis probably no more than so percent of this amount. In addition,there may be small contributions from municipal solid waste and,possibly, kelp.

Table 50.—Biomass Synthetic FuelsDevelopment Scenarios (MMB /DOE )

Year

Fuel 1980 1985 1990 1995 2000

Low estimateMethanol ab. . . . . . . . . . . – (e) (e) (e) 0.1Ethanol c . . . . . . . . . . . . . (e) (e) (e) (e) (e)Low- and

medium-energyfuel gas a . . . . . . . . . . . (e) (e) ( e ) 0 . 1 0 . 1

Biogas and methane d . . (e) (e) (e) (e)

Total . . . . . . . . . . . . . . (e) (e) (e) 0.2 0.3

High estimateMethanol ab. . . . . . . . . . . – ( e ) 0 . 1 0 . 3 0 . 6Ethanol c . . . . . . . . . . . . . (e) (e) 0.1 . .Low- and

medium-energyfuel gas. . . . . . . . . . . . ( e ) 0 . 1 0 . 3 0 . 7 0 . 9

Biogas and methane d . . (e) ( e ) 0 . 1 0 . 2 0 . 2

Total . . . . . . . . . . . . . . (e) 0.1 0.6 1.3 1.8aFr~rn ~OOd and plant herbage and possibly municiPal solid waste.bEthanol could also be produced from wood and plant herbage, but methanol

Is likely to be a less expensive liquid fuel from these sources.cFrom grains and sugar croPs.dFrom an{mal msnure, municipal solid waste, and, possibly, kelp.eLess than 0,1 MMB/DOE.


scenarios shown in figure 17. Coal-derived syn-fuels provide the largest potential. Ultimately,production of fossil synfuels is likely to be limitedby the demand for the various synfuel products,the emissions from synfuels plants, and the costof reducing these emissions to levels required bylaw. Beyond 2000, on the other hand, synfuelsfrom biomass may be limited by the resourceavailability; however, development of energycrops capable of being grown on land unsuitablefor food crops, ocean kelp farms, and other spec-ulative sources of biomass could expand the re-source base somewhat.

1980 1985 1990 1995 2000

YearSOURCE: Office of Technology Assessment.

Chapter 7


Contents

Page

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183

Current Situation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183

Future Demand . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 184

Fuel Switching . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185

Conservation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .., O........ 188

Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 189

Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191

TABLES

Table No. Page51. U.S. Petroleum Demand . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18452. 1980 Petroleum Demand . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18453.1990 EIA Petroleum Demand Forecast for Stationary Fuel Uses . . . . . . . . . 18454. Summary of Energy Requirements for Displacing 1990 Fuel 0il . . . . . . . . . 18655. Annual Replacement Energy Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . 18856. Investment Costs For Fuel Oil Replacement Energy . . . . . . . . . . . . . . . . . . . 18857. Energy Made Available by Conservation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 189

Chapter 7


INTRODUCTION

Stationary users–buildings, industry, and elec-tric utilities—consumed about 8.1 million barrelsper day (MMB/D) of petroleum products in 1980.While the potential for reducing oil use by thesesectors is well recognized, it has not received asmuch attention as oil reduction opportunities intransportation. Indeed, U.S. energy policies in the1970’s implicitly encouraged increased oil use forstationary purposes. Lately, however, policy ob-jectives have been set to encourage reduction inoil use by fuel switching and conservation. Theseobjectives include conservation goals and incen-tives to increase energy use efficiency by build-ings and industry, and fuel-switching goals to con-vert utility and large industrial boilers from oil andnatural gas to coal.

This section examines the current mix of petro-leum products used in the stationary sector andrecent trends. A Department of Energy forecastof stationary demand on petroleum products wasselected to serve as a baseline. This will be usedto provide estimates of the volume of fuel oil thatcan be saved by either conservation or conver-sion to new natural gas and electricity. Readersshould note that these estimates only describereductions in oil use that are technically and eco-nomically plausible. Whether the estimated re-ductions are actually realized depends on hownumerous energy users and producers react to

Photo credit: Department of Energy

Cracks and very narrow spaces, such as those aroundwindow framing are insulated to increase

energy use efficiency

economic incentives and other factors affectingtheir choices. Some of these factors are discussedat the end of this section.

CURRENT SITUATION

Table 51 shows petroleum use by the stationaryand transportation sectors since 1965.

Stationary sectors have accounted for 45 to 48percent of total petroleum demand over this peri-od. Demand growth has occurred in industry andelectric utilities as a result of natural-gas curtail-ments during the 1970’s, environmental restric-tions on coal, and the rapid increase in electrici-ty demand since about 1973. The type of petro-leum product used is also important, since we

are primarily concerned with products most read-ily converted to transportation fuels. Table 52shows the distribution of major petroleum prod-ucts for 1980 among the stationary and transpor-tation sectors.

As fuel, the stationary sectors consume prin-cipally middle distillates and residual oil. Themajor components of the other category includespetrochemical feedstocks, asphalt, petroleumcoke, and refinery still gas. These are unlikely

183

98-281 0 - 82 - 13 : ~IL 3


Table 51 .–U.S. Petroleum Demand (MMB/D)

Stationary Transportation

Year Industry Buildings Utilities Total Total

1965 2.2 3.0 0.3 5.5 6.01970 2.5 3.5 0.9 6.9 7.81975 2.8 3.2 1.4 7.4 8.91980 3.6 2.9 1.6 8.1 8.7SOURCE: Department of Energy, Energy Information Administration, fWAmrua/

Report to Congress, vol. Il.

candidates for conversion to transportation fuelsbecause modifications to refineries would be re-quired far beyond those needed to convert resid-ual fuel oil. ’ Further, some of these products,such as the petrochemical feedstocks and asphalt,could only be replaced by synthetic liquids. Theliquefied petroleum gases (LPGs) can be used di-rectly as a transportation fuel, as is the case withcars and trucks that have been modified to run

1 Refining Flexibility (Washington, D. C.: National Petroleum Coun-cil, 1980).

on propane. Widespread adoption will dependprincipally on the relative cost of propane com-pared with gasoline, diesel fuel, and methanolwhen the cost of motor vehicle conversion is in-cluded.

Since the current price of natural gas liquids–the major source of propane–is and is likely toremain as high as the price of domestic crude oil,a significant shift to propane-powered vehiclesis not likely. Therefore, LPG was not consideredin this analysis of stationary fuel use. The majortarget of fuel switching and conservation, then,is the 4.4 MMB/D of distillate and residual fueloil in current use. They can be used as a transpor-tation fuel, although it will be necessary to up-grade residual fuel to gasoline and middle distil-lates by modifying the refinery process. Suchmodification is under way but will require con-siderable time and investment.2

20il and Gas Journa[ Jan. 5, 1981, p. 43.

Table 52.—1980 Petroleum Demand (MMBD )a

Stationary Transportation

Product Buildings Industry Electricity Total Total

Gasoline. . . . . . . . . . . . . . . . 0 0 0 0 6.3Distillate. . . . . . . . . . . . . . . . 1.0 0.7 0.2 1.9 0.95Residual . . . . . . . . . . . . . . . . 0.5 0.65 1.3 2.45 0.4Jet . . . . . . . . . . . . . . . . . . . . . o 0 0 0 1.1LPG . . . . . . . . . . . . . . . . . . . . 0.3 0.7 0 1.0 0Other. . . . . . . . . . . . . . . . . . . 0.7 2.0 0 2.7 0.2%iven in terms of product equivalent–5.5 million Btu per Barrel.

SOURCE: Department of Energy, Energy Information Administration, 1980 Annua/ Report to Congress, vol. Il.

Over the next decade some of this 4.4 MMB/Dwill be eliminated by fuel switching and conserva-tion as the price of oil rises. Indeed, a declineof 1.1 MMB/D took place between 1979 and1980.3 How much more is possible by 1990 de-pends on future oil prices, the costs of alterna-tives, the ability to finance these alternatives, andenvironmental and regulatory factors. The 1980Energy Information Administration (EIA) estimateof 1990 demand is shown in table 53. This fore-

Table 53.—1990 EIA Petroleum a Demand Forecast forStationary Fuel Uses (MMB /D)b

Product Buildings Industry Electricity Total

Distillate. . . . . . . . 0.9 0.1 0.15 1.4Residual . . . . . . . . 0.3 0.2 0.9 1.15Total . . . . . . . . . . . 1.2 0.3 1.05 2.55

aThe “other” and LpG category of stationary petroleum demand is fOrecSSt toremain at its 1980 level of 3.8 MMB/D. This category, however, would then in-crease from 48 percent of all stationary uses in 1980 to 60 percent In 1990. Ithas proven to be much less eiastlc to fuel price increases since 1973 than thedistillate–residuai fuel categow.

bBarrei of product—5.5 MMBtu Per barrel.

J 1980 Annua/ Repoti to Congress, VOI. 2, DOEIEIA-01 73(80)/2

(Washington, D. C.: U.S. Department of Energy, Energy informa-tion Administration, 1980).

SOURCE: Department of Energy, Energy Information Administration, Arrnua/Report to Congress, vol. Ill.

Ch. 7—Stationary Uses of Petroleum • 185

cast is based on a 1990 oil price of $45 per bar-rel (bbl) (in 1980 dollars).

The forecast reduction of 1.8 MM B/D over thenext 10 years (from 4.4 MMB/D in 1980 to 2.6MMB/D in 1990) would be accomplished bymore efficient use, and by conversion to coal,electricity, and natural gas. Beyond 1990, con-tinued reduction can be expected, particularlyin the electric utility sector, if the economic ad-vantages of alternate fuels and conservation con-tinue.

For the purpose of this study, OTA determinedthe technology (alternate fuels, conservation) andinvestment necessary to eliminate this 2.6-MMB/D usage during the 1990’s. This is aboutthe same level of reduction that can be achievedby going from a new-car fleet average of 30 milesper gallon (mpg) in 1985 to an average of 65 mpgin 1995, * and is close to the target synthetic fuelsproduction level by 1992 set forth in the EnergySecurity Act.4 Therefore, it provides a good com-

*See ch. 5, p. 127.442 USC 8701, Energy Security Act, June 30, 1980.

parison for the remainder of the study. The restof this chapter describes how this eliminationmight be achieved and what it might cost.

First, fuel switching alone is considered, andsecond, conservation. OTA did not attempt toestimate a timetable other than to assume thatthe reductions take place throughout the 1990’s.This is consistent with the time needed to intro-duce similar savings from increased auto efficien-cy or from synthetic fuels production. * Wherepossible, serious time constraints that may ap-pear are mentioned. The focus, however, is oninvestment costs and resource requirements. Itis important to emphasize that because OTA’scalculations were based on the EIA 1990 forecast,costs of fuel switching and conservation neces-sary to go from the 1980 to 1990 levels of fueloil consumption were not counted. This some-what arbitrary decision will bias against conser-vation and fuel switching, since the least costlysteps are expected to be taken first, during the1980’s.

*See ch. 5, p, 127; ch. 6, p. 177.

FUEL SWITCHING

The prime candidates for eliminating this 2.6MMB/D of fuel oil by fuel switching are naturalgas, coal, and electricity from coal and naturalgas. Indeed, a considerable amount of the oilnow used by industry (about 20 percent) is aresult of converting from natural gas to oil dur-ing the mid-l970’s.5 This was partially a result ofthe Federal curtailment policy for natural gas thatgave low priority in many industrial applications(primarily boilers). Further, the uncertainty ofsupply that existed during that same periodcaused industry to switch other applications fromnatural gas to oil as well. In the buildings sector,“scarcity” of natural gas during the 1970’s, com-bined with the rapid rise in its price, caused atemporary halt in the growth rate of natural gas

51980 Annual Report to Congress, Vol. 2, op. cit., pp. 65 and107. This claim is inferred from these two tables which show a dropin natural gas consumption by industry between 1974 to 1979 ofover 20 percent and a corresponding increase in industrial oil con-su mption.

use. There was no corresponding growth in pe-troleum use, however, unlike the case with in-dustry, since electricity was the primary replace-ment energy.

Complete replacement of fuel oil in buildingsby natural gas alone would require about 2.4 tril-lion cubic feet per year (TCF/yr), assuming cur-rent end-use efficiency. If only electricity wereused, 425 billion kWh/yr of delivered electricenergy would be needed assuming an end-useefficiency increase of 67 percent. * By 1990, mostof the industrial processes that can use coal (pri-marily large boilers) will have been converted be-cause of the large difference in coal and oil pricesthat currently exists.6 If all the remaining fuel oil— — - . —

*Assuming heat pumps (water and space) with a seasonal per-formance factor of 1.25 compared to oil furnaces with a seasonalperformance factor of 0.75.

bcost and Qua/@ of Fuels for Electric Utility Plants, FebruaT 1981,DOE/EIA-0075(81 /02) (Washington, D. C.: U.S. Department ofEnergy, Energy Information Administration, 1981), p. 3.


used by industry were displaced by natural gasalone, 0.6 TCF/yr would be required, assumingno change in end-use efficiency. If electricityalone were used, about 120 billion kWh/yr of de-livered electric energy would be needed, assum-ing a 50-percent increase in end-use efficiency. *

For electric utilities, residual fuel oil is primarilyused in baseload steam plants, while distillate oilis used for peaking turbines. To replace the for-mer by coal would require 135 million tons, andto replace the latter by natural gas would require0.3 TCF/yr. ** Table 54 summarizes the amountof energy needed to replace oil in each sector,assuming that each substitute energy source isused exclusively.

The first question to ask is whether these substi-tute resources will be available. Forecasts for do-mestic natural gas production during the 1990’sby Exxon and EIA are about 14 to 16 TCF/yr.7

Of this, about 70 percent will come from existingreserves, while the remaining will come from newreserves including so-called unconventional gas;i.e., gas from tight sands, geopressured brine, coalseams, and Devonian shale. These latter re-sources are of particular interest since they arethe likely source of any domestic natural gas,above that now forecast, that would be neededto replace stationary fuel oil during the 1990’s.

*Assuming an end-use efficiency of 100 percent for electric heat,compared with 67 percent for oil-fired units. If combustion turbinesare replaced by electric motors for mechanical drive a similar in-crease will occur.

**It was assumed that there would be no change in conversionefficiency upon replacing the residual and distillate oil generationby coal and natural gas generation.

TEnergy Out/ook 19802&X.1 (Houston, Tex.: Exxon CO., U. S. A.,December 1980), p. 10; 1980 Anrwa/ Report to Congress, Vo/. 3,DOE/EIA 01 73(80)/3 (Washington, D. C.: U.S. Department of Energy,Energy Information Administration, 1980), p. 87.

EIA currently forecasts unconventional gas pro-duction increasing from 1.3 TCF in 1990 to 4.4TCF in 2000 at a production cost of about $5.50to $6.50/MCF (in 1980 dollars). 8 EIA predicts,however, that additional volumes of unconven-tional gas will become available in the 1990’s ifnatural gas reaches what would then be the worldprice of oil (about $56/bbl in 1980 dollars). TheNational Petroleum Council recently made a sim-ilar claim, predicting as much as an additional10 TCF/yr becoming available by 2000.9 There-fore, it appears that production of an additional3.3 TCF/yr (relative to that now forecast by EIAand Exxon) could be possible by the mid-l990’sat gas production prices equivalent to about$56/bbl of oil (1980 dollars).

These cost estimates are subject to a great dealof uncertainty, however, and the actual cost ofthis unconventional gas could be considerablyhigher. There is less uncertainty about the abilityto produce this gas increment from unconven-tional sources—particularly tight sands–but otheralternatives, including synthetic natural gas fromcoal, may be cheaper.

Next, consider electricity. Current generationcapacity in the United States is 600,000 MW (in-cluding 100,000 MW using oil) operating at anoverall capacity factor of 45 percent.10 Althoughincreasing the capacity factor to 65 percent whileconcurrently converting all oil units to coal wouldprovide the quantity of electricity needed (seetable 54), that path may not be practical. The pro-

‘i bid., p. 88.qUnconventiona/ Gas %urces, boecutive Summary, Washington,

D. C.: National Petroleum Council, December 1980), p. 5.IOE/~trjc Power Supp/y and Demand, 1981- 19W (Princeton, N. J.:

National Electric Reliability Council, July 1981).

Table 54.-Summary of Energy Requirements for Displacing 1990 Fuel Oil

Replacement energy sources

1990 petroleum forecast Natural gas Electricity Coal(EIA) (MMB/D) (TCF) (billion kWh) (million tons)

Buildings. . . . . . . . . . 1.2 2.4 425 —Industry. . . . . . . . . . . 0.3 0.6 120 —Utilities . . . . . . . . . . . 0.9 (residual) — — 135

0.15 (distillate) 0.3 — —

Total . . . . . . . . . . . 2.55 3.3 545 135SOURCE: Office of Technology Assessment.


file describing the fuel oil load of buildings forheating peaks rather sharply during the winter,and nearly all of the electric energy required toreplace fuel oil would need to be generated dur-ing the 5 months of the heating season. Since theload profile is the primary determinant of the ca-pacity factor, conversion to electric space andwater heating will likely do little to increase theoverall capacity factor. Therefore, as much as120,000 MW of new capacity may be needed.*,

The ease with which this much capacity couldbe added depends on the growth rate for elec-tricity for the remainder of the century in the ab-sence of this oil-to-electricity conversion (under-lying rate), and the financial health of the elec-tric utility industry. These two points are obvious-ly related because an industry which has difficultyraising capital, as is now the case,11 will have dif-ficulty meeting new generation requirements ofany kind. Currently forecasts of the electric de-mand growth rate range from near zero (by theSolar Energy Research Institute (SERl))12 to about3.8 percent per year (by the National Electric Re-liability Council).13

In the latter case, capacity additions becomeso great during the 1990’s that the full incrementof capacity needed for fuel oil replacement(120,000 MW) could be met if the underlying ratedropped to 3.2 percent per year but the utilitiescontinued building at 3.8 percent per year. Ifgrowth of electricity demand in the absence ofour hypothetical fuel switching dropped to 2 per-cent per year, a capacity addition rate of 3 per-cent per year would meet both underlying de-mand and the fuel-switching demand. Underthese conditions, annual capital requirementswould be approximately $25 billion to $35 billion(1980 dollars) and annual capacity additionwould average about 27,000 MW. These are val-ues below those attained by the utility industry

*This is the capacity required to produce 54s billion kWh oper-ating at the current coal-fired average capacity factor of 52 percent.

11 “The Current Financial Condition of the Investor-Owned Elec-tric Utility Industry and Possible Federal Actions to Improve It, ”Edison Electric Institute before the Federal Energy Regulatory Com-mission, Mar. 6, 1981.

IZ~u;/~;ng a S~sta;na~/e Future, Vo/. Z, prepared by the Solar

Energy Research Institute, Committee on Energy and Commerce,U.S. House of Representatives, Washington, D. C., April 1981, p.837.

13E/ectric power supply and Demand, 1981-1990, op. cit.

during the early 1970’s.14 This is manageable pro-vided the current financial problems are solved.If not, providing the replacement electricity fromnew capacity is unlikely.

Finally, there is the question of conversion ofthe electric powerplants that will still be burn-ing oil in 1990 to other fuels (primarily coal).There should be little difficulty producing the ex-tra coal for conversion of the plants burning resid-ual fuel oil.15 Further, as seen above, the naturalgas could be available as a fuel in those plantsburning distillate. There are barriers to convert-ing existing plants including environmental prob-lems of coal, the technical problems in actuallyconverting many of these powerplants, and diffi-culties in financing the conversion projects. Inmany cases it may be less costly to build a newpowerplant at a different site and retire the ex-isting oil-fired plant.

Considering the physical requirements alone,however, there could be adequate supplies, dur-ing the 1990’s, to replace 2.6 MMB/D of distillateand residual fuel oil by some combination of nat-ural gas and electricity along with coal (or pos-sibly nuclear) to replace the oil-fired electric utilityboilers.

Although the cost of this process is difficult tocalculate, it is possible to make an estimate bymaking several arbitrary, but plausible assump-tions based on the above analysis and currentoperating conditions. First, it is assumed thatnatural gas replaces all of the fuel oil used by in-dustry and utility combustion turbines, and halfthe heating oil used by buildings. Second, it isassumed that electricity is used to replace theother half of the heating oil used by buildings.Finally, all electric powerplants using residual fueloil are replaced by coal conversions or new coal-fired powerplants. Table 55 summarizes thereplacement energy requirements for this sce-nario.

To estimate the costs of eliminating this 2.6MMB/D of fuel oil by fuel switching, the follow-ing costs (in 1980 dollars) for the replacementenergy were used:

ldE/ectrjca/ World, Sept. 15, 1980, P. 69.

I SThe Direct Use of Coa[ OTA-E-86 (Washington, D. C.: U.S. con-gress, Office of Technology Assessment, April 1979).


Table 55.—Annual ReplacementEnergy Requirements a

ElectricBuildings Industry utilities Total

Natural gas (TCF) . . . 1.2 0.6 0.3 2.1Electricity (billion

kWh) . . . . . . . . . . . . 225 b – — 225Coal (million tons) . . – – 135 C 135aA99umes an increase in end-use efficiency of 67 percent when switching from

fuel oil to electdcity for space or water heating and no change when switchingfrom fuel oil to natural gee in any of the three sectors.

bRequ}re9 50,000 MW operating at a 50 percent capacity factor.c Assumes the l= oil-fired capacity, which is now forecast to OPerate at a 35

percent capacity factor (N ERC), can be replaced by 55,000 MW of coal-firedcapacity operating at 57.5 percent capacity factor.


1. New coal-fired electric powerplants cost$900/kW, including all necessary environ-mental controls.16

2. Investment costs for natural gas from uncon-ventional sources (tight sands) are approxi-mately $16,500/MCF per day. ” Operatingcosts, including transmission and distribu-tion, are about $1.30/MCF.18

3. The investment cost for new coal surfacemines is approximately $9,000/ton of coalper day. Operating costs are about $6.50/ton.19

4. The cost to convert oil fired capacity to coalis estimated at $600/kW.20

All of these costs are in 1980 dollars. Therefore,they will underestimate the actual costs of con-

IG~ec~~;Cal ASSeSS~e~t Guide, EPRI PS-1 201 -SR (Palo Alto, Calif.:

Electric Power Research Institute, July 1979), pp. 8-11.I TUnconventjona/ Gas Sources, Tight Gas Reservoirs, Part 1, op.

cit., pp. E-1 15.la’’ Natural Gas Issues” (Arlington, Va.: American Gas Associa-

tion, May 1981), p. 8.19’’Comparative Analysis of Mining Synfuels” (Los Angeles, Calif.:

Fluor Corp., 1981).20’’The Regional Economic Impacts on Electricity Supply of the

Powerplant and Industrial Fuel Use Act and Proposed Amend-ments” (Washington, D. C.: Edison Electric Institute, April 1980),p. 37.

version in the 1990’s to the extent there are realincreases in these costs between new and themid-1990’s. Cost estimates are also needed forthe end-use equipment. This was simplified byassuming no change is needed for industry andcombustion turbines in converting from distillateto natural gas. For buildings, heat pumps are usedwhen electricity is the new energy source, andnew gas furnaces are used for natural gas. Basedon current retail estimates these costs are $2,000and $1,200, respectively (1980 dollars), for unitscapable of delivering 100 million Btu of heat perheating season, and having the capacity to meetthe peak-hour heating Ioad.21 Table 56 summar-izes the investment costs per barrel per day re-placed for each sector.

The total investment, obtained by multiplyingthe per unit investment (table 56) by the amountof oil replaced (table 55), is about $230 billion(1980 dollars). These estimates include produc-tion of the energy resource (electric generatingplants, natural gas, and coal mines) and end-useequipment when needed. They do not includecosts to construct new transmission and distribu-tion facilities that might be needed. This omis-sion will be discussed below.

zlAcademy Airconditioning Co., Rockville, Md., private communi-

cation.

Table 56.—investment Costs ForFuel Oil Replacement Energy a b

Buildings Industry Utilities

Natural gas . . . . . . . . . . . . 121,000 100,000 90,000Electricity. . . . . . . . . . . . . . 110,000 – –Coal. . . . . . . . . . . . . . . . . . . – – 54,000aDollars per barrel of oil replaced per day.b Includes investment cost necessary to upgrade the replaced residual fuel oil

to gasoline and diesel where applicable. Cost is $14,000/bbi of residual per dayon the average, Purvin and Gertz, Inc., An Analysis of Potential for UpgradhrgOomest/c Refkrk!g Capacity, prepared for the American Gas Association, Ar-Ilngton, Va., March 1960.


CONSERVATION

The other major alternative for reducing oil use enough natural gas and electricity to substitutein the stationary sectors is conservation. Conser- for the remaining fuel oil. There have been nu-vation cannot completely eliminate fuel oil use merous estimates of conservation potential forby itself–but it can reduce it, and possibly free buildings and industry in the past several years.


The most detailed analysis is that recently com-pleted by SERI.22

The SERI estimates are used to examine thepossibility of and potential costs for eliminatingstationary uses of fuel oil over the period1990-2000. OTA has used the SERI analysis ofconservation measures in the buildings sector toobtain an approximation of the costs of eliminat-ing the remaining stationary uses of fuel oil in allsectors in the 1990-2000 period. These conserva-tion measures reduce the use of fuel oil, electri-city, and natural gas. The electricity and naturalgas saved is used to replace the fuel oil remain-ing after the conservation. In table 57, the SERIprojection for 2000 is given, by fuel, along withthe savings obtained relative to the 1990 baselinedemand (EIA forecast). The savings are the differ-ence between the SERI 2000 projection and theEIA 1990 forecast. * As shown in table 57, conser-vation could eliminate 67 percent of the fuel oilused by buildings and provide more than enough

llBUllding-l Sustainable Future, op. cit., pp. s and 6.

*By using the 1990 EIA forecast rather than the 1990 SERI pro-jection as the starting point, we have compressed some of the sav-ings calculated by SE RI. They assumed that much of the savingswould occur between 1980 and 1990 with a result that their 1990projection of fuel oil use is considerably below the EIA forecast.Our calculation does not change the net savings but only the periodin which they could occur.

natural gas to eliminate the remaining fuel oilused in both buildings and industry, as well asdistillate used by electric utilities. Finally, enoughelectricity would be saved to replace the elec-tricity produced by oil-fired generation (see table54). The amount of natural gas and electricityneeded to do this are given in the last columnof table 57. About 65 percent of the SERI esti-mated savings for the buildings sector alone willachieve the goal.

The SERI study estimated an investment costof $335 billion (in 1980 dollars) to achieve theirsavings goals for 2000. * 23 If we arbitrarily allocatethese costs to the portion of the savings neededsolely for fuel oil elimination, the cost investmentfor the scenario is about $215 billion (65 percentof total). In addition, investment is needed in con-verting end-use equipment from fuel oil to naturalgas in buildings for oil not eliminated by conser-vation. Using the procedure described in the pre-vious section, this amounts to about $10 billion.The total investment for these measures is $225billion–equivalent to $88,000/bbl of oil replacedper day.

*Adjusted to reflect savings not accounted for by OTA’S choiceof base case.

Zjlbld.

Table 57.—Energy Made Available by Conservation

SERI demand EIA demand Savings Savings used forEnergy source estimate (2000) estimate (1990) by 2000 oil substitution

Fuel oil (MMB/D) . . . . . . . . . 0.4 1.2 0.8 —Natural gas (TCF/yr). . . . . . . 4.5 7.8 3.3 1.7Electricity (billion kWh/yr). . 1140 1580 440 250SOURCE” Office of Technology Assessment.

DISCUSSION

The analysis described gives a plausible esti-mate of the technical and investment require-ments of eliminating stationary uses of oil be-tween 1990 and 2000. These requirements arenot forecasts, as mentioned above, but only de-scribe what is technically and economically with-in reason. There are several items, however, thatwere not considered in the calculation that willaffect these costs somewhat. In this section some

of the more important points are briefly dis-cussed.

In calculating the costs of conversion to naturalgas and electricity the possibility was ignored thatnew transmission and distribution equipmentwould be needed. This also holds for the conser-vation case, since natural gas freed by conserva-tion would need to be delivered to sites formerly


using fuel oil, and it is likely that new transmis-sion and distribution facilities would be neededfor some of those locations. A transmission anddistribution operating cost for all the natural gasconversions was included but this is not likely tocover new construction where it is needed.

Similarly, the electricity made available by con-servation may not be able to substitute for oil-fired capacity without the construction of newtransmission lines. In a previous OTA study onsolar energy,24 the construction and operationcosts of both electric and natural gas transmis-sion and distribution systems were calculated.Using those values updated to 1980,25 it wasfound in the worst case–electricity used toreplace oil for heat in buildings—that the cost ofoil replaced should be increased by about 20 per-cent. In all other cases the adjustment is less than10 percent under the unlikely assumption thatall the replacement energy requires new transmis-sion and distribution facilities.

Another point concerns the choice of conserva-tion estimates. The SERI study is the most optimis-tic of several analyses which attempt to calculatethe potential for conservation under least costconditions. The calculations in the SERI study,particularly for buildings, are based on the mostcomplete analysis to date of the thermal charac-teristics of buildings, and include extensive ex-perimental data. Therefore, these engineeringand cost estimates can be considered as attain-able.

Though the SERI calculations were used, it isnot explicitly or implicitly claimed that SERI con-servation targets will be reached. In an OTA studyon building energy conservation recently com-pleted it is estimated that only about 40 percentof the targets will be reached under currentconditions.26 A number of economic constraintsand choices—including restrictive financial con-ditions, uncertainty of results, and high owner dis-count rates—will reduce the probability that these

zdApp/iCation of solar Technology to Today’s Ener~ Needs, VOI.

1, OTA-E-66 (Washington, D. C.: U.S. Congress, Office of Technol-ogy Assessment, june 1978), p. 140.

zSOi/ and Gas ]ourna~ Aug. 10, 1981, p. 76.zbThe Energy Efficiency of Buildings in Cities, OTA-E-168 (Wash-

ington, D. C.: U.S. Congress, Office of Technology Assessment,March 1982).

goals can be met. Even though it was not neces-sary to use the entire SERI estimate of savings toachieve OTA’s hypothetical goal of eliminatingstationary fuel oil use, more than two-thirds wasstill required. Therefore, while it is technicallypossible to eliminate all stationary fuel oil usethrough conservation, this is not likely to happenunder current conditions.

The final point concerns conversion of oil-firedelectric generation capacity to coal. Although thehigh end of the range of estimates for conversioncosts was used, in some instances even this willbe insufficient. There will be sites where conver-sion is impossible because of lack of coal storagefacilities or inadequate coal transportation, orwhere excessive derating of the boiler would benecessary. In such cases, it will make more senseto retire the plant and replace it with new capaci-ty built elsewhere. For these cases, the replace-ment cost will equal the cost of new capacity,including any needed transmission costs. Itshould be expected, however, that new coal ornuclear generation will have a much higher ca-pacity factor than current oil generation becauseof the former’s lower cost of producing energy.

it should be remembered, however, that theload profile will dictate the capacity factor to agreat extent. With the large amount of capacityunder discussion, it can be expected that therewill be sufficient load diversity so that powertransfers between regions will allow for an in-crease in capacity factor to a level that now ex-ists for coal-fired powerplants (about 57 percent).

The possibility of power transfers was assumedand accounted for in the calculation by assum-ing the 57-percent capacity factor. The result isthat less capacity is needed to replace the elec-tricity produced by the oil-fired generation thatis expected to be on-line in 1990. To some degreethis capacity reduction will take care of some ofthe site-specific problems described above.

Another point to be considered is whether coal-fired electricity will be less expensive than thatgenerated from residual fuel oil in the 1990’s. Be-cause of the continuing decline in crude oil qual-ity27—i.e., lower gravity—it will be increasingly

zTOil and Gas Journa[ Apr. 20, 1981, p. 27.

Ch. 7—Stationary Uses of Petroleum ● 191

more costly to refine this oil up to the point whereno residual oil remains. Currently, the averagecost of converting residual fuel oil to middledistillates and gasoline is about $10,000 to$14,000/bbl/d. 28 The marginal cost, however, ismuch higher and will grow as more and moreresidual oil is transformed and as the crude oilfeed becomes heavier.——

ZeAn Ana/yS;~ of potentja/ for Upgrading Domestic Refining

Capacity, op. cit.

Consequently, it is possible that it would becheaper to use the residual oil directly in boilers,as it is used now, and produce the lighter fuelsfrom oil shale or by way of methanol from coal.To reach that point, residual fuel oil would haveto be priced below coal as a boiler fuel becausethe residual oil would have no other market. Noattempt was made to determine when or to whatextent this may occur.

SUMMARY

Elimination of fuel oil use in the stationary sec-tors (buildings, industry, electric utilities) appearsto be technically plausible by 2000. The cost foreither the conservation or fuel-switching scenariowould be high. As shown, the total cost for the1990-2000 period would be about $225 billionto $230 billion to eliminate the 2.6 MM B/D fore-cast still to be in use by 1990. These costs are con-sistent with estimates for synthetic fuels produc-tion and automobile efficiency improvement thatwould produce about the same amount of oil. *In addition, reduction from current use of 4.4MMB/D to the 1990 level will also require severaltens of billions of dollars. As noted earlier, the1980-90 costs were not taken into account in thecalculations.

Uncertainties about these estimates for station-ary fuel oil elimination by conversion arise fromchanges in powerplant construction costs, in coalprices, in the cost of producing natural gas fromtight sands, and in the discovery rate of new

*See ch. 5, p. 139; ch. 6, p. 172.

natural gas. All or any of these couId cause signif-icant swings in the cost of displacing oil, most like-ly upward. In the absence of information aboutthese uncertainties, the estimates given here,which represent the best analyses to date, canbe considered as reasonable. Similarly, for con-servation, uncertainties about the conservationpotential of buildings exist which can only becleared up as more and more buildings are ac-tually retrofit. Preliminary audits of buildingsalready retrofitted have indicated a range of en-ergy savings from 80 percent less than predictedto 50 percent more.29 The sample for this meas-urement was small, but it does indicate the levelof uncertainty.

The estimates that were derived are plausibletargets. They are not forecasts or even necessarilydesirable goals. That will have to be decided with-in the context of all the economic choices possi-ble and within the country’s policy objectivesabout oil imports.

ZgEnergy Effjc/ency of Buildings in Cities, op. cit.

Chapter 8

Regional and National Economic Impacts

Page

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 195

Economic impacts of Automotive Change . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 195Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 195Manufacturing Structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 196Manufacturer Conduct . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197Other Firms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 198

Economic lmpacts of Synfuels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 203The Emerging Industrial Structure of Synfuels. . . . . . . . . . . . . . . . . . . . . . . . . . 204Potential Resource Bottlenecks and Inflation . . . . . . . . . . . . . . . . . . . . . . . . . . 209Finance Capital and Inflation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 214Final Comments About inflation and Synfuels . . . . . . . . . . . . . . . . . . . . . . . . . 215

TABLES

Table No. Page

58. GM’s Major Materials Usage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19859. 1980 Motor Vehicles (MVs) and Parts Supplier Trade . . . . . . . . . . . . . . . . . . 19960. Examples of Supplier Changes and Associated New Capacity Investment . 20161. 1980 Auto Repair Facilities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20162. Potentially Critical Materials and Equipment for Coal Liquids

Development. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21263. Peak Requirements and Present Manufacturing Capacity for Heat

Exchangers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 213

FIGURES

Figure No. Page

18. Tire Industry Trends . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20019. Time Series Comparison: Construction Costs and Consumer Prices . . . . . . 216

Chapter 8

Regional and National Ecomomic Impacts

INTRODUCTION

This chapter examines the types, timing, anddistribution of economic impacts associated withboth development of a synthetic fuels industryusing national coal and oil shale resources, andimproved automobile fuel efficiency. identifyingand assessing these impacts are difficult because:impacts are not distributed evenly in time oracross regions, so that people may not receivebenefits in proportion to the adverse conse-quences they experience; impacts are not trans-latable into directly comparable terms (e.g., dol-lars); the evaluation of impacts is subjective,based on perceptions of the uncertain benefits

cumulative and may be difficuIt to monitor or at-tribute solely to a particular technology choice.

This chapter assesses the broad economic im-pacts of synfuels and changes in auto technology.Chapter 9 further analyzes employment effectsand discusses other social impacts of these tech-nological developments. Decisions about synfuelsand making cars more efficient will require trade-offs in terms of energy use, economic growth, andsocial welfare and equity. There will be both ben-eficial and adverse social consequences for theNation as it moves towards energy independ-

and costs of new technologies;

ECONOMIC

Overview

and impacts are ence.

IMPACTS OF AUTOMOTIVE CHANGE

The economic impacts of improving automo-tive technology result primarily from two factors:the large investments that will be required forassociated capacity, and changes in the goodsand services purchased by the auto manufactur-ers. Large investments increase financial risk, ex-haust profits, and influence the ability of firms toraise outside capital. Changes in goods and serv-ices used by manufacturers affect suppliers and,in turn, local economies. As automotive fueleconomy increases, the structure and conductof the auto industry and the relationship of thedomestic auto industry to the general economychange. Radical increases in demand for fueleconomy, induced either by changes in consum-er preferences or by Government mandates,would lead to greater industry change, most likelyin the form of acceleration or exacerbation of cur-rent trends.

Changes in the auto industry stem from bothtechnological developments and new markettrends, including strong competition from foreignmanufacturers. Large increases in demand for fueleconomy, and for small cars relative to large cars,

encourage the industry to improve the fuel econ-omy of all car classes and to invest in the pro-duction of small cars. These activities help domes-tic manufacturers to satisfy relatively new de-mands, but at the cost of diminished profits dur-ing at least the short term. Profits can fall whenmanufacturers prematurely write off large-car andother capacity investments and change their pric-ing strategies to replace large-car profits withsmall-car profits.

Meanwhile, manufacturers lose money whensales of their least efficient models decline. Highfixed costs and scale economies make their prof-itability vulnerable to sales declines of even a fewpercent. Profits would therefore also fall ifdomestic manufacturers lost market share to for-eign firms. Future opportunities to gain marketshare and profits will be limited by slowing mar-ket growth. *

*Th e u s auto market is nearly saturated (there were 0.73 cars

for every licensed driver in 1979, accordin g to the Motor VehicleManufacturers Association) and the U.S. population is growing slow-ly. Therefore, auto sales will grow at lower rates than in past dec-ades, probably averaging 1 to 1.5 percent per year.

195

—


Manufacturing Structure

The U.S. automotive industry includes threemajor manufacturers—General Motors (GM),Ford, and Chrysler-plus a smaller manufacturer,AMC (now almost half-owned by Renault, aFrench firm) and some very small specialty carmanufacturers. The three major manufacturershave historically been characterized by moderatelevels of vertical integration and broad productlines that include trucks and other vehicles as wellas automobiles. During the past few decades,GM’s operations have been the most extensiveboth vertically and horizontally; Chrysler’s havebeen the least extensive.

Because of the high costs of productionchange, U.S. auto manufacturers are becomingless vertically integrated, relying increasingly onsuppliers to. make components and other vehi-cle parts. For example, the Department of Trans-portation (DOT) reported that in late 1980 alone,domestic manufacturers announced purchasingagreements with foreign suppliers for over 4million 4-cylinder gasoline engines plus severalhundred thousand units of other engines andparts. Reliance on outside suppliers, referred toas “outsourcing,” relieves short-term spendingpressures on manufacturers. By spending less ini-tially to buy parts rather than new plants andequipment (in which to make parts), manufac-turers can afford to make more productionchanges while exposing less cash to the risk offinancial loss due to limited or volatile consum-er demands.

On the other hand, outsourcing may causemanufacturers to lose control over product qual-ity. Also, manufacturers may incur higher vehi-cle manufacturing costs in the longer term be-cause the price of purchased items includes sup-plier profits as well as production costs. Becauseof more severe financial constraints, Ford andChrysler tend to rely on suppliers more than GM.In the future, all domestic manufacturers mayoutsource more from domestic suppliers, foreignfirms, or foreign facilities owned by domesticmanufacturers as a means of reducing capital in-vestments and thus short-term costs.

Manufacturers are consolidating their opera-tions across product lines and engaging in joint

ventures, primarily with foreign manufacturers.While there appears to be no up-to-date sourceof data aggregating these changes, trade journalsand the business press report that American firmsare sharing production and research activitieswith foreign subsidiaries, with foreign firms i nwhich they have equity (Ford with Toyo Kogyo,GM with Isuzu and Suzuki, Chrysler with Mitsu-bishi and Peugeot, AMC with Renault), and withother foreign firms. Joint ventures are also increas-ingly common between non-American firms,which have historically been highly intercon-nected.

Cooperative activity among auto firms world-wide is likely to grow. Many firms will be unableto remain competitive alone, because of thegrowing costs and risks of improving automotivetechnology and increasing competition in mar-kets around the world. The quickest way for U.S.manufacturers to respond to a mandated or de-mand-induced fuel economy increase would beto use foreign automotive concepts directly, bylicensing designs, assembling foreign-made auto-mobile kits, or marketing imported cars undertheir own names. GM and Ford, for example, as-semble Japanese-designed cars in Australia andAMC sells Renaults in the United States.

Domestic companies can make profits by mere-ly selling foreign-designed cars. They can gain ad-ditional manufacturing profits without risking ad-ditional capital if they sell cars made by com-panies in which they have equity. Cooperativeactivity (and, in the extreme, mergers and acquisi-tions) allows firms to pool resources, afford largeinvestments in research and development (R&D)or in plant and equipment, gain scale economies,and spread large financial risks. It is consistentwith the reduction in the number of autonomousauto producers widely predicted by industry ana-lysts.

Although the number of automotive manufac-turing entities is declining worldwide, there maybe continued growth in the number of firms pro-ducing and selling in the United States. Already,Volkswagen produces cars in Pennsylvania andis building a plant in Michigan; Honda is plan-ning to build cars in Ohio; and Nissan is buildinga light truck plant in Tennessee. There are nowabout 23 different makes of foreign cars sold in

Ch. 8—Regional and National Economic Impacts ● 197

the United States, excluding “captive imports”sold under domestic manufacturers’ nameplates(e.g., the Plymouth Colt, which is made by Mitsu-bishi).’ Manufacturers of captive imports, includ-ing Isuzu and Mitsubishi, are already preparingto enter the U.S. market directly.

Manufacturer Conduct

U.S. auto manufacturers are fundamentally al-tering their product, production, and sales strat-egies as automobile technology and consumerdemand change. Several changes in product pol-icy include the following.

First, the number and variety of models is fall-ing. The highest number of models offered by do-mestic manufacturers was 375 in 1970; 255 wereoffered in 1980.2 Manufacturers might sharplyreduce the number of available models to in-crease fuel economy quickly, by producing rela-tively efficient models on overtime and ceasingproduction of relatively inefficient models.

Second, while cars of all size classes are shrink-ing in number, small cars are becoming moreprominent in number, share of capacity, and con-tribution to revenues relative to large cars. Re-cent changes in price strategy have led to smallerprofit differentials by vehicle size and higher ab-solute and relative small-car prices. As individualmodels become more alike in size, manufacturerswill differentiate models by visible options anddesign.

Third, manufacturers may introduce new, oil-conserving products such as very small “mini”cars (e.g., GM’s P-car and Ford’s Optim projects)and vehicles powered by electricity as well as al-ternating fuels.

Cost-reducing alterations to the physical andfinancial characteristics of individual firms–wide-Iy reported in trade journals, the business press,and company publications—help manufacturersadjust to declines in sales and profits and grow-ing investment requirements. Cost-cutting efforts

‘Automotive News, /98 1 Market Data nook (Detroit, Crain Com-munications, Inc., Apr. 29, 1981 ).

2Maryann N. Keller, “Status Report: Automobile Monthly Vehi-cle Market Review” (New York: Paine, Webber, Mitchell, Hutchins,Inc., February 1981).

include reductions in white-collar employmentand elimination of relatively inefficient or un-needed capacity. During the last couple of yearsGM, Ford, and Chrysler have sold or announcedplans to sell several manufacturing and officefacilities. One investment analyst estimates thatsales of assets may have provided over $600 mil-lion to GM and Ford during 1981.3

Efforts to reduce long-term costs focus on meas-ures to improve productivity and reduce laborcosts per unit. To improve productivity, manufac-turers (and suppliers) are already investigating andbeginning to use new types of equipment, plantdesigns, and systems for materials handling, qual-ity control, and inventory management. Industryanalysts and firms also expect that improved coor-dination between management and labor, ven-dors, and Government will be important meansfor improving productivity and competitiveness.Finally, manufacturers maintain that reductionsin hourly labor costs (wages and/or benefits) areessential for making U.S. cars competitive withJapanese cars. Whether, when, and how muchlabor costs are lowered depends on negotiationsbetween manufacturers and the United AutoWorkers union.

Another cost-cutting measure is reduction inplanned capital spending, Spending cutbacks af-fect firms differently, depending on their context.For example, Chrysler reduced 1980 plannedcapital spendings by $2 billion, halting a dieselengine project and others.4 GM has announcedcutbacks that take the form of spending defer-rals and cancellations of planned projects (withlittle effect on immediate cash flow, however).

Another factor which complicates the evalua-tion of cutbacks is that U.S. projects abroad are,and could be, used to supply the U.S. market.Foreign projects are relatively cheap where for-eign partners or foreign governments share in orsubsidize investments. Cost-cutting efforts areconsistent with growth in the share of U.S. autoinvestment and production abroad, because facil-ities in Central and South America, Asia, and inparts of Europe generally produce at lower costs

3MarYann N. Keller, personal communicat ion, 1981.

4Ward’s Automotive Reports, June 15, 1981.


and sell in home markets that are more profitablethan the U.S. market.

Some analysts believe that if extreme pressureswere placed on U.S. manufacturers to make siz-able investment in brief periods of time, Ford andGM (at least) would reduce U.S. production infavor of foreign production (Chrysler has divestedforeign facilities to obtain cash). U.S. manufac-turers and suppliers are already operating withhigh fixed costs, large investment requirements,weak demand, and labor costs higher than for-eign competitors. If there are sharp increases infuel economy demand, or if there are othersources of growth in perceived investment re-quirements–without offsetting changes in manu-facturing and demand/market share–these devel-opments might give auto firms additional incen-tives to curb, if not abandon, auto production inthe United States. If U.S. production were cur-tailed, it would affect production of new, veryefficient small cars while U.S. production of largerand specialty cars would probably continue.Large and specialty cars are characterized by con-sumer demand that is relatively insensitive toprice and in many cases limited to U.S. carbuyers.

Other Firms

Suppliers

Automobile suppliers manufacture a wide vari-ety of products, including textiles, paints, tires,glass, plastics, castings and other metal products,machinery, electrical/electronic items, and oth-ers. Changes in the volumes of different materialsused to produce cars and the ways in which carsare produced are changing the demands on sup-pliers. In the near term, for example, GM predictsthat the average curb weight of its cars will fall21 percent, from 3,300 lb in 1980 to about 2,600lb in 1985, with up to 67 percent more aluminum,48 percent more plastics, and 30 percent less ironand steel, by weight. Rubber use will also fall.GM predicts that steel will comprise a relativelyconstant proportion of car weight, while the pro-portion of iron will fall and aluminum and plas-

tics proportions may even double by 19855 (seetable 58).

Changes in demands for materials and othersupplies create pressures on traditional suppliersto close excess capacity and invest to developor expand capacity for new or increasingly impor-tant products. They also create new business op-portunities for firms whose products becomenewly important to auto manufacturers, such assemiconductor and silicone producers. The de-gree of hardship on individual traditional suppli-ers depends on how much of their business isautomotive and on their resources for change.Like the auto manufacturers, suppliers operatein the context of a cyclical market which cancause their cash flow to be unstable. Table 59indicates the dependence of different suppliergroups on automotive business as of 1980.

The steel and rubber industries have alreadybeen adversely affected by changing auto de-mands together with stronger import competi-tion. Tire manufacturers have suffered with therise in popularity of radial tires (which are re-placed less frequently than bias ply tires and re-quire different production techniques) and thefall in rubber use per vehicle. Between 1975 and1980, over 20 tire plants (about one-third of thedomestic total) were closed, one major tire manu-

5“GM Sees Big Gain for Aluminum, Plastics in ‘Typical’ 1985,”Ward’s Automotive Reports, Apr. 27, 1981.

Table 58.—GM’s Major Materials Usage(per typical car, 1980 v. 1985)

1980 1965

Percent PercentMaterials Pounds total Pounds totalIron. . . . . . . . . . . . . 500 15%0 250-300 10-12%Steel. . . . . . . . . . . . 1,900 58 1,450 58Aluminum . . . . . . . 120 4 145-200 6-8Glass . . . . . . . . . . . 92 3 60Plastics . . . . . . . . . 203 6 220-300 8-12Rubber. . . . . . . . . . 86 3 88 a 3 a

Other . . . . . . . . . . . 377 11 277 11

Total . . . . . . . . . . 3,300 100% 2,600 100%%M projects actual rubber use to be less than SS lb in 19S5.

SOURCE: General Motors Corp., reported in Wards Automotke Reports, Apr. 27,1981.


Table 59.—1980 Motor Vehicles (MVs) and Parts Supplier Trade

Percent of industry output Value of outputIndustry for MVs and parts for MVs and parts

Textiles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 + $4 billion+Wood products . . . . . . . . . . . . . . . . . . . . . . . . . 2 + 618 million +Nonhousehold furniture . . . . . . . . . . . . . . . . . 2.4 260 millionPaper and allied products. . . . . . . . . . . . . . . . 3 2.5 billion +Chemical . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 – 15 billion+Plastics, synthetic rubber, and synthetics. . 6 – 4 billion+Paints and allied products . . . . . . . . . . . . . . . 8 + 900 million +Tire and rubber products (OEM) . . . . . . . . . . 13 4 billion+Glass . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11– 1.3 billionSteel furnaces, foundries, and forgings . . . . 21– 24.6 billionAluminum and aluminum products . . . . . . . . 14.6 4 billion+Copper and other nonferrous metal

products . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11– 6 billion+Metal products and machine shop products 13– 22 billionMetalworkings and industrial machinery . . . 5.6 8 billion+Service industry machinery . . . . . . . . . . . . . . 12 3 billionElectrical and electronic equipment . . . . . . . 5.2 8 billionScientific and controlling instrument . . . . . . 7.5 900 millionNOTES: “’+” means “greater than” and “–” means “less than.” 4’ OEM” stands for “original equipment manufacturer. ”

SOURCE: The Automotive Materials Industry Council of the United States.

facturer (Mansfield) declared bankruptcy, and an-other (Uniroyal) suffered severe financial prob-lems (see fig. 18). Several steel plants were closedduring the same period. In both industries, addi-tional plant closings and continued import com-petition are likely in the 1980’s, although theelimination of excess and inefficient capacity isexpected by Government and private analysts toleave these industries financially healthier.6

Machinery and parts suppliers also face importcompetition and product demand changes. A re-cent Delphi survey of auto suppliers conductedby Arthur Andersen & Co. and the Michigan Man-ufacturers Association (hereafter referred to asA&M) predicted that these suppliers will be in-vesting together at least $2 billion per year in the1980’s, especially for new equipment (about 60percent of total investment). z Machinery invest-ments are needed both to make new types of sup-plied products and to help suppliers adapt to ashortage of skilled machinists. A recent study pre-pared for DOT by Booz-Allen & Hamilton de-scribes the types and levels of investments associ-

W .S. Department of Commerce, 1981 U.S. /nclustria/ Out/ook

(Wash ing ton , D.C: U.S. Government Printing Office, 1981).

7Arthur Andersen & Co. and the Michigan Manufacturers’ Associa-

tion, “Worldwide Competit iveness of the U.S. Automotive Industry

and Its Parts Suppliers During the 1980s” (Detroit: February 1981 ).

ated with different types of auto activities on anew-plant basis (see table 60).8

Analyses by A&M, Government agencies, andindustry analysts suggest that both appreciationof the types of supplier changes needed and abil-ity to make those changes are greater among larg-er supplier firms than among smaller ones. Mostsupplier firms are small- and medium-sized, al-though a few large firms have large shares of theauto supply business. Among GM’s total 32,000suppliers in the United States, for example, only4 percent have at least 500 employees while 52percent have at most 25.9

Auto product change and market volatility areleading large suppliers, in particular, to diversifyinto nonautomotive products. For example, be-tween 1978 and mid-1981 Eaton Corp., a majorsupplier, spent about $470 million to buy com-panies producing electronics, machinery, elec-trical parts, hydraulic systems, and other high-technology goods.10 Large suppliers are also

aBooz.Allen & Hamilton, Inc., Automotive Manufacturing prOc-

esses (Washington, D.C: U.S. Department of Transportation, Na-tional Highway Traffic Safety Administration, February 1981).

9“Supplier Conference, Ford/Europe Interview UnderscoreThreat, ” Ward’s Automotive Repor?s, june 15, 1981.

1°’’Eaton: Poised for Profits From Its Shift to High Technology,”Business Week, June 8, 1981.

98-281 0 - 82 - 14 : (/L, 3


260

240

220

200

Figure 18.—Tire Industry Trends

Consumption patterns for automotive tires

Year

80

20

Major tire manufacturer earnings fall,July 1979 to June 1980, as the table below shows:

Company

Armstrong. . . .Cooper . . . . . .Dunlop . . . . . .Firestone . . . .General . . . . . .Goodrich . . . . .Goodyear . . . .Mohawk . . . . .Uniroyal . . . . .

SOURCE: U.S. Department of Commerce, Bureau of Industrial Economics

strengthening their international operations,diversifying away from the U.S. market. Small-and medium-size firms are likely to follow automanufacturers in undertaking joint R&D and pro-duction ventures, while mergers and acquisitionsand even closings or bankruptcies are likely. * TheA&M survey predicted that decline in the num-bers of suppliers will lead to increased verticalintegration among suppliers, while strong import

Change in earnings

1981 U.S Industrial Outlook January 1981.

competition and other market changes will mo-tivate increases in supplier productivity.

Sales and Service

Other segments of the auto industry includedealers and replacement part and service firms.The latter group, which serves consumers afterthey buy their cars, is called the automotive after-market.

*According to Dun & Bradstreet, transportation equipment firms,primarily including auto suppliers, suffered financial failure at a rateof 101 per 10,000 in 1960, as compared with a rate of 42 per 10,000for all manufacturers.

Dealer sales activities are not necessarily af-fected by changing auto technology per se. Salesdepend on consumer income and general eco-

Ch. 8—Regional and National Economic Impacts • 201

Table 60.—Examples of Supplier Changes and Associated New Capacity Investment a

Approximate capital requirements forCharacteristics property, plant, and equipment

Foundries90 percent of auto castings use iron, 92 percent of which

are sand cast, and auto manufacturers operate about 20percent of U.S. sand casting capacity

Downsizing and production of smaller parts generatesexcess capacity

Materials substitution reduces sand casting with iron andincreases die casting with aluminum

Metal stampingAutos have had up to 3,000 stampings and auto

manufacturers produce about 60 percent of all stampingsby weight

Materials substitution decreases carbon steel, increaseshigh-strength steel and aluminum for stampings

Plastics processingInjection molding

Compression molding

Reaction injection molding

$21 million (typical independent die cast foundryproducing 15,000 tons/year)

$67 million (typical captive plant producing stampingsfor 175,000 cars/year; independent plants aresmaller and cheaper)

$31 million (typical plant producing 65 million lbparts/year)

$43 million (typical plant producing 60 million lbcompound/year

$19 million (typical plant producing 30 million lbparts/year)

aFi~u~~~ for completely new fdities.SOURCE: Booz-Allen & Hamilton, Inc.. Automotive Marrufactudna Processes, meDared for the Department of TransDotiation, National Hiahwav Traffic Safetv Adminis--.

tration, February 1961.

nomic conditions (including the availability ofcredit), demographic conditions (includinghousehold size), the price of fuel, and vehicleprice and quality attributes. Although consumershave responded to recent gasoline price increasesby demanding relatively fuel-efficient cars, the ex-perience of the recent recession illustrates thatoverall sales levels in a given year are primarilydetermined by consumer finances and not by ve-hicular technology.

There are about 300,000 automobile repair fa-cilities in the United States11 (see table 61). Newautomobile technology affects them becauseautomobile design and content are changing. Forexample, problems in new, computer-controlledcomponents will be diagnosed with computer-ized equipment, and plastic parts will be repairedwith adhesives rather than welding. Componentsare more likely to be replaced than repaired onthe vehicle or even at the repair shop. Whileautomobile service firms will have to invest innew equipment and skills to service new cars,continued service needs of older cars may easethe transition,

11 “Auto Repair Facilities Total 300,000, ” Warcl’s Automotive Re-

ports, Apr. 6, 1981.

Table 61 .—1980 Auto Repair Facilities

Type of facility QuantityDealers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26,000Auto repair shops (independent and franchised) 170,000Tire—battery-accessory outlets . . . . . . . . . . . . . 18,850Other auto and home supply stores . . . . . . . . . . 1,860Gasoline stations . . . . . . . . . . . . . . . . . . . . . . . . . . 70,000General merchandise stores . . . . . . . . . . . . . . . . . 3,500All others . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1,430

Total . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 292,240Total including facilities selling only parts,

accessories . . . . . . . . . . . . . . . . . . . . . . . . . . . 331,090

SOURCE: Wards Autornotlve Reports, Apr. 6, 1981.

However, the concurrent operation of very dif-ferent types of cars requires firms to double theirparts inventories to service both types. The dollarvalue of parts and the frequency of repairs arealso likely to differ between new and old cartypes. Manufacturers are attempting to curb serv-ice cost growth by designing cars for easy servic-ing. For example, the Ford Escort and Lynx andthe Chrysler K and Omni/Horizon cars were de-signed so that servicing during the first 50,000miles would cost less than $150.12

— - . —‘zFrancis j. Gaveronski, “Ford’s New Escort, Lynx Designed for

Easy Service,” Automotive News, May 19, 1980, and “Chrysler K-car to Stress Ease of Diagnosis, Repair, ” Automotive News, June

16, 1980.


The effects of new repair and service practiceson the structure of the aftermarket are uncertain.During the past decade, repair and service activ-ity shifted from dealers to service centers run bygeneral retailers (e.g., Sears) and tire retailers(e.g., Firestone) and to specialized franchised cen-ters for tune-ups, body work, or component serv-ice (e.g., AAMCO Transmissions and Midas Muf-fler). Both of these trends help to moderate serv-ice cost increases because of scale economies inplanning and management. However, the inti-mate and advance knowledge of new technolo-gies held by manufacturers is likely to help dealersregain repair and service business. While dealersnow perform about 20 percent of auto repairs,they are expected to gain a greater share by themid-1980’s. Meanwhile, scale economies in ad-vertising and inventory management may pro-mote consolidation among parts firms.13

Prospects

Further financial strain on the domestic autoindustry is not likely to lead to financial failureof major manufacturer and supplier firms (exceptperhaps Chrysler, but Government interventionmakes its future hard to predict). However, thecontinued viability of many smaller auto suppliersis becoming especially uncertain because auto-motive technology changes make products andcapacity obsolete. While the industry may con-tinue to contract, “collapse” of its leading firmsis not likely because major and even intermedi-ate-sized firms can make at least partial adjust-ments to automotive market changes; adjust-ments are already under way. Reduction in theU.S. activities of domestic firms and failure orcontraction of smaller firms would, nevertheless,severely affect employment and local economies.

In contemplating the future of the industry itis important to appreciate what financial failuremeans. In a technical sense, businesses fail whenthey are unable to make scheduled payments.If this inability is temporary, firms can usuallynegotiate with creditors or seek protection frombankruptcy courts to relieve immediate creditor

13M~Vann N. Keller, “Status Report: Auto Parts Industry Automo-tive Aftermarket Quarterly Review” (New York: Paine, Webber,Mitchell, Hutchins, Inc., July 29, 1980).

demands. In many cases, bankrupt firms are suc-cessfully reorganized, structurally as well as finan-cially. However, some firms find that the stigmaof bankruptcy makes producing and selling espe-cially difficult. * If selling a firm’s assets generatesmore value than using them for production bythe firm, the firm is fundamentally unviable, andthere are financial and economic grounds forliquidating it.

Barring Government support or merger, Chrys-ler is the large automotive firm most likely to failif viability in the U.S. market entails large invest-ments that it cannot afford. AMC has been at leasttemporarily rescued by the French Government-backed Renault. Because Chrysler’s financialweakness has been known for years, the magni-tude of the potential social and economic effectsof its failure has been diminishing as Chrysler hascut back its operations and suppliers have re-duced their dependence on Chrysler as a cus-tomer.

In mid-1979, when Data Resources, Inc., pre-pared for the U.S. Department of the Treasurya simulation of the macroeconomic effects of aChrysler bankruptcy and liquidation, it found thatonly temporary macroeconomic instability waslikely to result, although 200,000 people mightbe permanently unemployed. Dependence ofworkers and businesses on Chrysler has dimin-ished since that simulation was done, althoughsmall firms for which Chrysler is a primary cus-tomer remain vulnerable. If Chrysler were to liq-uidate, its exit from the U.S. market would pro-vide opportunities to domestic and foreign manu-facturers to expand market share and purchaseplant and equipment at relatively low cost. Thiscould relieve financial pressures on Ford and GM.

While contraction of the U.S. auto industry mayresult in fewer, healthier firms, employment andlocal economies will suffer.** Loss of jobs will re-

*When Lockheed and Chrysler appealed for Government aid,they both argued that their customers would not buy from firmsin bankruptcy. This is more likely to be a problem for automobile(or aircraft) manufacturers than for their suppliers, given the differ-ence in size of customer purchase and producer liability.

**Also, change in the amount of U.S manufacturer operationsin Canada (not considered “foreign”) could imply violation of ourobligations under the Automotive Products Trade Act agreementswith Canada.

Ch. 8—Regional and National Economic Impacts . 203

suit predominantly from supplier-firm difficuIties.Unemployment of auto industry workers mayalso affect the performance of the national econ-omy. Unemployment causes a more than pro-portionate decline in aggregate production,because slack demand reduces average hours perworker, output per worker, and entry into thelabor force, The reduction in disposable personalincome (DPI) because of unemployment reducespersonal consumption spending. Reduced per-sonal consumption (and business fixed invest-ment) spending reduces gross national product(GNP), causing DPI to fall, and so forth. Both per-sonal and corporate tax revenues decline, whiletransfer payments to unemployed workers andeconomically depressed communities rise.

The national economy can better adjust to autoindustry trauma than local and regional econo-mies because the national economy is more di-versified, and because, over time, national eco-

nomic sensitivity to auto industry problems hasbeen diminishing. Since World War II, manufac-turing employment in the Midwest (and North-east) has been declining as a percent of nationalmanufacturing employment; it has declined in ab-solute volume since 1970 because job opportu-nities have not been growing, foreign and domes-tic firms have located facilities in other regions,and other industries primarily located elsewherehave been growing in their importance to theeconomy. * In this context of structural change,the 1975 recession seems to have been a turn-ing point for traditional Midwest manufacturing,accelerating a trend of decline that was furtheraggravated by the 1979-80 oil crisis and recession.

*Electronics, computing equipment, chemicals and plastics, aero-space equipment, and scientific instruments have been the leadinggrowth industries in the postwar period. These industries are bothoutlets for diversification by auto-related firms and competitors totraditional auto-related firms in automotive supply.

ECONOMIC IMPACTS OF SYNFUELS

Because large blockssynfuels projects, theyeconomic activity even

of capital are required forwill be visible centers offrom a national viewpoint

and, in fact, for’ people outside of the synfuelsindustry, the economic costs and benefits of syn-fuels may be more easily understood in terms ofregional and national impacts.

Despite the absence of commercial experience,an outline of the synfuels industry emerges withcomparisons to coal mining, conventional oil andgas production, chemicals processing, and elec-tric power generation. By itself, this new industrialorganization is an important economic impact,as it changes the way economic decisions aremade regarding the supply of premium fuels. Fur-thermore, along with the technologically deter-mined menu of resource requirements, industrialorganization determines the major regional andnational economic impacts of synfuels deploy-ment.

potential regional and national economic im-pacts are then explored through comparisons ofaggregate resource demands and supplies. Sinceplans call for very large mines and processing

plants, and perhaps many construction projectsin progress at once, the emphasis is on potentialbottlenecks which could delay deploymentschedules and drive up project costs. If severeresource bottlenecks do occur, the resulting in-flation in the prices of these resources will spreadthrough the economy, driving up prices and costsfor a broad range of goods and services.

These resource costs add up in the next sec-tion of this chapter to financial requirements forprojects and for the industry as a whole. To theextent that the Federal Government does not in-tervene, individual firms must compete in finan-cial markets with all other products and all otherfirms for limited supplies of debt and equity cap-ital. With the important exception of methanoland ethanol from biomass, * the large scale andlong Ieadtimes of synfuels projects may make itdifficult to raise capital, especially during the next

*Ethanol from biomass is not included in this discussion becausewith current technology its potential production is limited by theavailability and price of feed-grain feedstocks. However, if economicprocesses for converting Iigno-cellulose into ethanol are developed,ethanol could compete with methanol as a premium fuel frombiomass.

204 • Increased Automobile Fuel Efficiency and Synthetic Fuels: Alternatives for Reducing Oil /reports

decade, when important technological uncertain-ties are likely to remain. Federal subsidies or loanguarantees will speed synfuels deployment–butonly by reducing capital available to other typesof investments, by reducing other Federal pro-grams, by increasing taxes, or by increasing theFederal deficit. Depending on general economicconditions, each of these different market inter-ventions may be inflationary.

Each of these areas of regional and nationaleconomic impacts—industrial structure, poten-tial resource bottlenecks, finance capital, and in-flation as related to synfuels development–is dis-cussed below.

The Emerging IndustrialStructure of Synfuels

Synfuels are fundamentally different from con-ventional oil and gas because they are manufac-tured from solid feedstocks and because synfuelseconomics may lead to the replacement of con-ventional fuels by methanol and low- or medium-Btu gas in the future. Liquids from coal and oilshale, the feedstocks with a natural resource basesufficient to fully displace petroleum in the longrun, involve economies of scale which encourageownership concentration. The methanol option,however, provides offsetting opportunities forlarge chemical firms to enter the liquid fuel busi-ness and, based on biomass feedstocks, it mayalso allow many small producers to supply localmarkets throughout the Nation.

The following discussion is broken down intothe four stages of synfuels production. While thisbreakdown is convenient, it should be under-stood that several stages of production may beperformed on the same site in order to minimizehandling, transportation, and management costs.

Mining Coal and Shale

Mining for synfuels will closely resemble min-ing for any other purpose except that the minesdedicated to synfuels production will be relativelylarge .14 It takes approximately 2.4 million tons of

IQFOr an extensive discussion of mining techniques and costs, see

The Direct Use of Coal: Prospects and Problems of Production andCombustion, OTA-E-86, (Washington, D. C.: U.S. Congress, Officeof Technology Assessment, June 1979), chs. Ill and IV.

coal per year to fire an 800 MWe generator andabout three times that much to feed a 50,000 bar-rels of oil equivalent per day (BOE/D) coal syn-fuels plant, and about four to eight times as muchoil shale (by weight) for the same output of liquidfuel produced by surface retorting. *

Capital costs for development of a coal minedepend primarily on the depth and thickness ofthe coal seam. Average investment cost data canbe misleading, since each mine is unique, but ittakes about $60 of investment per annual ton ofcoal mined underground (1981 dollars). Withcoal preparation and loading facilities, invest-ments at the mine site may approach $100 perannual ton, or about $750 million for capacitysufficient to supply a 50,000 bbl/d synfuels plant.Western surface mining may in certain cases besubstantially less expensive. * * Furthermore, sub-stantial synfuels production may be achieved onthe basis of existing excess mining capacity.***

In the absence of commercial experience, in-vestment cost estimates are unavailable for shalemining. It is clear, however, that they can beeither larger or smaller than for coal, dependingon two opposing factors. First, investments costscould be much higher because of the low energydensity of shale. Hence, much more materialmust be mined per barrel of oil equivalent. Sec-ond, shale investment costs could be lower be-cause major shale resources lie in very thick

*This range is determined by the Btu content of coal and shaleand by the efficiencies of converting a Btu of solids into a Btu offinished liquid fuels. If we just compare shale oil and methanol (thetwo liquid synfuel options which are best understood and probablyof least cost), conversion efficiencies are comparable, so the dif-ference in feedstock rates is entirely a matter of the energy densityof the feedstock. Coal has 16 to 30 MMBtu/T with Western coaltypically on the lower end of the range. Shale, which is presentlyconsidered suitable for retorting, has 3.6 to 5.2 MM Btu/T. Hence,the ratio of shale to coal inputs can be as low as 4.2 and as highas 8.3.

**Investment cost data were obtained from National Coal Associa-tion. Federal surface mine regulations have increased investmentrequirements in increasing the equipment required to operate amine and to reclaim land after coal has been removed, by increas-ing the amount of premining construction and equipment requiredto establish baseline data, and by extending the required develop-ment period.

* * *Th e National Coal Association estimates excess CiipaCity at

100 million to 150 million tons per year. The low end of the rangeis calculated on the basis of the number of mines closed and thenumber of workers working short weeks. The high end of the rangeis calculated by comparing peak weekly production to average an-nual output per week.


seams (in some areas over 1,000 ft thick) andoften relatively near the surface. Estimates for thefirst commercial shale project indicate that min-ing investments for a 50,000 bbl/d project maybe substantially below $750 million. * Further-more, if in situ retorting techniques fulfill optimis-tic expectations, shale mining could become rela-tively inexpensive as mining and retorting opera-tions are accomplished together underground.

Mine investment is important in project plan-ning, but its share in total investment is still usuallyless than a third. (Notice that in the estimated in-vestment costs for coal-based synthetics in ch.8, the cost of the mine was not included. It is in-cluded in ch. 4 in the discussion of total invest-ment costs. ) Beyond actual costs, the activity ofmining itself is important in the synthetic fuelcycle because of its previous absence in the U.S.oil and gas industry. I n fact, the entire sequenceof economic events associated with extraction ofcoal and shale contrasts sharply with the extrac-tion of conventional petroleum and natural gas.The key difference is that oil and gas reservesmust be discovered, with potentially large re-wards for the discoverer, whiIe the location andmorphology of coal and shale resources havebeen known for a long time.

A wildcat driller, looking for an oil or gasdeposit, can rent and operate a drilling rig witha relatively small initial investment. Since the mostpromising prospects have already been drilled inthis country, exploration typically is a high-riskgamble and, although investment is small com-pared with development of resources, it can stillrequire large sums of money in frontier areas suchas deep water or the Arctic. The uncertainty isa deterrent to investment, but potentially largepayoffs and special Federal tax incentives con-

“The Denver office of Tosco Corp. estimates that mine costs forthe Colony project, which is the first shale project to proceed withcommercial development, could be as low as $250 million. Thatparticular site has the advantage that large-scale open pit miningequipment can be used in an underground mine, since the seamis horizontal and the mine can be entered via portals opened ina canyon wall. This means that the reclamation costs of a surfacemine can be avoided as well as the costly mine shaft of a conven-tional underground mine. Furthermore, the site is propitious be-cause there is virtually no methane trapped in the shale, so safetymeasures are minimal. In the future, mine costs as well as conver-sion costs may be held down by in situ liquefaction, but this tech-nology remains unproven.

tinue to attract large numbers of investors andlarge sums of capital.15 Furthermore, the wildcat-ted can induce cooperation from landowners,local government officials, and any other power-ful local interests by promising royalty payments,or at least a rapid expansion of local business ac-tivity, without serious environmental impacts.Only after a substantial reservoir has been dis-covered is it necessary to make relatively largeinvestments in development wells, processingequipment, and pipelines.

In mining, there is nothing comparable to theopportunity and uncertainty of discovery wells.Most of the business parameters of a potentialmine site are evident to the landowner and toall potential mining companies, which means thatprofit margins are generally limited by competi-tive bidding.

As discussed below, mines also typically em-ploy more labor per million Btu of premium fuelproduced than oil and gasfields16 and they havemany more adverse environmental impacts (e.g.,acid drainage, subsidence, etc). For both reasons,interests external to the firm are more likely tooppose and perhaps interrupt mining operations.Investors realize such contingencies and see themas risks for which they expect compensation.

This discussion of relative payoffs and risks isby no means complete or conclusive, but it doessuggest that private investors may exploit min-

151 n 1979, approximately $12.5 billion was invested in explora-

tion for oil and gas in the United States. That includes (in billions),$5.4 for lease acquisition, !$4.5 for drilling, $2.3 for geological andgeophysical activity, and $0.3 for lease rentals. (See Capita/ /inves-tments of the World Petro/eum /ndustry, 1979, Chase Manha t tenBank, p. 20, and Basic Petro/eum Data Book, American PetroleumInstitute, vol. 1, No. 2, sec. Ill, table 8a). $12.5 billion is about 3percent of total gross domestic investment ($387 billion). (See 1980Statistical Abstract, p. 449.)

The oil and gas industry receives special tax treatment mainlyin terms of expensing intangible expenditures of exploration, eventhough they are surely treated as capital expenses in corporateaccounts.

IGA[though oil and gas has been closing steadily, in 1979 it took

approximately 14,500 workers (miners and associated workers) toproduce a Quad of coal and about 11,500 workers to produce aQuad of oil and gas. However, the labor intensity of mining forsynfuels is actually 160 to 200 percent greater than for coal alone,since only about 50 to 60 percent of the energy in coal feedstockremains in the finished synfuel product. See 1980 Statistical Abstract,p. 415, for employment data and 1980 Annual Report to Congress,Energy Information Administration, p. 5, for production data. Seenote 2, ch. 9 for further discussion of labor productivity.


ing prospects for synfuels much more slowly thanprospects for conventional oil and gas, or that in-vestors will accept much greater risks with con-ventional oil and gas prospects because of theoffsetting chances of striking it rich. Synfuelscapacity could still expand rapidly, but probablynot without very high profit incentives to reorientinvestors who have traditionally been in oil andgas exploration.

Conversion Into Liquids and Gases

During the second stage of production, solidfeedstocks are converted into various liquids andgases. Current synfuels project plans indicate thatcoal or shale conversion plants will resemblecoal-fired electric power stations in the sense thatboth convert a large volume of solid feedstockinto a premium form of energy. They will resem-ble chemical processing (in products such as am-monia, ethylene, and methanol from residual oilor natural gas) and petroleum refining facilitiesin their use of equipment for chemical conver-sions at high temperatures and pressures. *

Of the $2 billion to $3 billion (1981 dollars) re-quired overall for a 50,000 BOE/D shale project,between one-third and one-half goes into surfaceretorts which decompose and boil liquid kerogenout of the shale rock. A larger fraction of totalproject costs is required to obtain methanol fromcoal, but with subsequent avoidance of the up-grading and refining costs.** In general (but withthe exception of in situ mining shale), the con-version step alone requires investments compar-able to a nuclear or coal power station of 1 GWecapacity or to outlays for a 200 to 400,000 bbl/dpetroleum refinery .17

Factors other than economy of scale dominatethe economics of syngas production, as demon-

*Refineries typically use lower pressures than chemical plants andlower than what is expected for synfuels conversion.

* * For a breakdown of methanol costs, see ch. 8.I ZAS discussed in ch. 8, a[l synfuek capital cost e5timates are very

uncertain because none of these technologies has been used com-mercially. Furthermore, engineering cost estimates available to OTAtypically do not clearly differentiate costs by stages of production.Nevertheless, the conversion step, going from a solid feedstock toa gas or a liquid product, is undoubtedly the most expensive singlestep in synfuels production. For presentation of costs for electricpower stations see Technica/Assessment Guide, Electric Power Re-search Institute, July 1979.

strated by the existence of many small gasifica-tion plants across the country.18 Two factors ac-count for this. First, gasification is only the firststage in the production of either methane ormethanol, so costs of the second stage can beavoided and system engineering problems areless complex and more within the technical ca-pabilities of smaller users. Airblown gasifiers in-volve the least engineering, since they do not re-quire the production of oxygen, but only certainonsite end users such as brick kilns can use thelow-Btu gas. The second reason involves trans-portation and end-use economics.

In many industrial applications, natural gas(methane) has been the preferred fuel or feed-stock, but medium-Btu gas is an effective substi-tute in existing installations because it requiresrelatively minor equipment changes. Either low-or medium-Btu gas may be used in new installa-tions, depending on the industrial process andsite-specific variables. However, since thesemethane substitutes cannot be transported overlong distances economically, conversion facilitiesmust be located near the end users.

The size of the conversion facility is thereforedetermined by the number and size of gas con-sumers within a given area, and this often dic-tates conversion plants that are small in compari-son with a 50,000 bbl/d liquid synfuels plant.Consequently, industrial gas users may chooseto locate near coalfields in order to produce andtransport their own gas or to contract from dedi-cated sources. Either approach assures securityof supply and availability over many years.

Upgrading and Refining of Liquids

As discussed in chapter 6, raw syncrudes fromoil shale and direct liquefaction must be up-graded and refined to produce useful products.Technically, these activities are quite similar topetroleum refining, and this affords a competitiveadvantage to large firms already operating major,integrated refineries. This bias toward large, es-tablished firms is reinforced in the case of direct

Iasee National Coal Association, “Coal Synfuel Facility Survey,”August 1980, for a listing and discussion of between 15 to 20 low-Btu gas facilities coupled with kilns, small boilers, and chemicalfurnaces.


coal liquids by the apparent cost reduction if up-grading and refining are fully integrated with con-version, thus making it difficult for smaller firmsto specialize in refining as some do today. Up-graded shale oil, on the other hand, is a high-grade refinery feedstock that can be used by mostrefineries.

Downstream Activities: Transportation,Wholesaling, and Retailing

As long as synthetic products closely resembleconventional fuels, downstream activities will berelatively unaffected. However, medium- or low-Btu gas and methanol are sufficiently different torequire equipment modifications, and they maybe sufficiently attractive as alternative fuels to in-duce changes in location of business and struc-ture of competition.

Depending.on the market penetration strategy,methanol may be mixed with gasoline or han-dled and used as a stand-alone motor fuel. As amixture, equipment modifications will involve in-stallation of corrosion-resistant materials in thefuel storage and delivery system. As a stand-alonefuel, methanol may have its own dedicated pipe-line and trucking capacity and its own pump atretail outlets, and auto engines may eventuallybe redesigned to obtain as much as 20 percentadded fuel economy, primarily by increasingcompression ratios and by using leaner air-fuelmixtures when less power is required. *

If firms currently producing methanol for chem-ical feedstocks should enter fuel markets,19 driversstand to gain from the increased competitionamong the resulting larger number of major fuel-producing companies and by competition be-tween methanol and conventional fuel. Further-more, with coal-based methanol providing a criti-cal mass of potential supply, drivers across theNation may be able to purchase fuel from smalllocal producers (using biomass feedstocks), a sit-uation which has not obtained since the demiseof the steam engine.

*See ch. 9 for further information about methanol vehicles.lgAt the present time, approximately 1.2 x 109 gal barrels of meth-

anol (1 1 x 1 @ BOE) are produced domestically, primarily fromnatural gas, and used almost exclusively as a chemical feedstock.See Chemical and Engineering News, )an. 26, 1981.

Medium- or low-Btu gases are effective substi-tutes for high-Btu gas (methane) but, as discussedabove, their relatively low energy density pro-hibits mixing in existing pipelines and generallyrestricts the economical distance between pro-ducer and consumer (the lower the Btu contentthe shorter the distance). Hence, deployment ofthese unconventional gases will require dedicatedpipelines, relocation of industrial users closer tocoalfields, or coal transport to industrial gas-users.

Conclusion and Final Comment

Massive financial and technical requirementsfor synthetic liquids from oil shale and coal en-courage ownership that is more concentratedthan has been typical in conventional oil and gasproduction. Large firms, already established inpetroleum or chemicals, have three major advan-tages.

First, they can support a large in-house techni-cal staff capable of developing superior technol-ogy and capable of planning and managing verylarge projects. Second, they can generate largeamounts of investment capital internally, whichis especially important during the current periodof high inflation (inflation drives up interest onborrowed capital, making it much more expen-sive for smaller firms who must supplement theirmore limited internal funds).

Third, such firms already have powerful prod-uct-market positions where synthetic liquids mustcompete, so entry by new firms involves a greaterrisk that synthetic products cannot be sold at aprofit. * The second and third advantages may be

*Predicting investment behavior is always difficult, but barringFederal policy to the contrary, the most likely group of potentialinvestors are the 26 petroleum and chemical firms, each with 1981assets of $5 billion or more (see list below). Seven chemical firmswere included in this list primarily because they may be in a strongposition to produce and market fuel methanol, based on their ex-perience with methanol as a chemical feedstock.

Foffune, May 4, 1981, presents a listing of the 26 largest (in termsof total assets) petroleum and chemical firms: Exxon, Mobil, Tex-aco, Standard Oil of California, Gulf Oil, Standard Oil of Indiana,Atlantic Richfield, Shell Oil, Conoco, E. 1. du Pent de Nemours, *Phillips Petroleum, Tenneco, * Sun, Occidental Petroleum, Stand-ard Oil of Ohio, Dow Chemical, * Getty Oil, Union Carbide, * UnionOil of California, Marathon Oil, Ashland Oil, Amerado Hess, CitiesService, Monsanto, * W. R. Grace, * and Allied Chemical. * Asteriskindicates firm primarily in the chemicals industry.

This conclusion about the dominance of larger companies holdsdespite the fact that current synfuels projects planned or under study

(continued on next page)


nullified if several smaller firms can effectivelyband together into consortia, but it maybe muchmore difficult for a consortium to build a tech-nical staff which can develop superior technologyand manage large projects during the next dec-ade, when there will be many technical risks.

Ownership concentration is an important as-pect of industrial organization in an economy or-ganized on classical economic principles of anon-ymous competition, market discipline, and con-sumer sovereignty. Very large synfuels projectsowned by very large energy corporations andconsortia of smaller firms would not be anony-mous, even from the viewpoint of the nationaleconomy, and they would have leverage to dic-tate terms in their input and output markets.

Conversely, once companies have made verylarge investments in new synfuels projects, theybecome visible targets for political action whichmight significantly raise costs or reduce output.Visible producers may not in fact allocate re-sources much differently than if there were onlyanonymous competitors, but at least the oppor-tunity to manipulate markets exists where itwould not otherwise—and just the appearanceof doubt about the existence of consumer sover-eignty can raise serious political questions.

The capital intensity of synfuels will also changethe financial structure of the domestic liquid andgaseous fuel industry. Compared with invest-ments in conventional oil and gas during the last20 years, investment in synfuels per barrel of oilequivalent of productive capacity (barrels of oilper day) will increase by a factor of 3 to 5.2

0 While

involve many relatively small firms. For example, three of the fourmajor parties in the Great Plains Gasification Project are primarilyinvolved with either gas-distribution or transmission: American Nat-ural Resources Co., Peoples Energy Corp., and Transco Cos, Inc.American Natural Resources is associated with gas-distribution firmsoperating in Michigan and Wisconsin; Peoples Energy Corp. is asso-ciated with Northern Natural Gas, a major distributor in the Mid-west; and Transco is the parent company of Transcontinental GasPipeline Co., a major operator of transmission lines. The fourth part-ner, Tenneco, is also a major transmission company, but it was in-cluded in the group of top 26 firms listed above because of its chem-ical processing business. Undoubtedly, all four firms’ participationis predicated upon the existence of Government subsidies and loanguarantees, but that is especially true for the three smaller firms.

z~omparison based on data for total costs of oil and gas wells,plus estimated costs for predrilling activities over the period from1959-80. Capital outlays per barrel oil equivalent of reserves over

all such calculations are of necessity very impre-cise, the order of magnitude is confirmed by datacontained in the 1980 Annual Report of ExxonCorp. As of 1980, Exxon’s capitalized assets inU.S. production of oil and gas totaled $11.5 bil-lion, and its average daily production rate (ofcrude oil and natural gas) was about 1.4 millionBOE; so its ratio of capital investment to daily out-put was $8,200.21 A 50,000 BOE/D synfuels plantat $2.2 billion implies a ratio more than five timeslarger ($44,000/BOE/D).

In other words, switching from conventionalto synthetic liquids and gases amounts to a sub-stitution of financial capital (and the labor anddurable goods it buys) for a depleting stock ofsuperior natural resources. A parallel substitutionof investment capital for natural resources is oc-curring as conventional resources are increasinglyhard or expensive to find and develop becauseof the depletion of the finite stockpile of naturalresources.

As long as the United States could keep discov-ering and producing new oil and gas at relative-ly low cost, energy supplies did not impose seri-ous inflexibilities on our economy. When weneeded more we could get it without makingmuch of a sacrifice. With synfuels, it is necessaryto plan ahead, making sure that capital resourcesare indeed available to supply synfuels projects,and that product demand is also going to be avail-able at least a decade into the future so that largesynfuels investments can be amortized.

The current financial situation of many elec-tric utilities in the United States illustrates the risksentailed when plans depend on long-term priceand quantity predictions which may prove to bewrong. It was not long ago that utility investmentswere considered almost risk-free, and the industryhad for decades raised all the debt it wished atlow rates. Needless to say, the utility situation hasnow dramatically reversed as the result of sharplyrising costs embodied in new, long-lived gener-

the past 20 years averaged about $1.60. Depending on the syn-fuels option, a synfuels plant would have a comparable ratio of $5.4oto $7.00/BOE of “reserves.” Well-drilling and other exploration costswere obtained from Society of Exploration Geophysicists, AnnualReports and from joint Association Survey of the U.S. Oil and GasProducing Industry.

ZISee 1980 Annua/ Report of Exxon cOtpOrdtiOn, pp. 34, 44, 51.

Ch. 8–Regional and National Economic Impacts ● 209

ating capacity. while it may be premature todraw an analogy with synfuels, it is clear that syn-fuels will tie up capital in considerably largerblocks and for considerably longer periods thanwas true for conventional oil and gas reservesover the last 30 years.

Compared with synthetic petroleum, methanolpresents two opportunities to partially offset thetendency toward industrial concentration. First,as indicated above, its present use as a majorchemical feedstock provides an opportunity forlarge chemical firms to enter the liquid fuelbusiness. Second, since methanol can be pro-duced from wood and other solid biomass, small-scale conversion plants (approximately $1 O-mil-lion investments) operated by relatively small en-trepreneurs may be able to take advantage oflocal conditions across the country. * Assumingcost competitiveness, having a mixture of small-and large-scale methanol producers may rein-force the attractiveness of downstream equip-ment investments (e. g., retail pumps and engineimprovements), thus making it more likely thatdrivers will indeed have an attractive methanolopt ion.

Besides methanol, synthetic gases may attractadditional large and small firms from outside thepetroleum and chemical industries. Dependingon the deregulated “well head” price of naturalgas (relative to fuel liquids) and depending onregulatory policy regarding utility pricing, syn-thetic natural gas and synthetic medium-Btu gasmay become profitable investments for gas util-ities. Indeed, the first synthetic gas project toreach the final planning stage has substantial gasutility ownership, * * Syngas may become attrac-tive as a methanol coproduct or as a primaryproduct, in either case taking advantage of capitalsavings and higher conversion efficiencies thanif methanol or gasoline is the sole product of in-direct liquefaction.

*One domestic company, International Harvester, is presentlydeveloping technology to mass-produce this equipment and trans-port it to the purchaser’s location in easily assembled modules.

* *This compares with total private domestic investment in 1980of about $395 billion, and out of that total about $294 billion wentfor nonfarm investments in new plant and equipment. Also in 1980,two large blocs of energy investments were $34 billion for oil andgas exploration and production and $35 billion for gas and elec-tric utilities.

A final comment can be made about the loca-tion of the synfuels industry. Shale oil produc-tion will be concentrated in Colorado and Utah,since that is where superior shale resources ex-ist and since unprocessed shale cannot beshipped as a crushed rock without driving upcosts prohibitively, Coal-based synfuels offer thepossibility of spreading liquid fuel productionover a wider cross section of the Nation. This isespecially important for the Northeast and NorthCentral section of the United States, where thereremain substantial coal deposits in Pennsylvania,Ohio, and Illinois, States which have by this timedepleted most of their original petroleum re-serves.

Unlike their shale counterparts, coal-con-version facilities and subsequent upgrading andrefining plants need not be immediately adjacentto the mine mouth, since coal’s shipping costsper Btu are less than for shale. Location offacilities and, hence, their regional impacts willdepend on site-specific factors and the availablemodes of transportation. Location of facilities toconvert biomass into methanol will be deter-mined primarily by local availability and cost ofbiomass feedstocks. This restriction is imposedby the dispersed location of plant material, ratherthan by differences in energy density (biomassfeedstocks such as wood have an energy densi-ty only marginally lower than some Westerncoals).

Potential Resource Bottlenecksand Inflation

Technology, ownership concentration, and (incertain important cases) regional concentration,all combine to impose heavy demands on labor,material, and financial resources relative to cur-rent and potential new supplies of the same re-sources. If deployment plans fail to account forsupply limitations, long project delays and largecost overruns can occur.

Anytime a capital-intensive industry attemptsto start up quickly, temporary factor input short-ages can be expected—if not more extreme “bot-tlenecks” or chronic shortages which generallydisrupt construction schedules. Ideally, shortagesand, certainly, bottlenecks can be avoided by ad-

210 ● Increased Automobile Fuel Efficiency and Synthetic Fuels: Alternatives for Reducing Oil imports

vanced planning and giving suppliers purchasecontracts years in advance if necessary to ensureavailability. However, while such planning andlong-term commitments minimize shortage risks,they also increase risks of loss should plans betechnically ill-conceived and commitments aremade to projects with actual costs much largerthan planned. These two sets of risks must beweighed against each other, but at the presenttime technical risks clearly are more significant.

In order to predict resource bottlenecks andtheir impacts, the full array of supplier marketdynamics must be understood. In this limited dis-cussion, one can only begin to compare poten-tial demands and supplies for key synfuels re-sources.

As a final introductory remark, it should beclear that factor price inflation drives up costs inmany industries, not just for builders of synfuelsplants. Industries that appear most vulnerable toinflation resulting from synfuels deployment willbe identified. However, in general, a much largerstudy would be necessary to trace inflationarypressures through complex interindustry transac-tions.

Experienced Project Planners,Engineers, and Managers

As planned, the construction of oil shale andcoal liquids projects requires the mobilization ofthousands of skilled workers and massive quan-tities of equipment and materials. Of all these syn-fuels investment resources, the supplies of skilledengineers and project managers are the most diffi-cult to measure, and in the final analysis, it is leftup to the large investing firms to decide for eachproject when a critical mass of talent has beenassembled. While individual firms may have ex-cellent engineering departments, the possibilityof supply bottlenecks for chemical engineeringservices, across the full spectrum of chemicalprocessing industries, must be of concern be-cause of the potential financial risks due to designerrors and because of the length of time requiredto educate and train new people. *

● Well-trained engineers and project planners can still make majormistakes, but risks due to miscalculations and design errors are con-trolled by careful training and building up experience incremen-

t the present time, only one of the country’s10 major architectural and engineering (A&E)firms22 has actually built a synfuels plant. * Nocommercial-scale plant has been built. Given thisgeneral inexperience, and making the reasonableassumption that A&E firms will not be short ofwork worldwide, it seems highly unlikely that syn-fuels construction contracts for the first round ofa rapid deployment scheme will be able to holdbuilders to binding cost targets and completiondates. Consequently, those who would actuallytake investment risks may be extremely skepticalof builders’ qualifications and judgment, and thismay severely limit the apparent supply of quali-fied engineers and engineering firms.

Furthermore, if synfuels projects proceed aheadat a rapid pace despite the technical uncertain-ties and commercial inexperience, it could driveup the A&E costs for other large, new process-ing facilities which rely on the same limited groupof A&E firms and the same pool of skilled workers.Of all synfuels resource markets, the possibilities

tally. Commonly accepted periods for obtaining a bachelor’s degreeand subsequent on-the-job training range from 6 to 10 years.

Several recent examples illustrate that errors in the design of largemining and chemical processing plants do occur and can causesevere cost overruns and project delays. Perhaps the most extremecase was the Midwest (nuclear) Fuel Reprocessing Plant built forGeneral Electric. Construction started in 1968, with completionplanned for 1970 at an estimated cost of $36 million. Unfortunate-ly, expected time for major technical component failure in the newpIant was less than the time required to achieve stable operatingconditions. The project was abandoned and the company estimatedthat an additional expenditure of between $9o million and $130million would have been required to redesign and rebuild.

Additional examples include a municipal solid waste gasifier inBaltimore begun in 1973 which never achieved its major goal ofcommercial steam production, an oil sands project in Canada whichundetwent extensive retrofit when the teeth of its large mining shov-els were worn away in a matter of weeks by frozen oil sands, andso on. Clearly, major design errors have happened in the past andare likely in the future, with the number and severity of such errorsincreasing if a shortage of experienced design engineers develops,

For further information about these and other examples of designerrors, see Edward Merrow, Stephen Chapel, and ChristopherWorthing, A Review of Cost Estimation in New Technologies: Implkcation for Energy Process Plants and Corporations, july 1979.

ZZACCOrding to Business Wee/q Sept. 29, 1980, p. w, the 10 major

A&E firms, in order of their largest projects to date, are: Fluor, Par-sons, Bechtel, Foster Wheeler, C-E Lummus, Brown and Root,Pullman Kellog, Stone and Webster, CF Braun, and Badger.

*The Fluor Corp. built Sasol I and II in South Africa and will un-doubtedly sell this technology and its unique experience in theUnited States. However, different resource endowments can causevery different engineering economics in different countries, andthus this existing technical base may have to be adapted to theUnited States by investing in significant additional engineering.


for propagation of inflation from synfuels into therest of the economy is greatest here. Petrochemi-cals, oil refining, and electric power generationare all industries which depend on the same engi-neering resources in order to build new facilities.in 1979, these three industries accounted formore than 25 percent of the total investment innew plant and equipment.23

Mining and Processing Equipment,Including Critical Metals for Steel Alloys

The construction of massive and complex syn-fuels plants will require equally massive and di-verse supplies of processing equipment and con-struction materials. Some of this equipment mustmeet high performance standards for engineer-ing, metals fabrication, component casting, andfinal product assembly because it must withstandcorrosive and abrasive materials under high pres-sure and temperature.

Potential supply problems can be identified firstby comparing projected peak annual equipmentdemand (for each deployment scenario) to cur-rent annual domestic production. While projec-tions were not done specifically for OTA’s lowand high scenarios, useful information can be ex-trapolated from an earlier projection for the de-ployment of coal Iiquids.24 In that analysis, whichpostulated 3 million barrels per day (MMB/D) ofsynfuels by 2000, 7 of 18 input categories wereidentified as questionable because projected syn-fuels demands account for a significant fractionof domestic production. * Supply problems for

zJFor data see Statistical Abstract, 1980, p. 652.ZdData obtained from “A Preliminary Study of Potential impedi-

ments,” by Bechtel National, Inc., which is one part of a three-part compendium, Achieving a Production Goal of 1 Million BIDof Coal Liquids by 1990, TRW, March 1980. We can extrapolatefrom coal liquids to all other synfuels because subsequent research

(by E. 1. Bentz & Associates, OTA contractor) indicates that shaleoil, coal liquids, and coal gases are all quite similar in their totaluse of processing equipment per unit output (measured in dollars)and in their mix of processing equipment. Furthermore, the Bechtelstudy remains useful, despite its age, since subsequent incrementsin synfuels plant costs do not add items to this list or significantlyincrease demand requirements for the group of seven critical items.In other words, recent escalations in plant costs are primarily relatedto increases in the expected prices of components and to increas-ing demands for certain components which are insignificant whencompared with productive capacity nationwide.

*Significance in this case means that projected synfuels demandexceeds 1 to 2 percent of domestic production. Since this is arelatively low threshold, this list should stay about the same for bothscenarios.

chromium, the one item in this group of sevenwhich is not a manufactured piece of equipment,would not be caused by synfuels deployment,since synfuels requirements would amount to lessthan 3 percent of domestic consumption, butsupplies may nevertheless be difficult to obtainbecause U.S. supply is imported, much of it frompolitically unstable southern Africa.25

For the six types of equipment identified, theactual occurrence of bottlenecks will depend onthe ability of domestic industry to expand withsynfuels demand. In all cases, including draglinesand heat exchangers—where coal synfuels re-quirements exceed 75 percent of current domes-tic production even in the low scenario—thereappear to be no technical or institutional reasonswhy, if given notice during the required projectplanning period, supplies should not expand tomeet demand with relatively small price incen-tives.

In general, this optimistic conclusion is basedon the fact that Ieadtimes for expanding capac-

~ity to produce synfuels equipment are shorterthan the Ieadtimes required to definitely plan andthen build a synfuels plant.26 The fact that manyplants would be built at the same time does notnullify this basic comparison as long as all syn-fuels construction projects are visible to supplierindustries, as they should be. Furthermore, for-eign equipment suppliers can be expected tomake up for deficiencies in domestic supply if notactually displace domestic competitors.

For example, consider the case of heat ex-changers. As indicated in table 62, coal synfuels

zSThe chief use of chromium is to form alloys with iron, nickel

or cobalt. In the United States, deposits of chromite ore are foundon the west coast and in Montana. However, domestic produc-tion costs are much higher than in certain key foreign countries.In 1977, South Africa produced about 34 percent of total worldproduction, with the U.S.S.R. and Albania producing another 34percent. Other major producers are Turkey, the Phillipines, andZimbabwe. See Minerals in the U.S. Economy: Ten-Year Supply-Demand Profiles for Nonfuel Mineral Commodities (1968-77),Bureau of Mines, U.S. Department of Interior, 1979.

Zbone can never be certain about how well industrial systems

will adapt to rapidly expanding demand for a limited number ofhighly engineered types of equipment which must be producedwith stringent quality control, However, informal surveys of equip-ment manufacturers have not revealed substantial reasons whyequipment supplies should not be responsive to moderate priceincentives. See Frost and Su Ilivan, Coa/ Liquefaction and Gasifica-tion: Plant and Equipment Markets 1980 Z(X?0, August 1979.


Table 62.—Potentially Critical Materials and Equipment for Coal Liquids Development

(A) (B)Category Units Peak annual requirements U.S. production capacity (A)/(B) (percent)

1. Chromium . . . . . . . . . . . . . . . . . . . . . . . . . . . .2. Valves, alloys, and stainless . . . . . . . . . . . .3. Draglines. . . . . . . . . . . . . . . . . . . . . . . . . . . . .4. Pumps and drivers (less than 1,000 hp). . .5. Centrifugal compressors (less than

10,000 hp. . . . . . . . . . . . . . . . . . . . . . . . . . . . .6. Heat exchangers . . . . . . . . . . . . . . . . . . . . . .7. Pressure vessels (1.5 to 4 inch walls) . . . .8. Pressure vessels (greater than

4 inch walls) . . . . . . . . . . . . . . . . . . . . . . . . . .

tonstonsydhp

hpft 2

tons

tons

10,4005,9002,200

830,000

1,990,00036,800,000

82,500

30,800

070,000

2,50020,000,000

11,000,00050,000,000

671,000

240,000

—8

884

187412

13SOURCE: Achieving a Production Goal of f Million S/0 of Coal Liquids by 19%), draft prepared for the Department of Energy by TRW, Inc. and Bechtel National, Inc.,

March 1980, pp. 4-28. Although these projections apply to the achievement of 3 MM B/D of coal liquids, and not specifically to the low and high productionscenarios Postulated in this reDort, they nevertheless Indicate rouah orders of maanitude for eaui Dment demand. See footnote 16 of this chaDter for furtherdiscussion of alternative synfuels projmtlons.

requirements for the low scenario could accountfor about 75 percent of current domestic U.S.production. Extrapolation from table 62 indicatesthat requirements for the high scenario couldamount to 150 percent of current productionand, as data in table 63 indicate, even in the lowscenario, synfuels demand could exceed currentU.S. production for “fin type” heat exchangers.However, productive capacity can expand as rap-idly as machine operators and welders can betrained, which for an individual worker is meas-ured in terms of weeks and months. Additionalheat-treated steel and aluminum inputs will alsobe required, as well as manufacturing equipment,but in all cases supplies of these inputs shouldexpand with demand .27

This generally optimistic assessment does notmean that temporary shortages could not occurand temporarily drive up equipment prices ifprospects for synfuels deployment should im-

ZTcOrnpared With the full range of heat exchangers used in indus-

trial and utility applications, those likely to be used in synfuels plantswill operate at relatively low temperatures. Low-temperature unitsare made primarily out of carbon steel, low-alloy steel, and enamelsteel, all of which are readily available in commodity markets wheredemand for heat exchangers is a small fraction of the total. Hence,material inputs are unlikely to restrain expansion of heat exchangersupplies.

It is also unlikely that skilled labor or manufacturing plant andequipment will limit supplies, because the required welding andmachine operator skills can be learned in a period of weeks if nec-essary, and manufacturing facilities are not highly specialized. Back-ground information about the heat exchanger industry, and syn-fuels technology in particular, was obtained by private communica-tion with James Cronin, Manager of Projects, Air Preheater Divi-sion, Combustion Engineering, Wellsville, N.Y.

prove dramatically.28 However, as orders for newequipment skyrocket, new capacity should be-come available in time so that extremely highequipment prices can be avoided if project man-agers are willing to accept relatively brief (meas-ured in months) delays in delivery.

Skilled Mining and Construction Labor

Construction workers and their families canmove with employment opportunities, but mov-ing is costly and especially burdensome if jobsin an area last for only a period of months. Inorder to induce essential migration, synfuels proj-ects must incur high labor costs in the form oftravel and subsistence payments as well as

28A commonly cited example of a temporary inflationary spurt,

caused by a construction boom, occurred in the U.S. petrochemi-cals industry in 1973-75. Over the period from the mid-1 960’s tomid-1 970’s, the following three price indices show a distinctive pat-tern for chemical process equipment:

Chemical process All machineryYear equlpmenta Producer goodsb and Equipment1 9 6 7 . . . 100 100 1001 9 7 0 81 110 1111971 . 86 119 1181 9 7 2 74 135 1221 9 7 3 91 160 1391 9 7 4 139 175 1611975 . . . . 167 183 1711976 . . . . . . 188 194 1821977 ., . . . 154 209 206‘Data obtained from Annual Survey of Manufacturers, Bureau of Census, U S. Depan-ment of Commerce, SIC No. 35591 (xJ5, as reported In ASM-2.

bData obtalrlecj from U S, Statistical Abstract, 1979, PP 477-79.

In words, chemical process equipment prices reversed a decline

in 1973, increased by more than 150 percent through 1976, and

then tapered off again in 1977. This compares with a steady up-

ward trend from both producer goods and all machinery and equip-

m e n t .


Table 63.—Peak Requirements and PresentManufacturing Capacity for HeatExchangers (Million Square Feet)

Peakrequirementsfor 3 MMB/D

of coal liquids U.S. manufacturing(1985)” capacity

1. Process shellsand tubes. . . . . . . . 22.0 27

2. Fin type . . . . . . . . . 9.2 83. Condensers . . . . . . 4.4 15

36.8 50apeak requirements indicate maximum CapaCity requirements If synfuets Proj -

ects are to maintain production schedules.

SOURCE: Achieving a Production Goal of 1 Million B/Do fCoal Liquids by 19$X),draft prepared for the Department of Energy by TRW, Inc. and BechtetNational, Inc., March 1980, pp. 4-28.

“scheduled overtime. ” * However, while the in-flux of people and the relatively high paymentsto workers may cause severe local inflation, re-gional and national impacts should not be signifi-cant. Confidence in this conclusion is based pri-marily on the fact that training in constructionskills can be obtained in the period of weeks andmonths and that, if anything, there is an oversup-ply of people willing to enter these trades.29

Miners will be expected to move into a newarea and stay permanently. Although it wouldseem reasonable to suppose that workers would

—*Apparently, it is important for major employers to emphasize

that they do not pay premium wages and salaries for large construc-tion projects, but instead there are various special considerations.Whatever it is called, total worker remuneration appears to pro-vide an abnormally large incentive.

zgBechtel’s experience at nuclear powerplant sites in Michigan,

Pennsylvania, and Arizona has demonstrated that a person withlimited welding experience can be upgraded to “nuclear quality”in 6 to 12 weeks of intensive training. See Bechtel, “Productionof Synthetic Liquids, ” pp. 4-23. Actual training periods are influ-enced by various institutional factors. For further discussion of laborproductivity see K. C. Kusterer, Labor Productivity in Heavy Con-struction: Impact on Synfuels Program Employment, Argonne Na-tional Laboratory, ANL/AA-24, U.S. Department of Energy.

The supply of people willing to work on large construction proj-ects seems to be very price-elastic. In other words, large numbersof skilled or “able-and-willing-to-learn” workers will migrate to evenremote construction sites if wage incentives exceed going rates else-where in the Nation by 20-30 percent. Although it is difficult toconfirm this conclusion in published literature, it appears to be com-monly held among university-based experts as well as in the con-struction industry, Information was obtained from private com-munications with j. D. Borcherding, Department of Civil Engineer-ing, University of Texas in Austin; Richard Larew, Department ofCivil Engineering, Ohio State University; John Racz{ Synfuels Proj-ect Manager, Exxon USA in Houston; and Dan Mundy, BuildingConstruction Trades Department, AFL-CIO, in Washington, D.C.

be reluctant to mine underground, where work-ing conditions can be unpleasant and hazardous,historical experience suggests otherwise. In theEastern mines, with present wages about 140 per-cent of the national average in manufacturing,labor shortages have not been a serious prob-lem.30

Basic Construction Materials

Among all synfuels resources, basic construc-tion materials (primarily steel and concrete) areleast likely to cause serious bottlenecks. The morerapid the pace of deployment, the more likelya premium price must be paid for steel and ce-ment, but supplies of both should be highly re-sponsive to price incentives.

Mineral resources for the manufacture of Port-land cement (the class of hydrolic cement usedfor construction) are widely distributed across allregions of the Nation. The same is true for thesand and gravel that are mixed with cement andwater to make concrete. The only constraint onsupplies of cement or concrete is the time re-quired to construct new capacity, which takesat most 3 years for a new cement plant and muchless than that for a concrete mixing facility.31 Sincethese times are short relative to the constructionperiod for a synfuels project, cement shortagesshould not be a serious problem.

Steel supplies, on the other hand, may be in-sufficient in certain regions because required re-sources such as iron ore, scrap, and coking coalare not widely distributed. However, steel canbe shipped long distances without dramaticallyraising costs. For example, unfabricated structuralshapes and plates (e.g., 1 beams) are valued todayat approximately $25 per hundred pounds FOB

mAs prescribed in the new United Mine Workers/Bituminous OP-

erators Association contract, dated june 6, 1981, undergroundminers presently earn $10 to $11.76 per hour and surface miners$11.15 to $12.53. This compares with the national average wagein manufacturing of $7.80 and the average wage in constructionof $9.90, both calculated for March 1981. See Monthly LaborReview, May 1981, p. 84, for additional wage data. The generaliza-tion, that labor supply has not been a serious problem, is a con-clusion reached but stated only implicitly in an OTA report, TheDirect Use of Coa[ op. cit.

31 Information abut the resource base and construction leadtirnes

obtained by private communication with Richard Whitaker, Director

of Marketing and Economic Research, Portland Cement Associa-

tion, Skokie, III,


(freight on board at the factory) and they are com-monly shipped from Bethlehem, Pa., to Salt LakeCity, Utah, for another $4 per hundred pounds.In other words, even if local production is insuf-ficient to meet the needs of synfuels deployment,vast additional supplies from a national networkof suppliers can be shipped into the area withoutdriving up costs excessively .32

Final Comments

Despite OTA’s conclusions that resource short-ages other than engineering skills need not ob-struct synfuels deployment, it does not follow thatrapid synfuels deployment would not be inflation-ary for a broad range of resource inputs. Disre-garding the prospect of Federal intervention tospeed up deployment or to alleviate impacts, rap-id deployment could cause bursts of inflation inan economy where certain suppliers have domi-nant market positions at least within regions,where skilled workers are reasonably well orga-nized, and where people have grown accus-tomed to inflation. In such circumstances, itwould be surprising if those with power to negoti-ate their revenues and incomes did not exerciseit to their advantage when demand for their prod-uct and services is rapidly expanding.

Another caveat should also be made concern-ing the importation of processing equipment. Ifforeign suppliers compete successfully and be-come major suppliers of synfuels equipment, asthey have already demonstrated in the GreatPlains Gasification Project, rapid deploymentcould result in substantial foreign payments.33

Depending on the general balance of paymentspicture, this could devalue the dollar in foreign

JZData obtain~ from American Meta/Market, June 16, 1981, and

from Bethlehem Steel, Washington Office. It should be noted thatfabricated steel or steel which has been tailored to specific applica-tions can cost as much as $75 Wr hundred pounds and hence shipping costs may add much less to delivered costs (on a percentagebasis).

JJln this first major synfuels project, the japanese IOW bid was

substantially below apparent costs. Among other things, this in-dicates the competitive determination of at least one foreign sup-plier to capitalize on synfuels deployment. For related commentsby U.S. Steel firms, see Meta/s Dai/y, Sept. 4, 1980; and the ChicagoTribune, Aug. 30, 1980. For a general analysis of the U.S. steel in-dustry and its competition from abroad, see Technology and5tee//ndustry Competitiveness, OTA-M-122 (Washington, D. C.: U.S.Congress, Office of Technology Assessment, June 1980).

exchange markets and thus increase the price ofall imports into the United States. Perhaps offset-ting this concern about balance of payments, thesuccess of equipment imports may have a salutaryeffect on domestic producers by inducing themto improve their products and lower their costs.

Finance Capital and Inflation

In addition to potential shortages among re-source inputs, the deployment of synfuels capac-ity may be restrained by the limited availabilityof financial capital. Such a limit has already beenmentioned for small companies which cannotraise $2 billion to $3 billion and for any companytrying to borrow at presently inflated interestrates.

Limits may also be imposed by financial mar-kets that compare synfuels against all other typesof investments. If synfuels projects are indeed un-profitable, the number of projects funded maybe small or, if they are profitable, the number maybe large. In this sense, a market-based synfuelsdeployment scenario should be self-correcting,with the lure of profits attracting new investmentwhen expansion is warranted and the pain oflosses driving investors away and thus curtailingdeployment. Any of the previously discussedshortage possibilities, should they arise, will beperceived sooner or later by investors and thenumber of projects reduced as a result.

Whether or not deployment is by market incen-tives or Government policy, the adjustment andpossible disruption of financial markets requiredby synfuels deployment can be discussed in termsof gross investment data. Assume that on the av-erage, during its 5-year construction period, a$2.5-billion synfuels project requires $500 millionin outlays annually. This compares with total pri-vate domestic investment in 1980 of about $395billion, of which total about $294 billion went fornonfarm investments in new plant and equip-ment. Also in 1980, two large blocs of energy in-vestments were $34 bilion for oil and gas explora-tion and production and $35 billion for gas andelectric utilities.34

J4AII investment data except for oil and gas were obtained fromthe Survey of Current Business, Bureau of Economic Analysis, U.S.Department of Commerce, September 1981, pp. 9, S1. Oil and gas


In other words, 12 fossil synfuels plants underconstruction at the same time would account forabout 18 percent of the 1980 investments for theproduction of petroleum and natural gas, about17 percent of 1980 investments by electric util-ities, or about 5 percent of the total investmentin manufacturing. At this pace, assuming 5-yearconstruction periods, approximately 2 MMB/Dcapacity could be installed over the next 20 years(the low scenario). Almost three times this manyplants on the average must be under construc-tion at one time, and about three times as muchcapital must be committed to achieve the goalof just under 6 MM B/D by 2000 (high scenario).In either case, this average would be achievedby means of a relatively gradual startup, as tech-nologies are proven and experience is gained inconstruction, followed by a rapid buildup as allsystems become routine.

The question remains: Can funding be reason-ably expected for scenarios presented in thisreport? The answer depends on the future growthof GN P and the future value of liquid fuels relativeto other fuels and to all other commodities. With-out trying to predict the future, the question maybe partially answered by showing that such a di-version of funds to energy applications has prec-edents in recent history.

From 1970 to 1978, investments in oil and gasgrew at a rate of about 7.5 percent per year andinvestments in electric utilities grew at about 5percent per year, both in constant dollars.35 Aglance back at synfuels requirements as fractionsof existing energy investments shows that it wouldtake only about 2.5 years of 7.5 percent growthin oil and gas investments or about 3.5 years of5 percent growth in electric utility investmentsto provide sufficient incremental funds to supportthe low scenario, and about three times as manyyears of growth in each case to fund the high sce-nario.

investment data were obtained from Petro/eurn /rrc/ustry hwestrnentsin the 80’s, Chase Manhatten Bank, October 1981. The total of $34billion is broken down into $22 billion for service equipment, $6.3billion for lease bonuses, and $3.1 billion for geological andgeophysical data gathering.

35 Energy investment growth data obtained from 1978 AnnualReport to Congress, Energy Information Administration, p. 128.

In other words, another 5-year period of expan-sion in energy investments, similar to their growthin 1970-78 with oil and gas and electricity addedtogether, could provide more than enough fundsannually to reach the goal of about 6 MMB/D ofsynfuels by 2000 (high scenario), assuming thatthis higher level of investment were sustained forthe next 20 years. Furthermore, if such rapid de-ployment were economically justified (i.e., othercosts were rising sufficiently to make synfuels rel-atively low-cost options) there would be an eco-nomic incentive to divert funds to synfuels whichhad been devoted to conventional fuels.

Final Comments AboutInflation and Synfuels

In an inflating economy, all price incrementstend to be viewed as inflationary. However, thisappearance obscures the fact that some price in-creases are necessary adjustments in relativeprices in order to reduce consumption and to in-crease production. The latter will be true if syn-fuels place large, long-term, new demands onscarce human and material resources.

On the other hand, construction costs havegrown faster than the general rate of inflationsince the mid-1960’s.36 (See fig. 19.) Recently, thereverse has been true but there is reason to beconcerned that rapid synfuels deployment couldexacerbate what has been a serious inflationaryproblem. In any case, rising real costs of construc-tion has been one of the major reasons why “cur-rent” estimates of synfuels costs have more orless kept pace with rising oil prices. (See ch. 6for more detailed discussion.)

Finally, although most of this discussion has ex-plored how synfuels deployment may aggravateinflation, the cause and effect couId be reversedif deployment of first generation plants is tooslow. That is, if the promise of synfuels remains

JbcOrlSUrner price Index obtained from 1980 U.S. Statistic/

Abstract, p. 476. Construction Cost Index obtained from EngineeringNews Record, McGraw Hill, Dec. 4, 1981, Market Trends Section.The actual data series published in this journal has been convertedfrom a base year of 1916 to a base year of 1977. There are severalconstruction cost indices published by reputable sources, but onlythe ENR was reproduced here because the data available to OTAsuggest that all such series reflect more or less the same trends.

98-281 0 - 82 - 15 : c L 3


Figure 19.—Time Series Comparison: Construction in the distant future and conservation attemptsCosts and Consumer Prices

_ Construction cost index of Engineering I

News Record

--- Consumer Price Index

/

are clearly insufficient to balance oil supply anddemand worldwide, there will be no market-im-posed lid on the price of oil and no reason to ex-pect that sharp oil import price increases will notcontinue to destabilize domestic prices. In thatcase, the inflationary impacts of rapid deploymentmay appear to be much more acceptable.

SOURCE: U.S. Bureau of Labor Statistics, “Monthly Labor Review and Handbookof Labor Statistics,” annual, and “BM and ID Investment Manual,” /n-vestrnent Enghwedng, sec. 1, part 6, item 614, pg. 1, Apr. 16, 1961.

Chapter 9

Social Effects and Impacts

Page

Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 219

Social impacts of Changing Automotive Technology. . . . . . . . . . . . . . . . . . . . . . 219Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 219Employment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 220occupational and Regional Issues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 223

Social Impacts of Synfuels Development . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 225Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 225Manpower Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. ..:... 226Population Growth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 227private Sector Impacts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 230Public Sector Impacts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 231Managing Growth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 233

TABLES

Table No. Page

64. Initial and Lifecycle Costs of Representative Four-Passenger Electric Cars.. 22065. Auto Industry Employment Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22166. Factors to Consider in Locating Parts-Supplier Plants . . . . . . . . . . . . . . . . . . 22467. Manpower Requirements for a 50,000 bbl/d Generic Synfuels Plant . . . . . . 22768. Estimates of Regional Population Growth Associated With Fossil

Synfuels Development . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22869. Population Growth in SeIected Communities, 1970-80 . . . . . . . . . . . . . . . . . 22970. Statewide Population Estimates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 230

FIGURES

Figure No. Page

20. Auto, Steel and Tire Plant Changes, 1975-80 . . . . . . . . . . . . . . . . . . . . . . . . 22421. Census Regions and Geographic Divisions of the United States . . . . . . . . . 229

Chapter 9

Social Effects and Impacts

INTRODUCTION

Increased automotive fuel efficiency and pro- producing synthetic fuels will be felt primarily induction of synthetic fuels will both give rise to communities which experience rapid surges anda variety of social impacts. The impacts of increas- declines in population as plants are built anding fuel efficiency will occur primarily as changes begin to operate.in employment conditions, while the impacts of

SOCIAL IMPACTS OF CHANGINGAUTOMOTIVE TECHNOLOGY

Overview

The characteristics and uses of automobilessold in the United States indicate that historical-ly Americans have valued automobiles not onlyfor personal transportation but also as objects ofstyle, comfort, convenience, and power. Substan-tial increases in the costs of owning and operatingautomobiles that occurred during the 1970’s, andthat are expected in the future, are motivatingconsumers to change their attitudes and behaviorin order to reduce spending on personal transpor-tation. Some have purchased smaller, more fuel-efficient vehicles. Others have chosen to keeptheir present vehicles longer. Large numbers aresimply driving less. Since January 1979, the com-bined subcompact and compact share of totalsales has climbed from 44 to 61 percent, and gas-oline consumption has declined 12 percent.About one-half to three-quarters of these fuel sav-ings can be attributed to increased fuel efficien-cy of the automobile fleet.

Although about 12 percent of personal con-sumption expenditures has historically gone toautomobile ownership and operation, rising costsmay ultimately induce consumers to spend asmaller share of their budgets on automobiles or—at least—not to let that share increase. Recentincreases in the small-car proportion of new-carsales suggest that consumers are prepared totrade cargo space and towing capability for highfuel economy and the prospect of relatively lowoperating costs. in the future, instead of buyingvehicles designed for their most demanding trans-

portation needs, people may buy small vehiclesfor daily use and rent larger vehicles for infre-quent trips with several passengers, bulky orheavy cargo, or towing. The movement towardsmall cars is slowed by the tendency for peopleto retain cars longer than before. * Purchases offuel-efficient vehicles and ownership of severalvehicles, each suited for different transportationneeds, would be facilitated by improved econom-ic conditions.

Ridesharing and mass transit use have becomemore common and could increase further. Sincethe 1973-74 oil embargo public transit ridershiphas increased 25 percent.1 Ridesharing and transituse are limited by the dispersion of residencesand jobs, and, for transit, by the adequacy andavailability of facilities. Mass transit capacity islimited during peak commuting periods and oftenis unavailable or scheduled infrequently in areasoutside of central cities.

It should be noted that low-income people arelikely to have the fewest options for adjusting torising automobile costs. People with low incomesalready tend to own fewer vehicles, have relative-ly old vehicles (which were typically boughtused), travel less, and share rides or use publictransit more than the affluent.

Consumers are likely to respond differently toelectric vehicles (EVs) and small conventionally

*Thirty-five percent of private vehicles were over 5 years old in1969, 51 percent were over 5 years old in 1978.

‘American Public Transit Association.

219


powered cars (using internal combustion engines)because of different cost, range, and refuelingattributes (see table 64). The conventionallypowered car would have two significant advan-tages over an EV: unlimited range (with refuel-ing) and substantially lower first cost. The EV, onthe other hand, would offer the advantage ofbeing powered by a secure source of energy(electricity) and therefore assure mobility in theevent of disruption of gasoline supplies. It is notclear how the consumer would weigh these twooptions, although the degree to which EV man-ufacturers can reduce the cost differential is cer-tain to be very important.

Employment

In 1980, the Bureau of Labor Statistics estimatedthat there were fewer than 800,000 people em-ployed in primary automobile manufacturing andautomotive parts and accessories manufacturing.This compares with employment levels over900,000 during the peak automobile productionperiod, 1978-79.2 These figures, however, pre-sent an incomplete picture of employment. Al-though the Bureau of the Census counts employ-ees in industries producing various primary prod-

2Bureau of Labor Statistics, Current Employment Statistics Pro-gram data.

ucts, it does not identify how many workers con-tribute to intermediate products used in auto-mobiles or other finished goods. Thousands ofautomotive people perform work in support ofautomobile manufacturing within industriesotherwise classified—producing, for example,glass vehicular lighting, ignition systems, storagebatteries, and valves. Thus, the Department ofTransportation estimated that during 1978 to 1979about 1.4 million people were employed by autosuppliers overall.3

Historically, the growing but cyclical nature ofthe auto market resulted in a pattern of periodicgrowth and decline in auto-related employment(see table 65). Current and projected trends forstrong import sales, decline in the growth rateof the U.S. auto market, increased use of foreignsuppliers and production facilities, and adoptionof more capital-intensive production processesand more efficient management by auto manu-facturers and suppliers will contribute to a generaldecline in auto industry employment.

Specific changes in employment will dependon the number of plants closed or operating un-

3U. S. Department of Transportation, The U.S. Autorrrobi/e /ndus-try, 1980: Repori to the President from the Secretary of Transporta-tion (Washington, D.C: Department of Transportation, January1981 ).

Table 64.—initial and Lifecycle Costs of Representative Four-Passenger Electric Cars

Near term Advanced

Pb-Acid Ni-Fe Ni-Zn Zn-CL 2 (ICE) Zn-CL 2 Li-MS (ICE)

Initial cost, dollars . . . . . . . . . . . . . . . . . . . . . . . . . 8,520Vehicle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6,660Battery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1,860

Lifecycle cost, cents per mile . . . . . . . . . . . . . . . 23.9Vehicle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.0Battery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.0Repairs and maintenance . . . . . . . . . . . . . . . . . 1.5Replacement tires . . . . . . . . . . . . . . . . . . . . . . . 0.6Insurance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2Garaging, parking, tolls, etc.. . . . . . . . . . . . . . . 3.1Title, license, registration, etc.. . . . . . . . . . . . . 0.7Electricity. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3Fuel and oil . . . . . . . . . . . . . . . . . . . . . . . . . . . . . —Cost of capital . . . . . . . . . . . . . . . . . . . . . . . . . . 5.5

8,4005,9502,45024.9

4.54.81.50.52.23.10.62.2—5.5

8,1305,7202,41026.6

4.37.01.50.52.23.10.62.0—5.4

8,1205,5402,58022.0

4.22.31.50.52.23.10.62.2—5.4

4,7404,740

—21.4

4.3—3.90.42.23.10.6—4.03.0

7,0505,4101,64019.4

4.11.41.50.42.23.10.51.7—4.5

6,8105,1801,63020.1

3.92.61.50.42.23.10.51.5—4.4

5,1405,140

—21.8

4.7—3.90.42.23.10.5—3.73.3

NOTE: All costs are In mid-19S0 dollars. Annual travel 10,000 milesAssumptions: Car end battery salvage value 10 percentElectrlclty price S0.03 per kilowatt-hour Cost of capitalGasoline price $1.25 per gallon

10 percent per yearCar and battery purchases are 100 percent financed over their useful lives.

Electrlc vehicle life 12 years Electricity cost includes a road use tax equal to that paid by typical gasolineInternal combustion engines vehicles of equal weight vla State and Federal geso[lne taxes.

vehicle life 10 years

SOURCE: General Research Corp. Cost categories and many entries, such as tire% Insurance, 9araOin0, etc., are based on periodic cost analyses by the Departmentof Transportation (see footnote 13). All costs shown were computed by the Electric Vehicle Weight and Cost Model (EVWAC) (see footnote 14).

Ch. 9—Social Effects and Impacts . 221

Table 65.—Auto Industry Employment Data

(2)(1) Average annual

Average annual employment in primaryunemployment rate in auto manufacturing and

the motor vehicle parts and accessoriesindustry SIC 371 manufacturing, SIC

Year (percent) 3711 and SIC 3714 (000)

1970. . . . 7.0 733.41971 . . . . 5.1 781.31972 . . . . 4.4 798.21973. . . . 2.4 891.51974 . . . . 9.3 818.91975. . . . 16.0 727.81976. . . . 6.0 814.91977 . . . . 3.9 869.51978. . . . 4.1 921.71979. . . . 7.4 908.61980 . . . . 20.3 775.6SOURCE Column 1 data are from the Bureau of Labor Statistics, household sam-

ple survey Column 2 data are from the Bureau of Labor Statisticsestablishment survey, Data in the two columns are not directly com-parable. “SIC” refers to “Standard Industrial Classification.”

der capacity, the capacity of the plants, and thedegree to which production at affected plants islabor-intensive. The long-term effects on workersdepend on personal characteristics such as skills(many production workers have few transferableskills), the levels of local and national unemploy-ment, information about job opportunities, andpersonal mobility (greatest for the young, theskilled, and those with some money).

The Department of Transportation estimatesthat each unemployed autoworker costs Federaland State Governments almost $15,000 per yearin transfer payments and lost tax revenues. Thisestimate implies, for example, that if 100,000 to500,000 manufacturer and supplier workers areunemployed for a year their cost to governmentis about $1.5 billion to $7.5 billion. During 1980,payments to unemployed workers of GeneralMotors (GM), Ford, and Chrysler in Michigan in-cluded about $380 million in unemployment in-surance, $100 million in extended benefits, and$800 million in “trade adjustment assistance”(provided by a program established in the TradeExpansion Act of 1962 and modified by the TradeAct of 1974).4

Growing use of labor-saving machinery by automanufacturers and major suppliers to implement

4Michigan Employment Security Commission, personal communi-cation.

complex technologies, cut costs, and improveproduct quality is reducing job opportunities inthe auto industry. GM, for example, expects toinvest almost $1 billion by 1990 for 13,000 newrobots for automobile assembly and painting andparts handling. A new robotic clamping andwelding system developed by GM and RobogateSystems, Inc., will enable GM to reduce laborcosts for welding by about 70 percent, improvewelding consistency, and reduce vibration andrattling in finished automobiles. s

MacLennan and O’Donnell, analysts at DOT,have calculated that today’s new and refurbishedplants can assemble 70 cars/hour with an averageemployment level of 4,500, while older plantstypically produce 45 to 60 cars/hour using about5,400 workers. Such plant modernization impliesthat three fewer assembly plants and 23,000 few-er workers are needed to assemble 2 million carsannually. 6 The United Auto Workers estimatesthat labor requirements in auto assembly, whichhas been a relatively labor-intensive aspect ofauto manufacture, will be reduced by up to 50percent by 1990 through the use of robots andother forms of automation. ’

Foreign-designed automobiles manufactured inthe United States also provide jobs. Current andanticipated local production by foreign firms (onlyVolkswagen (VW) to date) largely involves vehi-cle assembly, using primarily imported compo-nents and parts. VW’s Pennsylvania plant em-ploys 7,500 workers to assemble over 200,000cars and contributes to about 15,000 domesticsupplier jobs; 8 a comparably sized domestic--owned plant would support a total of about35,000 domestic jobs. New U.S. manufacturingand supplier jobs will grow with local produc-tion and purchasing from U.S. suppliers by for-eign firms in proportion to the amount of localproduction content in the automobiles. Theplanned increase in local content for Rabbitsmade here by VW—from 70 percent in model

5“GM’s Ambitious Plans to Employ Robots, ” Business Week, Mar.

16, 1981.

‘Carol MacLennan and ]ohn O’ Donnell, “The Effects of the Auto-

motive Transition on Employment: A Plant and Community Study”(Washington, D.C: U.S. Department of Transportation, December1980).

7Business Week, op. cit.8Department of Transportation, op. cit.

year 1981 to 74 percent in model year 1983–implies more work in the United States.9

The Departments of Labor and Transportationestimate that there are between one and two sup-plier jobs overall for each primary auto manufac-turing job.10 Change in supplier employment as-sociated with declining manufacturing employ-ment is uncertain, and will depend on the natureof the supplied product, how it is made, and theamount that auto manufacturers buy. While somesupplier jobs, like auto manufacturing jobs, de-

9“VW Projects Increases in U.S. Content,” Ward’s AutomotiveReports, May 27, 1981.

IOMac Lennan and O’Donnell, Op. cit.

penal on production volume, other supplier jobs(e.g., in machine tool manufacture and plasticsprocessing) are tied to the implementation of newtechnology. Trends toward foreign sourcing andvehicle production and automation among sup-pliers suggest that supplier employment overallwill decline.

Steel and rubber industry jobs are especiallyvulnerable to automotive weight and volume re-ductions. Many of these supplier jobs have al-ready been lost with automotive weight reduc-tions during the 1970’s. For example, MacLen-nan and O’ Donnell estimate that reduced auto-motive use of iron and steel in 1975 to 1980 ledto a permanent loss of 20,000 jobs, a loss only

Ch. 9–Social Effects and Impacts Ž 223

partially offset by a gain of 8,000 jobs in process-ing plastics and aluminum for automotive use. 11

During the same period, employment in the tireand rubber industry declined at a compound an-nual rate of 4.1 percent. 12

Jobs with parts and component manufacturersare also relatively vulnerable, although, again,many have already been lost. Mac Lennan andO’Donnell estimate that the closing of almost 100materials, parts, and component plants in 1979to 1980 eliminated over 80,000 supplier jobs. 13

Because of the predominance of small firmsamong auto suppliers, near-term supplier joblosses may occur incrementally.

Automobile importation supports some domes-tic jobs and results in the loss of others. Thereare over 125,000 people employed by importers,primarily in dealerships. 14 Growth in import-re-lated employment stems from increases in thenumber and market shares of importers, in thenumber of dealerships per importer, and in em-ployment per dealership. On the other hand, im-ports cause loss of industrial jobs. DOT estimatesthat loss of 100,000 vehicle sales to imports resultsin the loss of about 8,500 primary manufactur-ing and 13,000 to 16,000 supplier jobs. ’ 5 This im-plies, for example, that the almost 400,000-unitincrease in import sales in 1978 to 1980 causeda loss of 34,000 jobs i n automobile manufactur-ing and up to 64,000 supplier jobs.

Employment in automotive services, includingrepair, parking, renting and leasing, washing, andother services (Standard Industrial Classification75) grew at a compound annual rate of 5.7 per-cent in 1975 to 1980 to a total level of almost540,000 people, according to the Department ofCommerce.16 Employment in these areas is ex-pected to continue to grow.

I 1 Ibid.

1 ZU ,S. Depaflment of commerce , 1981 U.S. /ftdu5tria/ OUt/00~

(Washington, D.C: Department of Commerce, 1981).1 jMac Len nan and O’ Donnell, Op. cit.

14 Patricia Hinsberg, “Study Finds Imports Create U.S. jobs,” Automotive News, Aug. 20, 1979.

15Depaflment of Transportation, oP. cit.IGDepaflment of COmmerce, oP. cit.

Occupational and Regional Issues

Improvements in automotive technology causechanges in the skills required for production jobs.Major design and technology changes increasedemands for engineers, who have been in shortsupply, while cost-cutting strategies eliminateother white-collar positions. GM, for example,eliminated about 10,000 white-collar jobs begin-ning in 1980 to save about $300 million, and mayeliminate up to 20,000 more. ’ 7 Automation re-duces the number of routine and hazardoustasks, while increasing equipment maintenanceand service tasks. GM, for example, plans to haveequal numbers of skilled and unskilled workersby the 1990’s, although it presently has oneskilled worker for each five to six unskilledworkers.18

Auto production jobs are concentrated inMichigan, Ohio, Indiana, New York, and Illinois(see fig. 20). The geographic distribution of auto-related jobs is likely to change somewhat for sev-eral reasons. First, some nontraditional auto sup-pliers are located away from traditional areas ofauto production. Major plastics-producing States,for example, include California, New Jersey, andTexas as well as Ohio and Illinois. Furthermore,many of today’s suppliers are located abroad andmany U.S. suppliers are opening plants abroad.Second, foreign or domestic firms may establishproduction facilities outside of the East-NorthCentral area to gain lower labor and utility costs,For example, Nissan chose to build a plant in Ten-nessee. Third, domestic firms are closing ineffi-cient and unneeded plants. Table 66 summarizesthe factors considered in locating parts-supplierplants.

Automotive plant closings primarily affect em-ployment in the East-North Central region, be-cause automobile production is concentratedthere. Ongoing and future losses of automobile-related employment in this region are largely areflection of the structural changes in the autoindustry described earlier in this chapter, al-though there will continue to be cyclical changes

17”GM May Chop Another 19,000 Salaried jobs, ” AutomotiveNews, Mar. 2, 1981.

lsWa// Street Journa/, January 1981.


Figure 20.–Auto, Steel, and Tire Plant Changes, 1975=80

SOURCE: U.S. Department of Transportation, “The U.S. Automobile industry, 1980,” DOT-P-1 O-81-O2, p. xviii, January 1981.

Table 66.—Factors to Consider inLocating Parts-Supplier Plants

Relative ranking

Factor 1970’s 1980’s

Availability and cost of skilled labor . . . .Availability and cost of energy . . . . . . . . . 2;State and local taxes, incentives . . . . . . . 3Availability and cost of raw materials . . . 2Work ethic of area . . . . . . . . . . . . . . . . . . . 4State and local permits, regulations . . . . 5Worker’s compensation insurance . . . . . . 7Availability and cost of capital . . . . . . . . . 7Right-to-work state (union relations) . . . . 6Freight costs . . . . . . . . . . . . . . . . . . . . . . . . 10Quality of living environment . . . . . . . . . . 8Community attitude ., . . . . . . . . . . . . . . . . 9Customer service . . . . . . . . . . . . . . . . . . . . 12Available land . . . . . . . . . . . . . . . . . . . . . . . 11

123456789

1011111112

SOURCE: Arthur Andersen & Co. and the Michigan Manufacturer Association,Woddwlde Corrrpetltlveness of the U.S. Automotive Industry and ItsParts Suppliers During the 1930s, Februa~ 1981.

in automobile-related employment. Because oflocal and regional employment dependence onone industry—motor vehicles—other businesses(such as retail and service establishments) andtheir employment also suffer as employment inthe local population declines.

Unemployment and out-migration will jeopard-ize other businesses and strain local tax bases,and Michigan will be especially vulnerable. Lossof employment, population, and business recent-ly induced Moody’s Investers Service, Inc., tolower bond ratings for Akron, Ohio, which hasdepended on the tire and rubber industry for itseconomic vitality; bond ratings for other auto-dependent cities have also been lowered.

Ch. 9—Social Effects and Impacts ● 225

Photo credit: Michigan Employment Security Commission

Shifts in plant location associated with investments in fuel efficiency may create unemployment problems intraditional manufacturing centers

SOCIAL IMPACTS OF SYNFUELS DEVELOPMENT

Overview

The principal social consequences of develop-ing a synthetic fuels industry arise from largeand rapid population increases and fluctuationscaused by the changing needs of industry for em-ployees during a facility’s useful life. Such popula-tion changes disproportionately affect small, ruralcommunities that have limited capacity to absorband manage the scale of growth involved; thesetypes of communities characterize the locationswhere oil shale and many coal deposits arefound.

In general, whether the consequences ofgrowth from synfuels development will be

beneficial or adverse will depend on the abilityof communities to manage the stresses which ac-company rapid change. Although impacts can begenerally characterized, the extent and nature oftheir occurrence will be site-specific dependingon both community factors (size, location, taxbase) and technology-related factors (the loca-tion, size, number, and type of synfuels facilities;the rate and timing of development; and associ-ated labor requirements).

Growth will tend to concentrate in establishedcommunities where services are already availa-ble, if they are within commuting distance to syn-fuels facilities. New towns may be established to


accommodate growth in some areas. Large townswill serve as regional service centers. Isolatedcommunities will more likely experience greaterimpacts than areas where well-linked communi-ties can share the population influx. Energy con-version facilities which are sited near mines willresult in the greatest concentration of local im-pacts.

Most synfuels production from oil shale in theNation will be concentrated in four Westerncounties, affecting about a dozen communitiesin sparsely populated areas of northwestern Col-orado and northeastern Utah, and eventuallysouthwestern Wyoming. These communities arewidely separated, are connected by a skeletaltransportation network, and have had historicallysmall populations. Oil shale cannot be economic-ally transported offsite because of the large quan-tities of shale involved per barrel of product.

Coal presents a more flexible set of options thanoil shale with respect to the location of conver-sion facilities in relation to mines. The coal usedfor synfuels production will most likely be dis-persed among all the Nation’s major coal regions.

In the West, coal sites will be in the oil shaleStates (Colorado, Utah, and Wyoming) as wellas in Montana, North Dakota, and New Mexico.Most of the increase in Western coal productionfor synfuels will be in Wyoming and Montana.19

Midwestern sites will most likely be in Illinois,western Kentucky, and Indiana. The coal coun-ties to be most severely affected in Appalachiawill be in rural parts of southwestern Pennsyl-vania, southern West Virginia, and eastern Ken-tucky. Parts of Illinois will also be affected. In cen-tral Appalachia, communities are typically small,congested, and in rural mountain valleys.

The major differences between the Eastern andWestern coalfields, in general, are that in theWest, counties are larger, towns are smaller andmore scattered, the economic base is more diver-sified, more land is under Federal jurisdiction,water is relatively more scarce, and the terrainis less rugged and variable. To the extent that coal

lgE. j. Bentz & Associates, Inc., “Selected Technical and EconomicComparisons of Synfuel Options,” contractor report to OTA, April1981.

is transported, there could be additional environ-mental and safety hazards, noise, and disruptionor fragmentation of communities, farms, andranch lands.

The social consequences of producing synthet-ic fuels from biomass are discussed in detail ina previous OTA report, Energy From BiologicalProcesses. Unlike the social consequences associ-ated with fossil fuels, the social impacts of bio-mass arise from production rather than process-ing. For example, 90 percent of the employmentimpacts from biomass are expected from cultiva-tion and harvesting (mostly forestry).

Manpower Requirements

Manpower requirements for synfuels produc-tion are generally of two types: 1) labor is re-quired for the construction of energy facilities andsupporting service infrastructures, and 2) workersare needed for the operation and maintenanceof facilities. As discussed in chapter 8, the abilityto attract and retain an adequate labor force—particularly experienced chemical engineers andskilled craftsmen, who are already in short supply—could become a constraint on synfuels develop-ment.

Construction manpower requirements for sin-gle projects lead to large, rapid, yet temporary,increases in the local population. The construc-tion phase usually lasts 4 to 6 years, peaking overa 2- to 4-year period as construction activitiesnear completion.20 The shorter the scheduledconstruction period, the higher the peak laborforce.21 Labor requirements will change signifi-cantly during the construction phase, in terms ofboth size and occupational mix. Labor require-ments for the daily, routine operation and mainte-nance of a plant are relatively stable during theuseful life of the facility; scheduled yearly and ma-jor maintenance work would cause only brief in-creases in the operations labor force.

Estimates of manpower requirements for gener-ic 50,000 barrel per day (bbl/d) synthetic fuel

Zolbid.21 peter D. Miller, “stability, Diversity, and Equity: A Comparison

of Coal, Oil Shale, and Synfuels,” in Supporting Paper 5: Socio-political Effects of Energy Use and Policy, CONAES, Washington,D. C., 1979.

Ch. 9—Social Effects and Impacts • 227

plants are shown in table 67. They are highly un-certain, in large part because of the lack of expe-rience with commercial-size plants. In addition,major components of uncertainty in the construc-tion manpower estimates include such unpredict-able situations as regulatory delays, lawsuits, de-lays in the receipt of materials, labor unrest, andthe weather; and major components of uncertain-ty in the estimates of operations manpower relateto the age of the plant, maturity of the technol-ogy, and novelty of the plant design.

Even for well-known technologies such as coal-fueled electric powerplants, initial estimates ofthe peak construction labor force required for se-lected rural-sited plants have varied from about50 to 270 percent of the actual peak levels.22 Thisrange of uncertainty may be applicable to the esti-mates of construction manpower requirementsfor synfuels plants in general, but should proveto be overly broad when considering a specifictechnology. The uncertainty surrounding require-ments for operational manpower is expected tobe narrower, perhaps on the order of + 25 per-cent.23

The estimates shown in table 67 are plant-gateemployment requirements; other synfuels-relatedactivities such as mining, beneficiation, and trans-portation are not included unless indicated. Themanpower requirements for these additional ac-tivities will be site-specific and could alter the rel-———

Zzjohn S. Gilmore, “Socioeconomic Impact Management: AreImpact Assessments Good Enough to Help?” paper presented atthe Conference on Computer ModeIs and Forecasting Impacts ofGrowth and Development, University of Alberta, jasper Park Lodge,Alberta, Apr. 21, 1980 (revised June 1980).

2 3 G e o r g e Wang, Bechtel Group, Inc., PfXSOnal Communication.

ative ordering of alternative technologies. For ex-ample, on the national average, production perminer per day is approximately three times great-er in surface mines than in underground mines.This ratio can be expected to vary, dependingon many factors including types of methods andequipment used and geology for undergroundmining, and geology and environmental consid-erations for surface mining.24

Population Growth

Local population will grow where synfuels areproduced because workers directly employed atthe synfuels plants, employees in secondary in-dustries and services, and accompanying familieswill move into these areas. Population growthrates will depend on the nature of the area wherethe plant is located and on the phase of plant de-velopment.

Estimates of population growth due to synfuelsdevelopment usually assume that for each newworker entering an area, the population increasesbetween three and five persons.25 The demand

zdThe average national production per miner per day was 8.38

tons in underground mines and 25.78 tons in surface mines forbituminous and lignite in 1978. Nationwide, productivity varied:for underground mining, from approximately 2 to 15 average tonsper miner per day and, for surface mining, from approximately 7to 98 average tons per miner per day. (Depatiment of Energy, EnergyInformation Administration, Bituminous Coa/ and Lignite Produc-tion and Mine Operations— 1978, Energy Data RepoR, June 16,1980.)

25AS an example, White, et al., use a population/employment mul-tiplier of 3.0 for the construction phase and 4.0 for the operationphase (Energy From the West, Science and Public Policy Program,University of Oklahoma, prepared for the Environmental Protec-tion Agency, March 1979). Miller uses a uniform “conservative”popuIation/employ merit multiplier of 5.o (see footnote 21 above).

Table 67.—Manpower Requirements for a 50,000 bbl/d Generic Synfuels Plant

Liquefaction

Direct Indirect Coal gases Oil shale

Total construction (person-years). . . . . . . . . . 11,000 20,000 11,000 11,000 a

Peak construction (persons) . . . . . . . . . . . . . . 3,500 6,800 b 3,800 3,500 a

Operations and maintenance (persons) . . . . . 360C 360C 360C

2,300 d 3,800 d 1 ,200 d 2,000 e

asurf~e reto~lng WiII generally have higher construction manpower requirements than modified in-Situ Processes.bM anpower requirements of 17,000 persons have been projected by Fluor Corp. based on a SASOL tYPe coal conversion Plant(“A Fluor Perspective on Synthetic Liquids: Their Potential and Problems”).

cDaily, routine O&M requlrments (E. J. Bentz & Associates, Inc., “Selected Technical and Economic Comparisons of SynfuelsOptions,” April 19S1).

dAnnual aggregation of scheduled yearly and major maintenance work. Technology SpeCifiC (Bechtel NationaJ, inc., “production

of Synthetic Liquids From Coal: 19S0-2000,” December 1979).coil shale requirements include mining, Modified in.situ (MIS) processes will generally have higher O&M requiret?leflk than

suriace retorting (e.g., MIS Involves an ongoing mining process).



for support services in nearby communities in-creases with the absolute size of the work forceduring the peak construction period. The largerthe work force required during peak construc-tion relative to that required for operations andmaintenance, the greater the likelihood that near-by communities will experience large populationfluctuations.

In general, if several facilities are located in thesame area, the impacts from population growthand fluctuation could be disproportionately largeunless construction and operation activities arecoordinated; on the other hand, populationgrowth associated with construction can be stabi-lized if an indigenous construction work forcedevelops. *

Estimates of population increases associatedwith the fossil synfuels development scenariospresented in chapter 6 are shown in table 68. Ona regional basis, population growth associatedwith oil shale will be concentrated in only severalcounties in the Mountain Region (see fig. 21).population increases associated with coal-basedsynfuels will be dispersed throughout the Nation,with the East North Central experiencing the big-gest population increases and the West SouthCentral experiencing the smallest population in-creases.

Table 69 shows energy-related populationgrowth during the last decade in selected com-munities. In small communities, and in sparselypopulated counties and States, energy-relatedpopulation growth could represent a significantproportion of future population growth. For ex-ample, official population projections by the Col-orado West Area Council of Governments(CWACOG) show increases by 1985, relative to1977, of up to 400 percent in Rio Blanco County(1977 special census population was 5,100) and300 percent in Garfield County (1 977 special cen-sus population was 18,800), assuming the indus-try develops according to the 1979 plans of com-panies active in the area.

*A succession of projects in an area should lead to an indigenousand more stable construction manpower work force, dependingon whether workers perceive a permanence of industrial expan-sion in the area. Some proportion of the construction work forcemay also be employed in operations and maintenance activitiesonce construction is completed.

Table 68.—Estimates of Regional PopulationGrowth Associated With Fossil Synfuels

Development a (thousands)

1990 1995 2000

Low estimate:South Atlantic . . . . . . . . . . . 4-6 12-20 30-51East North Central . . . . . . . 14-24 46-76 115-192East South Central . . . . . . . 5-9 17-28 42-71West North Central . . . . . . . 5-9 17-28 42-71West South Central. . . . . . . 2-3 6-10 15-25Mountain:

Coal . . . . . . . . . . . . . . . . . . 7-12 23-38 58-96Shale . . . . . . . . . . . . . . . . . 66-110 90-150 81-135

Total . . . . . . . . . . . . . . . . . 103-173 211-350 383-641

High estimate:South Atlantic.. . . . . . . . . . . 11-18 33-55 86-144East North Central . . . . . . . 33-56 105-174 275-458East South Central . . . . . . . 11-18 33-55 86-144West North Central . . . . . . . 17-29 54-90 141-235West South Central. . . . . . . 5-8 15-25 39-65Mountain:

Coal . . . . . . . . . . . . . . . . . . 19-32 60-100 122-203Shale . . . . . . . . . . . . . . . . . 132-220 213-355 108-180

Total . . . . . . . . . . . . . . . . . 228-381 513-854 857-1,429aEstimate are relative to 1985 (for plants coming online in the Year shown) and

are based on OTA’s development scenarios presented in ch. 8. Populationmultipliers of 3 and 5 were applied to develop the ranges shown. Aggregatedestimates should not be extrapolated to determine the ability of any State orlocality to absorb this population.

bproduction iS distributed among the regions, according to the low and highscenario distributions used In the Bechtel report for, respectively, the low andhigh scenarios developed herein (Bechtel National, Inc., December 1979). It isfurther assumed that direct and indirect iiquids wiil be representad equally. Onlydally, routine O&M requirements are included.Regional estimates are for coal processes unless other wise indicated.


Under CWACOG’s high-growth scenario(500,000 bbl/d in 1990 and 750,000 bbl/d in 1995and 2000), increases of up to 800 percent in RioBlanco County and 350 percent in Garfield Coun-ty are projected.26 In three counties in Kentuckywhere the construction of four major synfuelsplants had recently been planned to commence(H-Coal, SRC-1, W. R. Grace, and Texas Eastern),the expected maximum number of synfuels work-ers (excluding accompanying family members)was projected to increase 1980 population levelsby about 3 percent in Daviess County (1 980 cen-sus population was 86,000) to over 30 percentin Breckinridge County (1980 census populationwas 17,000) and over 50 percent in HendersonCounty (1980 census population was 41 ,000);population increases during the operation phase

ZbAn Assessment of oil Shale Technologies, OTA-M-1 18

(Washington, D. C.: U.S. Congress, Office of Technology Assess-ment, June 1980).

Ch. 9—Social Effects and Impacts . 229

Figure 21 .—Census Regions and Geographic Divisions of the United States

- ..

0

H

SOURCE: U.S. Department of Commerce, Bureau of the Census.

Table 69.–Population Growth in Selected Communities, 1970-80

Percent increasePopulation b

Average annualState City Energy resource impact a 1970 1980 Total (compounded)

West Virginia

KentuckyUtah

Wyoming

North Dakota

Montana

Colorado

Buckhannon, Upshur County(Union District)

Caseyville, Union CountyHuntington, Emergy CountyOrangeville, Emery CountyHelper, Carbon CountyDouglas, Converse CountyGillette, Campbell CountyRocksprings, Sweetwater County

Washburn, McLean CountyBeulah, Mercer CountyForsyth, Rosebud CountyHardin, Big Horn CountyColstrip, Rosebud County c

Craig, Moffat CountyRifle, Garfield CountyHayden, Routt County

coal

coalcoal, powerplantcoal, powerplant

coalcoal, uranium, oil, gas

coalcoal, oil, gas, trona,powerplant, uranium

coal, powerplantcoal, powerplantcoal, powerplant

coalcoal, powerplantcoal, powerplant

oil shale, minerals, coalcoal, powerplant

248 587 136.7

87.0857 2,316 170.2511 1,309 156.2

1,964 2,724 38.72,677 6,030 125.37,194 12,134 68.7

11,657 19,458 66.9

8041,3441,8732,733<2004,2052,150

763

1,767 119.82,878 114.12,553 36.33,300 20.73,500 1,650.08,133 93.43,215 49.51,720 125.4

9.0

6.510.59.93.38.55.45.3

8.27.93.21.9

33.16.84.18.5

aldentified by the Department of community Development within the respective States.bBureau of the Census, 19s0 Cansus of Population and Housing, Advance RepOffs.CEStirnates by sunllght Development, Inc.


230 ● Increased Automobile Fuel Efficiency and Synthetic Fuel Alternatives for Reducing Oil Imports

were projected to be respectively 0.4, 4, and 15percent (excluding accompanying family mem-bers). 27 Table 70 shows statewide population esti-mates, based on an extrapolation of only demo-graphic trends, for some of the States that aremost likely to experience population increasesfrom synfuels development.

Small rural communities (under 10,000 resi-dents) that experience high population growthrates are vulnerable to institutional breakdowns.Such breakdowns could occur in the labor mar-ket, housing market, local business activities, pub-lic services, and systems for planning and financ-ing public facilities. Symptoms of social stress(such as crime, divorce, child abuse, alcoholism,and suicide) can be expected to increase.

The term “modern boomtown” has been usedto describe communities that experience strainson their social and institutional structure fromsudden increases and fluctuations in the popula-tion. Communities are also concerned about thepossibility of a subsequent “bust.” Large fluctua-tions in population size could lead to a situationwhere a community expands services at onepoint in time only to have such services under-utilized in the future if demands fail to materializeor be sustained.

ZZC. Gilmore Dutton, “Synfuel Plants and Local Government FiscalIssues, ” memorandum to the Interim Joint Committee on Appropria-tions and Revenue, Frankfort, Ky., Dec. 18, 1980.

Private Sector Impacts

The principal social gains from synfuels devel-opment in the private sector are increased wagesand profits; direct and secondary employmentopportunities will be created and expanded; dis-posable income will increase; profits from energyinvestments should be realized; and local tradeand service sectors will be stimulated. The abil-ity of the private sector to absorb growth will de-pend, in large part, on the degree of economicdiversification already present. Western commu-nities, in general, have more diversified econo-mies and broader service bases than those in theEast, where many communities (as in central Ap-palachia) have historically been economicallydependent on coal.

Many private sector benefits, however, will notbe distributed to local communities, at least dur-ing the early periods of rapid growth. For exam-ple, synfuels development would be located inareas where the required manpower skills arealready scarce; unemployment in local communi-ties may thus not be significantly lowered unlesslocal populations can be suitably trained. Wheresynfuels development competes with other sec-tors for scarce labor, fuel and material inputs, andcapital resources, traditional activities may be cur-tailed and the price of the scarce resources in-flated. Local retail trade and service industriesmay experience difficulties in providing and ex-panding services to keep pace with demands, re-

Table 70.—Statewide Population Estimates

Total State Statewide population percent Projected State

population 1980 a increase 1970-80 population 2000 b

State (millions) Total Annual compounded 1990 2000Kentucky . . . . . . . . . 3.66 13.7 1.3 4.08 4.43West Virginia. . . . . . 1.95 11.8 1.1 2.08 2.20Colorado . . . . . . . . . 2.89 30.7 2.7 3.50 4.00Montana. . . . . . . . . . 0.79 13.3 1.3 0.90 0.98North Dakota. . . . . . 0.65 5.6 0.5 0.70 0.73Wyoming . . . . . . . . . 0.47 41.6 3.5 0.54 0.60Utah . . . . . . . . . . . . . 1.46 37.9 3.3 1,73 1.95aeureau of the Census, 1980 census of Population and Housing, Advanced Reports.beureau of the census, ///ustrat/ve Projections of State Populations by Age, Race, and Sex: 1975 fO ~. Projected estimates

are from Series II-B which assumes a continuation from 1975 through 2000 of the civilian non-college interstate migrationpatterns by age, race, and sex observed for the 1970-75 period. Has been corrected by the percent difference between the1980 projections and the 19S0 census. Note that these projections are extensions of recent trends with respect to demographicfactors only.



cruiting and retaining employees, and competingwith out-of-State concerns. Both the Eastern andwestern sites for synfuels development have gen-erally depended on imported capital, and prof-its to and reinvestments of the energy companiesare likely to be distributed to locations remotefrom plant sites.

High local inflation rates often accompany rap-id growth, due to both excess demand for goodsand services and high industrial wage rates. Localinflation penalizes those whose wages are inde-pendent of energy development and those onfixed incomes.28

Housing can be a major problem for the privatesector in areas that grow rapidly from synfuelsdevelopment: the existing housing stock is usuallyalready in short supply and often of poor qual-ity; local builders often lack the experience andcapability to undertake projects of the large scalerequired; shortages of construction financing andmortgage money are common; and land may notbe available for new construction because of ter-rain, land prices, or overall patterns of owner-ship. Housing shortages have already led to dra-matic price increases in the Western oil shaleareas. The need for temporary housing for con-struction workers aggravates housing supplyproblems, and mobile homes are often used byboth temporary and permanent workers.

Public Sector Impacts

Communities experiencing rapid growth arevulnerable to the overloading of public facilitiesand services, due both to large front-end capitalcosts and to constraints which limit a communi-ty’s ability to make the necessary investments ina timely fashion. * Ability of a community to ab-sorb and provide for a growing population willbe community-specific and depend on many fac-tors—such as the size of the predevelopment taxbase, availability of developable land, existing

ZoAn Assessment of Oil Shale Technologies, Op. Cit.

*investments in the public sector can be constrained by, as ex-amples, existing tax bases, debt limitations, bonding capacity, andthe 2- to 5-year Ieadtime typically required for planning and imple-menting services. In addition, ceilings are often established on therate of expansion of local government budgets, and there is a tend-ency either not to tax or to undervalue undeveloped mineral wealth.

social and institutional structure, extent and rateof growth of demand for public services, localplanning capabilities and management skills, andpolitical attitudes.

In the long run, local governments should ben-efit from expanded tax bases arising from the cap-ital intensity of energy facilities and the establish-ment of associated economic activity. In the ag-gregate, sufficient additional tax revenue shouldbe produced to pay for the upgrading and expan-sion of public facilities and services as requiredfor the growing population.29

In the short run,however, raising local revenues under conditionsof rapid growth is often made difficult becauseof the unequal distribution of incurred costs andrevenue-generating capacity among different lev-els of government.

For example, energy development activities aretypically sited outside municipal boundaries, withthe result that revenues go to the county, schooldistrict, and/or State. However, the populationgrowth accompanying this energy development,and hence the need for services, typically occurswithin cities and towns that do not receive addi-tional revenues from the new industry. The sepa-ration between taxing authority and public serv-ice responsibility can also occur across State lines.In addition, the availability of local tax revenuescan lag behind the need for services, because in-dustrial taxes are often based on assessed prop-erty values and are not received until full plantoperation. * Note also that the total tax burdenon the mineral industry and the proportion ofState taxes distributed to localities vary from Stateto State.

There is no clear consensus on the cost of pro-viding additional new public facilities and servicesin communities affected by energy development.The economics of the decision to expand froman existing service base, or to build a new town,will depend on such factors as the availability ofland, accessibility to employment, extent and

ZgManagement of Fue/ and Nonfuel Minerals in Federal Lands,

OTA-M-88 (Washington, D. C.: U.S. Congress, Office of TechrrcdogyAssessment, April 1979).

*Note that mobile homes generate little, if any, property taxes;and local governments have had difficulty in providing services tosuch sites.

98-281 0 - 82 - 16 : QL 3


Synfuels development will require the creation of new communities in sparsely populated areas


recovery or other funding/revenue mechanismshave been applied.

Health care is particularly vulnerable to over-loading from rapid growth because rural commu-nities often have inadequate health services priorto development and experience difficulties in at-tracting and retaining physicians. Synfuels devel-opment will change the health care needs of localcommunities because of the influx of young fam-ilies, the increase in sources of social stress, andnew occupational environments that will give riseto special health care needs. Hospital facilitiesas well as health, mental health, and social serv-ices will be required. Educational facilities are alsolikely to be overloaded. Both Eastern coal com-munities and Western oil shale communities arepresently having difficulty in attracting and retain-ing personnel and in funding the provision andexpansion of facilities and programs,

Public sector dislocations caused by synfuelsdevelopment on Indian lands could be more se-vere than on other rural areas. Tribes have limitedability to generate revenues, there will be largecultural differences between tribal members andworkers who immigrate to an area to work ona project, and land has religious significance tosome tribes and individual landownership is com-monly prohibited (so that, for example, conven-tional patterns of housing development may notbe possible).

Most reservations are also sparsely populated,with few towns, and public services and facilitiesare severely inadequate and overburdened. Sig-nificant amounts of coal are owned by Indiansin New Mexico and Arizona, lesser amounts inMontana, North and South Dakota, and Colo-

rado. Although in the aggregate current coalleases represent only a fraction of the total coalunder lease, Indian leases are important becauseof their size and coherence. About 8 percent ofthe oil-shale mineral rights in the Uinta Basin areowned by Indians, but most of the associateddeposits are of low grade.32

Managing Growth

Unmanaged growth, although not well under-stood, appears nevertheless to be a leadingsource of conflict and stress associated with ener-gy development. All involved parties-the Feder-al, State, and local governments; industry; andthe public—have an interest in and are contribut-ing in varying degrees to growth management byproviding planning, technical, and financial assist-ance to communities experiencing the effects ofsynfuels development. These mechanisms, whichvary among States in terms of their scope, detail,and development, are discussed in detail in previ-ous OTA reports.33 3435 I n general, the effective-ness of existing mechanisms has yet to be testedin the face of rapid and sustained industrial ex-pansion. Major issues to be resolved include whowill bear the costs of and responsibilities for bothanticipating and managing social impacts, andhow up-front capital will be made available whenneeded to finance public services.

32u .s. Geological survey, Synthetic Fuels Development, Earth-

Science Considerations, 1979.JJAn Assessment of Oil Shale Technologies, Op. cit.

JaManagernent of Fue/ and Nonfuel Minerals in Federal Lands,

op. cit.M The Direct Use of coal: Prospects and Problems of Production

and Combustion, OTA-E-86 (Washington, D. C.: U.S. Congress, Of-fice of Technology Assessment, April 1979).

Chapter 10

Environment, Health, and SafetyEffects and Impacts

Contents

Page

Introduction . . . . . . . . . . . . . . . . . . . . . . . . 237

Auto Fuel Conservation. . . . . . . . . . . . . . . 237Motor Vehicle Safety . . . . . . . . . . . . . . . 237Mining and Processing New Materials . 243Air Quality . . . . . . . . . . . . . . . . . . . . . . . 244

Electric Vehicles. . . . . . . . . . . . . . . . . . . . . 245Power Generation . . . . . . . . . . . . . . . . . 245Resource Requirements . . . . . . . . . . . . . 246Noise . . . . . . . . . . . . . . . . . . . . . . . . . . . . 247Safety . . . . . . . . . . . . . . . . . . . . . . . . . . . . 247Occupational and Public Concerns . . . 247

Synthetic Fuels From Coal . . . . . . . . . . . . 248Mining . . . . . . . . . . . . . . . . . . . . . . . . . . . 250Liquefaction . . . . . . . . . . . . . . . . . . . . . . 253Conventional Impacts . . . . . . . . . . . . . . . 253Environmental Management . . . . . . . . . 261Transport and Use . . . . . . . . . . . . . . . . . 264

Oil Shale . . . . . . . . . . . . . . . . . . . . . . . . . . . 266

Biomass Fuels . . . . . . . . . . . . . . . . . . . . . . . 268Obtaining the Resource . . . . . . . . . . . . . 268Conversion . . . . . . . . . . . . . . . . . . . . . . . 269

Appendix 10A. —Detailed Descriptionsof Waste Streams, Residuals ofConcern, and Proposed ControlSystems for Generic indirect andDirect Coal Liquefaction Systems . . . 271

TABLES

Table No. Page

71 . Material Substitutions for VehicleWeight Reductions . . . . . . . . . . . . . . . 243

72.

73.

74.

75.

76.

77.

78.

79.

Emissions Characteristics ofPage

Alternative Engines . . . . . . . . . . . . . . . 245Increase in U.S. Use of Key Materialsfor 20 Percent Electrification ofLight-Duty Travel . . . . . . . . . . . . . . . . . 247Major Environmental Issues forCoal Synfuels . . . . . . . . . . . . . . . . . . . . 249Fatality and Injury Occurrencefor Selected Industries, 1979 . . . . . . . 252Two Comparisons of theEnvironmental impacts of Coal-BasedSynfuels Production and Coal-FiredElectric Generation . . . . . . . . . . . . . . . 257Potential Occupational Exposuresin Coal Gasification . . . . . . . . . . . . . . . 258Occupational Health Effects ofConstituents of indirect LiquefactionProcess Streams . . . . . . . . . . . . . . . . . . 259Reported Known Differences inChemical, Combustion, and HealthEffects Characteristics of SynfuelsProducts and Their PetroleumAnalogs . . . . . . . . . . . . . . . . . . . . . . . . . 265

FIGURES

Figure No. Page

22.

23.

Passenger-Car Occupant Deathper 10,000 Registered Cars by CarSize and Crash Type; Cars 1 to 5Years Old in Calendar Year 1980 . . . 238Size Ranges of New Coal-FiredPowerplants With Hourly EmissionsEqual to 50,000 bbl/day SynfuelsPlants . . . . . . . . . . . . . . . . . . . . . . . . . . 254

Chapter 10

Environment, Health, and Safety Effects and Impacts

INTRODUCTION

There are major differences in the risks to pub-lic health and the environment associated withthe alternative approaches to reducing the de-pendence of the U.S. transportation sector on for-eign oil. Depending on the level of development,the production and use of synthetic fuels implymassive increases in mining (and agriculture andforestry for biomass), construction and operationof large conversion plants producing substantialquantities of waste products (some of which aretoxic), and fuel products that may be differentfrom the fuels now in commerce and that maythus represent different risks in handling and use.

Electrification of autos would require large in-creases in electric power production, which inturn imply major increases in powerplant fuel useand emissions. Also, the use of electric cars woulddecrease the use of conventional vehicles andthus yield reductions in vehicular emissions aswell as changes in vehicle materials and operatingcharacteristics.

Increased automotive fuel efficiency would in-volve changes in vehicle size, materials, operatingcharacteristics, and emissions. All the strategieswould reduce the use of petroleum that wouldotherwise have been imported, and adverse ef-fects associated with the strategies should be par-

tially offset by the resulting environmental benefitof reductions of oil spills and other hazards.

This section identifies potential effects on theenvironment and human health of these three al-ternative (or complementary) approaches to re-ducing or eliminating oil imports. Because of sig-nificant uncertainties in the precise characteristicsof the technologies to be deployed, their poten-tial emissions and the control levels possible, andfuture environmental regulations and other im-portant predictive factors, the approach of thisevaluation is relatively informal and qualitative.We attempt to put the alternatives into reason-able perspective by identifying both a range ofpotential effects and, given the availability of con-trols and incentives to use them, the most likelyenvironmental problems of deployment. The ma-jor emphasis in the discussion of synthetic fuelsis on coal-based technologies. OTA has recentlypublished reports on biomass energy1 and oilshale,2 both of which contain environmentalassessments.

‘Energy From Biological Processes, OTA-E-124 (Washington, D. C.:U.S, Congress, Office of Technology Assessment, July 1980).

2An Assessment of Oil Shale Technologies, OTA-M- 118(Washington, D. C.: U.S. Congress, Office of Technology Assess-ment, June 1980).

AUTO FUEL CONSERVATION

Some measures taken to improve the fuel econ-omy of light-duty vehicles might have significanteffects on automobile safety and the environ-ment. Major potential effects include changes invehicle crashworthiness due to downsizing andweight reduction, environmental effects fromchanges in materials and consequent changes inmining and processing, and possible air-qualityeffects from the use of substitutes for the spark-ignition engine,

Motor Vehicle Safety

The shift to smaller, lighter, more fuel-efficientcars has led to heightened concern about a possi-ble increase in traffic injuries and fatalities. Partof this concern stems from evidence that occu-pants of smaller cars have been injured and killedat rates considerably higher than the rates asso-ciated with larger cars. The National HighwayTraffic Safety Administration (N HTSA) has recent-

237


Iy estimated that a continuing shift to smallervehicles could result in an additional 10,000 traf-fic deaths per year (with total annual road fatal-ities of 70,000) by 1990 unless compensatingmeasures are taken.

In light of these concerns, OTA examined avail-able evidence on the relationship between vehi-cle size and occupant safety in today’s auto fleet,and reviewed some attempts—including theNHTSA estimate—to extrapolate this evidence toa future, downsized fleet.

Occupant Safety and VehicleSize in Today% Fleet

Much of the current concern about the safetyof small cars is based on statistical analysis of na-tional data from the Fatal Accident Reporting Sys-tem (FARS), which contains information on fatalmotor vehicle accidents occurring in the UnitedStates. For example, an analysis of FARS data onautomobile occupant deaths conducted by theInsurance Institute for Highway Safety (IIHS) (fig.22) shows that deaths per registered vehicle in-crease substantially as vehicle size (measured bylength of wheelbase) decreases.4 Furthermore,this trend occurs for both single- and multiple-

3National Highway Traffic Safety Administration, Traft7c SafetyTrends and Forecasts, DOT-HS-805-998, October 1981.

Zlnsurance Institute for Highway Safety, Status Report, vol. 17,No. 1, Jan. 5, 1982.

Photo credit: National Highway Traffic Safety Administration

Crash tests sponsored by the National Highway Traffic SafetyAdministration are an important source of information for

understanding the mechanics of crashes andevaluating auto safety features

vehicle crashes. The trend is so strong that theannual occupant deaths per registered small sub-compact are more than twice as high as the ratefor full-size cars–3.5 per 10,000 cars comparedwith 1.6 per 10,000.

The relationships illustrated in figure 22 temptone to conclude that small cars are much less safethan large cars in virtually all situations. For a vari-ety of reasons, however, the information in thefigure must be interpreted with care. First, the re-cent crash tests sponsored by NHTSA5 (new carswere crashed head-on into a fixed barrier at 35miles per hour) seemed to indicate that the differ-

5The crash tests are described in several references. A useful, clearreference is “Which Cars Do Best in Crashes?” in the Apdl 1981issue of Consumer Reports. Also see M. Brownlee, et al., “implica-tions of the New Car Assessment Program for Small Car Safety, ”in proceedings of the Eighth International Technical Conferenceon Experimental Safety Vehicles, Wolfsburg, Germany, Oct. 21-24, 1980, NHTSA report.

Figure 22.—Passenger-Car Occupant Death per10,000 Registered Cars by Car Size and Crash Type:

Cars 1 to 5 Years Old in Calendar Year 1980-

2

1

0Subcompact Intermediate

Small Compact Full-sizesubcompact

SOURCE: Insurance Institute for Highway Safety.

Ch. 10—Environment, Health, and Safety Effects and Impacts ● 239

ences in expected occupant injuries between ve-hicles in the same size class–i.e., differencescaused by factors other than size—can be greaterthan any differences between the size classes. im-portantly, the results imply that relatively minorchanges in engineering and design, such as inex-pensive improvements in the steering columnand changes in the seatbelt mechanisms, can pro-duce improvements in vehicle crashworthinessthat may overwhelm some of the differencescaused by size alone. The results of the tests canbe applied only to occupants wearing seatbelts(11 percent of total occupants), however, andonly to new cars in collisions with fixed objects.

Another reason to be cautious is that the IIHSanalysis may be overlooking the effect of variablesother than car size. For example, the age of driv-ers and occupants is a critical determinant of fatal-ity rates. Younger drivers tend to get into moreserious accidents,7 and younger occupants areless likely than older ones to be killed or seriouslyinjured in otherwise identical crashes.8 Becausethe average age of drivers and occupants is notuniform across car size classes—it is believed thatsmaller cars tend to have younger drivers andoccupants—the observed differences in fatalityrates may be functions not only of the physicalcharacteristics of the cars but also of differencesin the people in those cars.

Other variables that should be considered ininterpreting injury and fatality statistics includesafety belt usage (drivers of subcompact cars havebeen reported to use seatbelts at a significantlyhigher rate than drivers of intermediate and largecars9), the average number of occupants per car,and differences in maneuverability and brakingcapacity (i. e., crash avoidance capability) be-tween big and small cars. *

‘R. H. Stephenson and M. M. Finkelstein, “U.S. GovernmentStatus Report,” in proceedings of the Eighth International TechnicalConference on Experimental Safety Vehicles, op. cit.

7H. M. Bunch, “smaller Cars and Safety: The Effect of Downsiz-

ing on Crash Fatalities in 1995, ” HSRI Research Review (Universityof Michigan), vol. 9, No. 3, November-December 1978.

elbid.%tephenson and Finkelstein, op. cit.*The effect of improved crash avoidance capability and other

safety factors may be perverse. To the extent that drivers may takemore chances in reaction to their perception of increased safety,they can negate the effectiveness of safety improvements. The tend-ency of drivers of large cars to use seatbelts at a lower rate thandrivers of small cars may be an indicator of such a reaction.

Several analyses have tried to account for theeffect of some of these variables.10 However,these analyses use different data bases (e.g., Statedata such as that available from North Carolina,and other national data bases such as the Nation-al Accident Sampling System and the NationalCrash Severity Study), different measures of vehi-cle size (wheelbase, the Environmental Protec-tion Agency (EPA) interior volume, weight, etc.),different formulations of safety (e.g., deaths per100,000 registered vehicles, deaths per vehicle-mile driven, deaths per crash), and in additiontheir data reflect different time frames. Few anal-yses correct for the same variables. Consequently,it is extremely difficult to compare these analysesand draw general conclusions.

Also, credible data on total accident rates forall classes of cars, and more detailed data on ac-cident severity, are not widely available. This typeof data would allow researchers to distinguish be-tween the effects of differences in crashworthi-ness and differences in accident avoidance capa-bility in causing the variations in fatalities meas-ured in the FARS data base. For example, studiesof accident rates in North Carolina indicate thatsubcompacts are involved in many more acci-dents than large cars.11 Consequently, the rela-tionship between fatalities per registered vehicleand car size, and that between fatalities per crashand car size could be significantly different forthis data set, with the latter relationship indicatingless dependence between safety and vehicle sizethan appears to be the case in the former. Unfor-tunately, such data are available only in a fewjurisdictions and cannot be used to draw nation-wide conclusions.

Finally, the existing data base reflects only cur-rent experience with small cars. In particular, thedata reflect no experience with the class of ex-tremely small sub-subcompacts that currently aresold in Japan and Europe but not in the UnitedStates. it is conceivable that widespread introduc-tion of such cars into the U.S. fleet, triggered by

ICIA variety of these are described in j. R. Stewart and J. C. Stutts,

“A Categorical Analysis of the Relationship Between Vehicle Weightand Driver Injury in Automobile Accidents, ” NHTSA reportDOT-HS-4-O0897, May 1978.

llj. R. Stewart and C. L. Carroll, “Annual Mileage Comparisonsand Accident and Injury Rates by Make, Model, ” University ofNorth Carolina Highway Safety Research Center.


their lower sales prices or by renewed oil priceincreases, could have severe safety conse-quences. NHTSA engineers are concerned thatoccupants of sub-subcompacts might be endan-gered not only by the increased decelerationforces that are the inevitable danger to the smallervehicle in multicar crashes, but also by problemsof managing crash forces and maintaining passen-ger-compartment integrity that are encounteredin designing and building cars this small.12

Despite these problems, some conclusionsabout the relationship between vehicle size andsafety can be drawn. For example, the strong pos-itive relationship between vehicle size and safe-ty in a//accidents combined and in car-to-car col-lisions has been confirmed in virtually all analy-ses.13 However, the size/safety relationship doesnot appear to be as “robust” for single-car colli-sions, which accounted for about half of all pas-senger-car occupant fatalities in 1980. Althoughseveral studies conclude that there is a strong pos-itive relationship between car size and safety inthis class of accidents,14 l Q and the IIHS analysesshow a very strong relationship,15 some studieshave concluded that this positive relationship dis-appears among some size classes when the dataare corrected for driver age and other variables.16

However, even these studies show that subcom-pacts fare worse than all other size classes insingle-vehicle accidents. ”

Forecasting Future Trends in Auto Safety

Attempts to forecast the effects on traffic safe-ty of a smaller, more fuel-efficient fleet—a resultof further downsizing within each size class aswell as a continued market shift to smaller sizeclasses—are confronted with severe analytical dif-ficulties. First, if the forecast is to account for theeffects of important vehicle and driver-related

“J. Kanianthra, Integrated Vehicle Research Division, NHTSA,personal communication, March 1982.

llstewa~ and Stutts, OP. cit.IAFor example, several studies cited in Stewart and St@tS, op, cit.;

also, j. H. Engel, Chief, Math Analysis Division, NHTSA, “An investi-gation of Possible Incompatibility Between Highway and VehicleSafety Standards Using Accident Data,” staff report, April 1981; also,J. O’Day, University of Michigan Highway Safety Research Institute,personal communication, March 1982.

‘S IIHS, op. cit.lbstewart and Stutts, Op. cit.

17 bid.

variables, the forecasters must predict how thesevariables will change in the future—e.g., for eachsize class, forecasters must predict future valuesof average driver age, vehicle miles driven, occu-pancy rates, seatbelt usage, etc. And they musteither estimate future size dimensions in each carclass and the number of vehicles in each classin the fleet, or else postulate these values. Sec-ond, forecasters must construct a credible mod-el that describes the relationship between trafficsafety (e.g., injury and fatality rates) and key vehi-cle and driver-related variables in such a way thatthe model will remain valid over the time periodof the forecast.

The models used by NHTSA18 and others19 toproject future safety trends generally use simplestatistical representations of the relative risk of ac-cidents or injuries and fatalities. The traffic fatalityprojections examined by OTA all relied on acci-dent data that included older design automobileseven though few such vehicles are likely to re-main in the fleet when the date of the projectionarrives.

In particular, NHTSA’s widely disseminated es-timate of 10,000 additional annual traffic deathsby 199020 assumed that exposure to fatality riskis a function only of vehicle weight and the num-ber of registered vehicles in each weight class.No account is taken of the effect of recent vehi-cle design changes, age and behavior of drivers,differences in crash avoidance capabilities, differ-ences in annual vehicle-miles driven and vehi-cle occupancy rate between various automobilesize classes, and other variables. Similar short-comings exist in the other projections. The result-ing projections of future changes in traffic injuriesand fatalities should be considered as only rough,first-order estimates.

—ILWHEA, op. cit. The model briefly described in this report ap-

pears to be similar to the forecasting model used in j. N. Kanianthraand W. A. Boehly, “Safety Consequences of the Current Trendsin the U.S. Vehicle Population, ” in proceedings of the Eighth inter-national Technical Conference on Experimental Safety Vehicles,op.cit.

19W. Dreyer, et al., “Handling, Braking, and Crash Compatibil-ity Aspects of Small Front-Wheel Drive Vehicles, ” Society of Auto-motive Engineers Technical Paper Series 810792, ]une 1981. Also,Bunch, op. cit. Also, j. Hedlund, “Small Cars and Fatalities–Com-ments on Volkswagen’s SAE Paper, ” internal NHTSA memoran-dum, Feb. 4, 1982.

20 NHTSA, op. cit.

Ch. 10—Enviromnent, Health, and Safety Effects and Impacts • 241

Because of the weaknesses in available quanti-tative projections of future fatality rates, OTA ex-amined current injury/fatality data and othersources for further evidence of whether or r-totdownsizing and a mix shift to smaller size classeswould have a significant effect on safety. In par-ticular, the following observations are importantto answering this question:

1.

2.

A safety differential between occupants ofsmall and large cars in multiple-car collisionsdoes not necessarily imply that reducing thesize of all cars will result in more deaths inthis class of accidents. Although availabledata clearly imply that reducing a vehicle’ssize will tend to increase the vulnerabilityof that vehicle’s occupants in a car-to-carcollision, the size reduction also will makethe vehicle less dangerous to the vehicle itcollides with. Under some formulations ofaccident exposure and fatality risk, these twofactors may cancel each other out. For ex-ample, Volkswagen has calculated the effectof increasing the proportion of subcompactsin today’s fleet. Using FARS data and fore-casting assumptions that are well within theplausible range, Volkswagen concluded thatan increase in subcompacts would actuallylead to a decrease in traffic fatalities in car-to-car collisions.21 Other models using differ-ent formulations and data bases might cometo different conclusions. For example, mod-els using traffic safety data from NorthCarolina probably would arrive at a differentresult. In this historical data set, subcompactscolliding with subcompacts have been foundto have a considerably greater probability ofcausing a fatality than collisions between twofull-size cars.22 Presumably, models usingthis data set would be likely to forecast thata trend toward more subcompacts wouldlead to an increase in car-to-car crash fatal-ities.If small cars are less safe than large cars insingle-car accidents, then a decrease in theaverage size of cars in the fleet with no com-

21 Dreyer, et al., Op. Cit.2*K. Digges, “Panel Member Statement,” Panel on ESV/RSV Pro-

gram, in proceedings of the Eighth International Technical Confer-ence on Experimental Safety Vehicles, op. cit.

pensatory improvements in crashworthinessclearly should imply an increase in injuriesand fatalities in this class of accidents. As justdiscussed, some studies suggest that a con-sistent relationship between size and safetydoes not exist for compact, midsize, and full-size cars in single-car accidents.23 On theother hand, subcompacts do fare worse thanthe other classes in these studies.24 Conse-quently, if these studies are correct, a generaldownsizing of the fleet might have only asmall effect on fatalities in single-car acci-dents, while a drastic shift to very small carscould cause a large increase in such fatalities.

The results of these studies may not bewidely applicable. Other studies observe adefinite size/safety relationship across all sizeclasses. 25 And some factors tend to favor thisalternative conclusion. For example, thehigher seatbelt usage in smaller cars shouldtend to make small cars appear safer in theraw injury data, and thus tend to hide orweaken a positive size/safety relationship.Taking differences in seatbelt usage into ac-count might expose or strengthen such a re-lationship.26 Also, analysis of FARS data thatincludes only vehicles up to 5 years old pro-duces a stronger size/safety relationship thananalysis of the whole fleet.27 Most studies usethe whole fleet, but the more limited dataset might prove to be better for a projectionof the future because it reflects only newer-design automobiles. Finally, as discussed inchapter 5, the larger crush space and passen-ger compartment volume available to thelarger cars should give them, at least theoret-ically, a strong advantage in the great major-ity of accidents. On the other hand, an op-posing factor favoring those studies show-ing less dependence between vehicle sizeand crashworthiness is the limited evidenceof increasing accident rates with decreasingcar sizes. 28 This offers a reason other than

ZJStewa~ and Stutts, op. cit.Wbid.25 Supra 14,ZGstephenSOn arid Finkelstein, Op. Clt,27 Based on a comparison of II HS’S analysis, op. cit., and Engel’s

analysis, op. cit.za’jtewa~ and Carroll, op. `cit.


3.

4.

(or in addition to) differences in crashworthi-ness for the differences in fatalities amongthe various auto size classes.Although most arguments about downsizingand traffic safety have focused on vehicle oc-cupants, the inclusion of pedestrian fatalitieswill affect the overall argument. About 8,000pedestrians were killed by motor vehicles in1980,29 and analysis of FARS data indicatesthat pedestrian fatalities per 100,000 regis-tered cars increase as car size increases30—i.e., reducing the average size of cars in thefleet might decrease pedestrian fatalities be-cause of the reduced “aggressiveness” ofsmaller cars towards pedestrians. If policyconcern is for total fatalities, this effectshould lessen any overall adverse safety ef-fect of downsizing the fleet.Much of the available data implies that traf-fic fatalities will rise if the number of colli-sions between vehicles of greatly differentweights increases. This points to three dan-gers from a downsized fleet. First, for a Iim-ited period of time, the number of collisionsof this sort might increase because of thelarge number of older, full-size cars left inthe fleet. This problem should disappearwithin a decade or two when the great ma-jority of these older cars will have beenscrapped. Second, a more permanent in-crease in fatalities could occur if large num-bers of very small sub-subcompacts–carsnot currently sold in the U.S. market—wereadded to the passenger vehicles fleet. Thepotential for successful large-scale sales ofsuch vehicles will depend on their prices—they may be significantly less expensive thancurrent subcompacts—as well as future oilprices and public perceptions of gasolineavailability. Third, car-truck collisions, whichtoday represent a significant fraction of occu-pant fatalities (car-to-other-vehicle accidentsaccount for about 25 percent of total occu-pant fatalities31), may cause more fatalitiesunless the truck fleet is downsized as well.Subcompacts fare particularly poorly in car-

29 NWSA, Fatal Accident Reporting System 1980.JoBased on an analysis of data presented in Engel, OP. cit.31 NHTSA, op. cit.

truck collisions, and a large increase in thenumber of vehicles in this size class couldcreate substantial problems.

The available statistical and physical evidenceon auto safety suggest that a marked decrease inthe average vehicle size in the automobile fleetmay have as a plausible outcome an increase invehicle-occupant fatalities of a few thousand peryear or more. This outcome seems especially like-ly during the period when many older, heaviervehicles are still on the road. Also, such an out-come seems more likely if the reduction in aver-age size comes mainly from a large increase inthe number of very small cars in the fleet, ratherthan from a more general downsizing across thevarious size categories in the fleet.

The evidence is sufficiently ambiguous,however, to leave open the possibility that onlya minor effect might occur. And, as discussed inthe next section, improvements in the safetydesign of new small vehicles (possibly excludingvery small sub-subcompacts) probably couldcompensate for some or all of the adverse safetyeffect associated with smaller size alone. Someautomobile analysts feel that significant safety im-provements are virtually inevitable, even withoutadditional Government pressures. For example,representatives of Japanese automobile com-panies have stated32 that the present poor recordof Japanese cars in comparison with Americansmall cars is unacceptable and will not be al-lowed to continue. Major improvements in Japa-nese auto safety would seem likely to force aresponse from the American companies. Also,General Motors has begun to advertise the safe-ty differentials between its cars and Japanesemodels, an indication that American manufac-turers may have decided that safety can sell. Onthe other hand, because of its severe financial dif-ficulties, the industry may be reluctant to pursuesafety improvements that involve considerablecapital expenditures.

Safer Design

Increases in traffic injuries and fatalities neednot occur as the vehicle fleet is made smaller in

JzRepOrted in the April 1981 Consumer Reporfs, OP. cit.

IJlbid., and IIHS, op. cit.

Ch. 10—Enviromnent, Health, and Safety Effects and Impacts Ž 243

size. Numerous design opportunities exist to im-prove vehicle safety, and some relatively simplemeasures could go a long way towards compen-sating for adverse effects of downsizing and shiftsto smaller size classes.

Increased use of occupant restraint systemswould substantially reduce injuries and fatalities.NHTSA analysis indicates that the use of air bagsand automatic belts could reduce the risk of mod-erate and serious injuries and fatalities by about30 to 50 percent. 34

Simple design changes in vehicles may substan-tially improve occupant protection. As noted inevaluations of NHTSA crash tests, design changesthat are essentially cost-free (changing the loca-tion of restraint system attachment) or extreme-ly low cost (steering column improvements to fa-cilitate collapse, seatbelt retractor modificationsto prevent excessive forward movement) 35 appearto be capable of radically decreasing the crashforces on passenger-car occupants.

A variety of further design modifications to im-prove vehicle safety are available. As demon-strated in the NHTSA tests,36 there are substan-tial safety differences among existing cars of equalweight. One important feature of the safer cars,for example, is above-average length of exteriorstructure to provide crush space. Also, the Re-search Safety Vehicle Program sponsored by theDepartment of Transportation shows that smallvehicles with safety features such as air bags, spe-cial energy-absorbing structural members, anti lac-eration windshields, improved bumpers, doorsdesigned to stay shut in accidents, and other fea-tures can provide crash protection considerablysuperior to that provided by much larger cars.

Two forms of new automotive technology in-troduced for reasons of fuel economy could alsohave important effects on vehicle safety. First, theincorporation of new lightweight, high-strengthmaterials may offer the automobile designer newpossibilities for increasing the crashworthiness of

MR . ). Hitchcock and C. E. Nash, “Protection of Children andAdults in Crashes of Cars With Automatic Restraints, “ in Eighth inter-national Technical Conference on Experimental Safety Vehicles,op. cit.

jSBrownlee, et al., Op. cit.

3blbid.

the vehicle. Because some of the plastics andcomposite materials currently have problems re-sisting certain kinds of transient stresses, however,their use conceivably could degrade vehicle safe-ty unless safety remains a primary considerationin the design process. Second, the use of elec-tronic microprocessors and sensors, which is ex-pected to become universal by 1985 to 1990 tocontrol engine operation and related drivetrainfunctions, could eventually lead to safety devicesdesigned to avoid collisions or to augment driverperformance in hazardous situations.

Modifications to roadways can also play a sig-nificant role in improving the safety of smaller ve-hicles. For example, concrete barriers and road-way posts and lamps designed to protect largervehicles have proven to be hazards to subcom-pacts37 in single-vehicle crashes, and redesign andreplacement of this equipment could lower futureinjury and fatality rates.

Mining and Processing New Materials

Aside from the beneficial effects of downsizingon the environmental impacts of mining—by re-ducing the volume of material required–vehicledesigners will use new materials to reduce weightor to increase vehicle safety. Table 71 shows fourcandidates for increased structural use in automo-biles and the amount of weight saved for every100 lb of steel being replaced.

It appears unlikely that widespread use of thesematerials would lead to severe adverse impacts.Magnesium, for example, is obtained mostly fromseawater, and the process probably has fewerpollution problems than an equivalent amountof iron and steel processing, Most new aluminum

3711HS, op. cit.

Table 71 .—Material Substitutionsfor Vehicle Weight Reductions

Weight saved/100 lbStructural material steel replaced

Magnesium . . . . . . . . . . . . . . . . . . . . . . 75Fiberglass-reinforced composites. . . 35-50Aluminum . . . . . . . . . . . . . . . . . . . . . . . 50-60High-strength low-alloy steel . . . . . . . 15-30

SOURCE: M. C. Flemming and G. B. Kenney, “Materials Substitution andDevelopment for the Light-Weight, Energy-Efficient Automobile,” OTAcontractor report, February 1980


probably would be obtained by importing bauxiteore or even processed aluminum, rather than ex-panding domestic production. if kaolin-type claysare used for domestic production, waste disposalproblems could be significant; however, the costof producing aluminum from this source currentlyis too high to make it economically worthwhile.

Use of high-strength low-alloy steel will likelylead to slightly lowered iron and steel produc-tion because of the higher strength of this materi-al, with a positive environmental benefit. Final-ly, the use of plastics and reinforced compositeswould substitute petrochemical-type processingfor iron and steel manufacture, with an uncer-tain environmental tradeoff.

Air Quality

Regulation of automobile emissions under theClean Air Act of 1970 (Public Law 91-614) andsubsequent amendments has sharply reduced theamount of pollutants from automobile exhaustin the atmosphere. Assuming that present stand-ards and proposed reductions in permissible lev-els of hydrocarbons (HC), carbon monoxide(CO), and nitrogen oxides (NOX are met, the ag-gregate of automobile emissions by 1985 will beroughly half of what they were in 1975 despitean increase of 25 percent in the number of carson the road and a corresponding rise in totalmiles of vehicle travel. * By 2000, if the 1985standards have been maintained and compliedwith, the aggregate of automobile emissions ofHC, CO, and NOX will be 33, 32, and 63 percentof today’s levels, respectively. Particulate emis-sions would be about one-half of today’s levels—and possibly much lower, depending on theprogress in control of particulate emissions fromdiesels.

The reductions expected by 1985 will havebeen brought about by a combination of two ba-sic forms of emission control technology—meth-ods of limiting the formation of pollutants throughcontrol of fuel-air mixture, spark timing, and other

*These projections, based on an earlier study by OTA38 have beenadjusted to account for more recent data on automobile use andthe lower projected growth rates used in this study.

38changeS in the Future Use and Characteristics of the Automobile

Transpofiation System, OTA-T-83 (Washington, D. C.: U.S. Congress,Office of Technology Assessment, February 1979).

conditions of combustion in the engine, and sys-tems to remove pollutants from the exhaust be-fore it is discharged into the atmosphere. The ef-fectiveness of both techniques has been greatlyenhanced by the advent of electronic engine con-trols in recent years.

By 1985, when electronic engine controls willbe virtually universal in passenger cars, the man-dated levels of 3.4 grams per mile (gpm) CO, 0.41gpm HC, and 1.0 gpm NOX can probably be metby spark-ignition engines with little or no penaltyin fuel economy beyond that associated with thelower engine compression ratios dictated by (low-octane) lead-free gasoline. * And although thisfuel penalty may be charged to the control of CO,HC, and NOX emissions because lead-free gaso-line is required to protect catalytic converters, thereduction in lead additives to gasoline may alsobe justified on the basis of its beneficial effect inreducing lead emissions and, consequently, thelevel of lead in human tissue. Assuming that re-ducing lead in gasoline is desirable even withoutthe catalytic converter requirement, the much-argued tradeoff between fuel economy and emis-sions that seemed so compelling in the 1970’s isunlikely to remain a major issue with the spark-ignition engine by the last half of the 1980’s.

A shift to still smaller vehicles and the introduc-tion of new engines (and substantial increases inthe use of current diesel technology) may affectthe tradeoff between air quality and control costs.Because lower vehicle weights and lesser perfor-mance requirements will allow substantiallysmaller engines, the grams per mile emissionstandards should be easier to meet for mostengine types. And, although manufacturers canbe expected to respond to this opportunity bycutting back on emission controls, there will bean enhanced potential for eventually loweringemissions still further. On the other hand, someof the engines—e.g., the gas turbines and diesels—may pose some control problems, with NOX

and particulate especially.

*Although high-octane lead-free gasoline can be, and is, manu-factured, the fuel savings it might allow from higher compressionengines may be counterbalanced by additional energy required forrefining. The exact energy required to produce higher octane lead-free gasoline will be very specific to the refinery, feedstock, andrefinery volumes.


Table 72 briefly describes some of the emissionscharacteristic of current and new engines forlight-duty vehicles. The potential emission prob-lem with diesels appears to be the major short-term problem facing auto manufacturers todayin meeting vehicle emission standards. There ap-pears to be substantial doubt that diesels cancomply with both NO X and particulates standardswithout some technological breakthrough, be-cause NO X control, already a problem in diesels,conflicts with particulate control. This problemis especially significant because diesel particulateare small enough to be inhaled into the lungs andcontain quantities of potentially harmful organiccompounds.

The effect on human health of a substantial in-crease in diesel particulate emissions is uncertain,because clear epidemiologic evidence of adverseeffects does not exist and because there is doubtabout the extent to which the harmful organicsin the emissions will become biologically avail-able—i.e., free to act on human tissue *—afterinhalation.39 However, a sharp increase in thenumber of diesel automobiles to perhaps 25 per-

*Initially, the organics adhere to particulate matter in the exhaust.In order for them to be harmful, they must first be freed from thismatter. In tissue tests outside the human body (“in vitro” tests),they were not freed, i.e., they did not become biologically active.This may be a poor indicator of their activity inside the body,however.

JgHealth Effects Panel of the Diesel Impacts Study Committee

(H. E. Griffin, et al.), National Research Council, /-/ea/th .Effects ofExposure to Diesel Exhaust, National Academy of Sciences, Wash-ington, D. C., 1980.

Table 72.—Emissions Characteristicsof Alternative Engines

Current spark ignition. — Meets currently defined 1983 stand-ards.

Current (indirect injection) diesel.—Can meet CO and HCstandards, but NO X remains a problem. NO X control con-flicts with HC and particulate control. Future particulatestandards could be a severe problem.

Direct-injection diesel. —Meets strictest standards proposedfor HC and CO. NO X limit 1 to 2 g/mile depending on vehi-cle and engine size. Possible future problems with particu-Iates, odor, and perhaps other currently unregulatedemissions.

Direct-injection stratified-charge.— Needs conventionalspark-ignition engine emission control technology to meetstrict HC/CO/NO x standards. Better NO X control thandiesel. In some versions particulate likely to be problem.

Gas turbine-free shaft.—Attainment of 0.4 g/mile NO X limita continuing problem, appears solvable, maybe withvariable geometry. Other emissions (HC, CO) no problem,

Sing/e shaft. —Same basic characteristics as comparable freeshaft. Better fuel economy may help lower NO X emissions.

Sing/e shaft (advanced). – N OX emissions aggravatedbecause of higher operating temperatures.

Stirling engine (first generation).—Early designs have hadsome NO X problems, but should meet tightest proposedstandards on gasoline, durability probably no problem,emissions when run on other fuels not known.

SOURCE: Adapted from: J. B. Heywood, “Alternative Automotive Engines andFuels: A Status Review and Discussion of R&D Issues,” contractorreport to OTA, November 1979.

cent of the market share, which appears possi-ble by the mid-1990’s, probably should be con-sidered to represent a significant risk of adversehealth effects unless improved particulate con-trols are incorporated or unless further researchprovides firmer evidence that diesel particulateproduce no special hazard to human health.

ELECTRIC VEHICLES

The substitution of electric vehicles (EVs) fora high percentage of U.S. automobiles and lighttrucks may have a number of environmental ef-fects. The reduction in vehicle-miles traveled byconventional gasoline- and diesel-powered vehi-cles will reduce automotive air pollution, whereasthe additional requirements for electricity will in-crease emissions and other impacts of power gen-eration. Changes in materials use may have envi-ronmental consequences in both the extractiveand vehicle manufacturing industries. The use oflarge numbers of batteries containing toxic chem-icals may affect driver and public health and safe-

ty. The different noise characteristics of electricand internal combustion engines imply a reduc-tion in urban noise levels, while differences insize and performance may adversely affect driv-ing safety. Finally, there may be a variety of lessereffects, for example, safety hazards caused by in-stallation and use of large numbers of chargingoutlets.

Power Generation

As discussed in chapter S, utilities should haveadequate reserve capacitv to accommodate high


levels of vehicle electrification without addingnew powerplants. For example, if the utilitiescould use load control and reduced offpeakprices to confine battery recharging to offpeakhours, half of all light-duty vehicular traffic couldbe electrified today without adding new capac-ity. Given the probable constraints on EVs, how-ever, a 20-percent share probably is a more rea-sonable target for analysis. *

The effects on emissions of a 20-percent electri-fication of vehicular travel are mixed but generallypositive. If present schedules for automotive pol-lution control are met and utilities successfullyrestrict most recharging to offpeak hours, this lev-el of electrification would, by the year 2010, leadto the following changes in emissions** com-pared with a future based on a conventional fossilfuel-powered transportation system:

less than a l-percent increase in sulfur diox-ides (SO2),about a 2-percent decrease in NOX,about a 2-percent decrease in HC,about a 6-percent decrease in CO, andlittle change in particulates.40

The positive effects on air quality may in real-ity be more important than these emission figuresimply. The addition of emissions due to electricityproduction occurs outside of urban areas, andthe pollution is widely dispersed, while the vehi-cle emissions that are eliminated occur at groundlevel and are quite likely to take place in denseurban areas. Thus, the reduction in vehicularemissions should have a considerably greater ef-fect on human exposure to pollution than thesmall increase in generation-related emissions.Also, any relaxation of auto emissions standardswill increase the emissions reductions and airquality benefits associated with “replacement”of the (more polluting) conventional autos. Onthe other hand, future improvements in automo-bile emission controls–certainly plausible given

*As noted elsewhere, however, this is still an extremely optimisticmarket share even for the long term, unless battery costs are sharplyreduced and longevity increased, or gasoline availability decreases.

**Assuming existing emission regulations for powerplants.40W. M. Carriere, et al., The Future Potentia/of E/ectric and Hybrid

Vehic/es, contractor report by General Research Corp. to OTA,forthcoming.

progress during the past decade–might decreasethe air quality benefits of electrification. *

Other effects of increased electricity demandmust also be considered. Most importantly, a 20-percent electrification of cars will lead to substan-tial increases in utility fuel use, especially for coal.Although the extent of increased coal use will de-pend on the distribution of EVs, if the vehicleswere distributed uniformly according to popula-tion, coal would supply about two-thirds of theadditional power necessary in 2010,41 requiringthe mining of about 38 million additional tons peryear. ** If the EVs replaced gasoline-powered carsgetting 55 mpg, the gasoline savings obtained bythe coal-fired electricity—about 36 billion gal/yr–could also have been obtained by turning thesame amount of coal into synthetic gasoline.***

Resource Requirements

EVs will use many of the same materials, in sim-ilar quantities, as conventionally powered vehi-cles, but there will be some differences whichmay create environmental effects. EVs, for exampie, will require more structural material thantheir conventional counterparts because of thesubstantial weight of the batteries (at least withexisting technology). More importantly, the bat-teries themselves will require some materials inquantities that may strain present supply. Table73 shows the increase in U.S. demand for bat-tery materials for 20-percent electrification oflight-duty vehicular travel by 2000.

The effect on the environment of increases inmaterials demand is difficult to project becausethe increased demand can be accommodated ina number of ways. In several cases, although U.S.

“It is equally reasonable to speculate about future improvementsin powerplant emission controls. For example, more stringent con-trols on new pIants as well as efforts to decrease SOZ emissions fromexisting plants in order to control acid rain damages could increasethe benefits of electrification.

41 Ibid.**Assumptions: 12,000 Btu/lb coal; vehicle energy required =

0.4 kWh/mile at the outlet; total 2010 vehicle miles = 1.55 trillionmiles, 20 percent electric; electrical distribution efficiency = 90percent; generation efficiency = 34 percent.

***Assuming a synfuels conversion efficiency of coal into gasolineof 50 percent.

Ch. 10—Enviroment, Health, and Safety Effects and Impacts . 247

Table 73.—increase in U.S. Use of Key Materialsfor 20 Percent Electrification of Light-Duty

Travel (year 2000)

PercentBattery type Material increase a

Lead-acid . . . . . . . . . . . . . . . . . . . . . . LeadNickel-iron . . . . . . . . . . . . . . . . . . . . . Cobalt

LithiumNickel

Nickel-zinc . . . . . . . . . . . . . . . . . . . . . CobaltNickel

Zinc-chloride . . . . . . . . . . . . . . . . . . . GraphiteLithium metal sulfide . . . . . . . . . . . . Lithium

31.218.214.321.331.834.350.0

103.6aAssuming l(Jopercent of the batteries are of the category shown ... the parcent

increases thus are rrot additive for the same materials.

SOURCE: W. M. Carrier, et al,, The Future Potential of Electric and HybridVehicles, contractor report to OTA by General Research Corp., forth-coming.

and world identified reserves currently are insuf-ficient, increased demand probably will be metby identifying and exploiting new reserves. Theenvironmental effects would then be those of ex-panding mining and processing in the UnitedStates or abroad. In other cases, mining of sea-bed mineral nodules or exploitation of lowerquality or alternative ores (e.g., kaolin-type claysinstead of bauxite to produce aluminum) couldoccur. Supplies of some materials may be madeavailable for cars by substituting other materialsfor nonautomotive demands.

In general, the potential for finding additionalresources and the long-range potential for recy-cling indicate that major strains on resources—and, consequently, environmental impacts of un-usual concern—appear to be unlikely with levelsof electrification around 20 or 30 percent. Localareas subject to substantially increased mining ac-tivity could, however, experience significant im-pacts.

Noise

EVs are generally expected to be quieter thancombustion-engine vehicles, and electrificationshould lower urban noise. The effect may not,however, be large. Although automobiles ac-count for more than 90 percent of all urban traf-fic, they contribute only a little more than halfof total urban traffic noise and a lesser percent-age of total urban noise. A recent calculation ofthe effect on noise levels of 100-percent conver-sion of the automobile fleet to electric vehicles

predicts a reduction in total traffic noise of only13 to 17 percent.42

Safety

EVs will affect automotive safety because oftheir lower performance capabilities and differentstructural and material configuration. Lower ac-celeration and cruising speed, for example, couldpose a safety problem because it could increasethe average velocity differential among highwayvehicles and make merging more difficult. ManyEVs will be quite small and, as discussed in thesection on auto fuel conservation, this may de-grade safety. On the other hand, compensatingchanges in driver behavior or redesign of roadsin response to EVs could yield a net positiveeffect.

Similarly, the net effect of materials differencesis uncertain. The strong positive effect of remov-ing a gas tank containing highly flammable gaso-line or diesel fuel will be somewhat offset by theaddition of the battery packs, which containacids, chlorine, and other potentially hazardouschemicals. Collisions involving EVs may result inthe generation of toxic or explosive gases or thespillage of toxic liquids (e.g., release of nickel car-bonyl from nickel-based batteries). Finally, thenecessity to charge many of the vehicles in loca-tions that are exposed to the weather creates astrong concern about consumer safety from elec-trical shock.

Occupational and PublicHealth Concerns

In addition to the potential danger to drivers(and bystanders) from release of battery chemi-cals after collisions, there are some concernsabout the effects of routine manufacture, use, anddisposal of the batteries. Manufacture of nickel-based batteries, for example, may pose problemsfor women workers because several nickel com-pounds that may be encountered in the manufac-turing process are teratogens (producers of birthdefects). Also, because many potential batterymaterials (lead, nickel, zinc, antimony) are per-

4ZW M. Carriere, General Research Corp., perSOnal communica-tion, June 19, 1981.

98-281 0 - 82 - 17 : QL 3

248 . Increased Automobile Fuel Efficiency and Synthetic Fuel Alternatives for Reducing Oil Imports

sistent, cumulative environmental poisons, theprevention of significant discharges during manu-facture as well as proper disposal (preferably byrecycling) must be assured. Finally, routine vent-ing of gases during normal vehicle operationsmay cause air-quality problems in congestedareas.

These risks do not appear to pose difficult tech-nological problems (most have been rated as“low risk” in the Department of Energy’s (DOE)Environmental Readiness Document for EVS

43)

and existing regulations such as the ResourceConservation and Recovery Act provide an op-portunity for strict controls, but institutional prob-lems such as resistance to further Governmentcontrols on industry obviously could increase thelevel of risk. Recycling could pose a particularproblem unless regulations or scale incentives re-strict small-scale operations, which are often diffi-cult to monitor and regulate.

43u.s. Department Of Energy, Etw;mmnental Readiness fkKwment, Electric and Hybrid Vehicles, Commercialization Phase III

P/arming DOE/ERD-0004, September 1978.

SYNTHETIC FUELS FROM COAL

Development of a synthetic fuels industry willinevitably create the possibility of substantial ef-fects on human health and the environment froma variety of causes. A 2 million barrels per day(MMB/D) coal-based synthetic liquid fuels indus-try will consume roughly 400 million tons of coaleach year, * an amount equal to roughly half ofthe coal mined in the United States in 1980. Theseveral dozen liquefaction plants required to pro-duce this amount of fuel will operate like largechemical factories and refineries, handling multi-ple process and waste streams containing highlytoxic materials and requiring major inputs ofwater and other valuable materials and labor.Transportation and distribution of the manufac-tured fuels not only require major new infrastruc-ture but are complicated by possible new dangersin handling and using the fuels. Table 74 listssome of the major environmental concerns asso-ciated with coal-based synfuels. Note that theseverity of these concerns is sharp/y dependenton the level of environmental control and man-agement exerted by Government and industry.

*This corresponds to an average process efficiency of about 55percent and coal heat content of about 20 x 10 6 Btu/ton. The ac-tual tonnage depends on the energy content of the coals, the con-version processes used and the product mixes chosen. Process effi-ciencies will vary over a range of 45 to 65 percent (higher if largequantities of synthetic natural gas are acceptable in the productstream), and coal heat contents may vary from 12 million to 28million Btu/ton.

The health and environmental effects of thesynfuels fuel cycle can be better understood bydividing the impacts into two kinds. Some of theimpacts are essentially identical in kind (thoughnot in extent) to those associated with more con-ventional combustion-related fuel cycles such ascoal-fired electric power generation. These “con-ventional” impacts include the mining impacts,most of the conversion plant construction im-pacts, the effects associated with population in-creases, the water consumption, and any impactsassociated with the emissions of environmentalresiduals such as SO2 and NOX that are normal-ly associated with conventional combustion offossil fuels.

Another set of impacts more closely resemblessome of the impacts of chemical plants and oilrefineries. These include the effects of fugitive HCemissions and the large number of waste andprocess streams containing quantities of tracemetals, dangerous aromatic HCs, and other tox-ic compounds. These are referred to as “noncon-ventional” impacts in this section.

This distinction between “conventional” and“nonconventional" impacts is continuedthroughout this discussion. In particular, for theconventional impacts, synfuels plants are explicit-ly compared with coal-fired powerplants. A fur-ther understanding of the scale of coal-fired pow-erplants should allow this comparison to better


Table 74.—Major Environmental Issues for Coal Synfuels

Land use andwater quality Air quality Ecosystems Safety and health Other

MiningShort- and long-term Fugitive dust

land use changes, (especially in theerosion, and West)uncertainty ofreclamation in aridWest

Aquifer disturbanceand pollution

Nonpoint source waterpollution (acid minedrainage—East;sedimentation—West)

Subsidence

Llquefaction and refiningPotential surface and Emission of “criteria

ground water pollutants” (i.e.,pollution from NOX, SO 2,holding ponds particulate, etc.

Wastewater discharges Fugitive emission of(East) carcinogenic

Disposal of large substancesamounts of solid Possible release ofwastes trace elements

Local land use Releases duringchanges “upset” conditions

Construction on flood Possible localized odorplains problems

Product transport and end-useProduct spills from Changed automotive

trains, pipelines, and exhaust emissionsstorage (increase in some

pollutants, decreasein others)

Increased evaporativeemissions frommethanol fuels

Toxic productvaporization

Disruption of wildlifehabitat and changedproductivity of theland

Siltation of streamsHabitat fragmentation

from primary andsecondarypopulation growth

Air pollution damageto plants

Contributions to acidrain

Wildlife habitatfragmentation frompopulation increases

Contribution to the“greenhouse” effect

Acute and chronicdamages from spills

Mining accidents Increased water useOccupational diseases for reclamation

in underground Coal transportationmining (e.g., black impacts on roadlung) traffic and noise

Occupational safety Water availabilityand health risks issues (especially infrom accidents and the West)toxic chemicals

Carcinogens in directprocessintermediates andfuel products

Exposure to spills Potential change inUncertain effects of fuel economy

trace elements and Methanol corrosionHCs and reduction of

existing enginelongevity

SOURCE: M. A. Chartock, et al., Environmental Issues of Synthetic Transportation Fuels From Coal, OTA contractor report, forthcoming

serve the reader. A 1,000-MWe plant, for exam-ple, serves all the electrical needs (including re-quirements for industry) of about 400,000 peo-ple. A plant of this size would be large but notexcessively so for a new facility, because manycurrently planned coal-fired plants are larger than600 MWe, and the nationwide average capacityof planned units is 433 MWe.44 Existing plants are,on the whole, much smaller than these newplants, with an average capacity of only 57

AAR. W. Gilmer, et ai., “Rethinking the Scale of Coal-Fired Elec-tric Generation: Technological and Institutional Considerations, ”in Office of Technology Assessment, The Direct Use of Coal, Vo/-ume 11, Pa r t A , 1979.

MWe.45 Some existing plants, however, are verylarge: Arizona Public Service Co.’s Four Cornersplant in New Mexico, for example, has a capacityof 2,212 MWe.46

This comparison is intended to place the envi-ronmental and health impacts of a synfuels plantside by side with the impacts of a technology thatmay be more familiar to readers. We stress, how-ever, that this comparison is not relevant to acomparison of coal liquids and coal-based elec-tricity as competing alternatives.

451bid..% Federal Energy Regulatory commission, Steam-E/ectric P/ant Air

and Water Quality Control Data, Summary Report, October 1979.


Such a comparison can be made only by care-fully considering the end uses for the competingenergy forms, which we have not done. For usein automobile travel, however, a synthetic fuelmay prove to be as efficient in its utilization ofcoal energy as a powerplant producing electricityfor EVs (see “Electric Vehicles, Power Genera-tion” in this chapter), In this case, to the extentthat a synfuels production facility produces fewer(or more) impacts than a powerplant processingthe same amount of coal, the impact of the en-ergy-production stage of the “synfuels to motorfuel” fuel cycle may be considered to be envi-ronmentally superior (or inferior) to the samestage of the “electric auto” fuel cycle.

It is also stressed that the “nonconventional”effects associated with the toxic waste streamsproduced by synfuels plants are essentially impos-sible to quantify at this time, because of signifi-cant uncertainties associated with the type andquantity of toxic chemicals produced, the rateat which these chemicals might escape, the effec-tiveness of control systems, the fate of any escap-ing chemicals in the environment, and finally, thehealth and ecological impacts of various expo-sures to the chemicals.

Because of these uncertainties, there may bea temptation to judge synfuels production main-ly on the basis of its “conventional,” and morequantifiable, impacts. In OTA’s opinion, this isa mistake, because the toxic wastes pose difficultenvironmental questions and also because themagnitudes of several of the more conventionalimpacts are themselves quite uncertain.

Mining

A large coal-based synfuels industry will con-sume a significant portion of U.S. coal output.Although actual coal-production growth duringthe remainder of this century is uncertain, severalsources agree that total production on the orderof 2 billion tons per year is possible by 2000.47

At this level a 2 MMB/D coal synfuels capacitywould require roughly 20 percent of total U.S.production in 2000.

dTThp Direct use of Coal: Prospects and Problems of Production

and Combustion, OTA-E-86 (Washington, D. C.: U.S. Congress, Of-fice of Technology Assessment, April 1979). Also available fromBallinger Publishers in a March 1981 edition.

The impacts of a mine dedicated to synfuelsproduction should be essentially the same asthose from other large mines dedicated to powerproduction and other uses, and thus these im-pacts fit into the “conventional” category. Al-though the coal requirement for a unit plant witha 50,000 barrel per day (bbl/d) output capacity—at Ieast 5 million tons per year*—is high by to-day’s standards, mines are already tending to-wards this size range where it is feasible (e.g.,eight mines in the Powder River Basin producedmore than 5 million tons of coal each in 198048).On the other hand, it is not clear that the geo-graphic distribution (and thus the distribution ofimpacts) of synfuels coal production and produc-tion for other uses will be similar. Because it isdifficult to predict where a future synfuels industrywill be located, the nature of any differences be-tween mining for synfuels and mining for otheruses is uncertain.

As discussed in another OTA report,49 althoughmany of coal mining’s adverse impacts have beenmitigated under State and Federal laws, impor-tant environmental and health concerns remain.The major concerns are likely to be /and reclama-tion failure, acid mine drainage, subsidence ofthe land above underground mines, aquifer dis-rupt ion, and occupational disease and injury.

Mining for synfuels conversion will experienceall of these impacts, although not at all sites.

The folIowing discussion of mining impacts re-lies primarily on the OTA report:

Reclamation.– The use of new mining methodsthat integrate reclamation into the mining proc-ess and enforcement of the Surface Mine Con-trol and Reclamation Act (SMCRA) should reducethe importance of reclamation as a critical nation-al issue. However, concern remains that a combi-nation of development pressures and inadequateknowledge may lead to damage in particularly

*The potential range is about 5 million to 18 million tons per year.The 5-million-ton extreme represents a 65-percent efficient process(not truly a Iique faction process because half of its output is syngas;the upper limit of efficiency for processes producing primarily liq-uids is about 60 percent) using very-high-value (28 million Btu/ton)Appalachian coal. The 18-million-ton extreme represents a 45-per-cent efficient process using low-energy (12 million Btu/ton) lignite.

d8An Assessment of the Development Potential and Production

Prospects of Federa/ Coa/ Leases, OTA-M-1 50 (Washington, D. C.:U.S. Congress, Office of Technology Assessment, December 1981).

@ T h e Direct Use of C’oa[ Op. cit.

Ch. 10—Environment, Health, and Safety Effects and Impacts • 251

vulnerable areas—arid lands and alluvial valleyfloors in the West, prime farmland in the Mid-west, and hardwood forests, steep slope areas,and flood-prone basins in Appalachia. Althoughmost of these areas are afforded special protec-tion under SMCRA, the extent of any damage willdepend on the adequacy of the regulations andthe stringency of their enforcement. Recent at-tempts in the Congress to change SMCRA andadministration actions to reduce the Office of Sur-face Mining’s field staff and to transfer enforce-ment responsibilities to State agencies have raisedconcerns about the future effectiveness of this leg-islation.

Acid Mine Drainage.Acid mine drainage, if notcontrolled, is a particularly severe byproduct ofmining in those regions—Appalachia and partsof the interior mining region (indiana, Illinois,Western Kentucky) —where the coal seams arerich in pyrite. The acid, and heavy metals leached

into the drainage water by the acid, are directlytoxic to aquatic life and can render water unfitfor domestic and industrial use. Zinc, nickel, andother metals found in the drainage can becomeconcentrated in the food chain and cause chronicdamage to higher animals. An additional impactin severe cases is the smothering of stream bot-tom-dwelling organisms by precipitated iron salts.

Acid drainage is likely to be a significant prob-lem only with underground mines, and only afterthese mines cease operating. Assuming strong en-forcement of SMCRA, acid drainage from activesurface and underground mines should be col-lected and neutralized with few problems. onlya very small percentage of inactive surface minesmay suffer from acid seepage. Undergroundmines, however, are extremely difficult to seal

off from air and water, the causal agents of aciddrainage. Some mining situations do not allowadequate permanent control once active mining


and water treatment cease. A significant percent-age of the mines that are active at present or thatwill be opened in this century will present aciddrainage problems on closure.

In a balancing of costs and benefits, it may notbe appropriate to assign to synfuels developmentthe full acid damage associated with synfuelsmines, even though these mines will have aciddrainage problems. This is because drainageproblems may taper off as shallower reserves areexhausted and new mines begin to exploit coalseams that are deeper than the water table. Manyof these later mines will be flooded, reducing theoxidation that creates the acid drainage. It ispossible that many or most acid drainage-pronemines dedicated to a synfuels plant would havebeen exploited with or without synfuels develop-ment.

Subsidence.–Another impact of undergroundmining that will not be fully controlled is subsi-dence of the land above the mine workings. Sub-sidence can severely damage roads, water andgas lines, and buildings; change natural drainagepatterns and river flows; and disrupt aquifers. Un-fortunately, there are no credible estimates ofpotential subsidence damage from future under-ground mining. However, a 2 MMB/D industrycould undermine about a hundred square milesof land area (about one-tenth the area of RhodeIsland) each year, * most of which would be apotential victim of eventual subsidence.

Subsidence, like acid drainage, is a long-termproblem. However, SMCRA does not hold devel-opers responsible for sufficient time periods toensure elimination of the problem, nor does itspecifically hold the developer responsible to thesurface owner for subsidence damage. The major“control” for subsidence is to leave a large partof the coal resources—up to 50 percent or more—in place to act as a roof support. There is obvi-ously a conflict between subsidence preventionand removal of the maximum amount of coal.Moreover, the supports can erode and the roofcollapse over a long period of time. The resultingintermittent subsidence can destroy the value ofthe land for development. An alternative mining

*Assuming half of the coal is produced by eastern and centralunderground mining, 18,000 acres undermined per 10 15 Btu of coal.

technique called Iongwall mining deals with someof these problems by actually promoting subsi-dence, but in a swifter and more uniform fashion.Longwall mining is widely practiced in Europebut is in limited use in the United States. It is notsuitable for all situations.

Aquifer Disruption.–Although all types ofmining have the potential to severely affectground water quantity and quality by physical dis-ruption of aquifers and by leaching or seepageinto them, this problem is imperfectly under-stood. The shift of production to the West, whereground water is a particularly critical resource,will focus increased attention on this impact. Aswith other sensitive areas, SMCRA affords specialprotection to ground water resources, but theadequacy of this protection is uncertain becauseof difficulties in monitoring damages and enforc-ing regulations and by gaps in the knowledge ofaquifer/mining interactions,

Occupational Hazards.–Occupational haz-ards associated with mining are a very visible con-cern of synfuels production, because coal work-ers are likely to continue to suffer from occupa-tional disease, injury, and death at a rate wellabove other occupations (see table 75), and thetotal magnitude of these impacts will grow alongwith the growth in coal production.

The mineworker health issue that has receivedthe most attention is black lung disease, the non-

Table 75.—Fatality and Injury Occurrencefor Selected Industries, 1979

Fatalities Nonfatal injuries

Number Rate a Number Rate a

Undergroundbituminous. . . . . . . 105 0.09 14,131 12.30

Surface bituminous . 15 0.02 2,333 3.47All bituminous coal c

(and lignite) . . . . . . . 137 0.064 16,464 10,20Other surface mining b

(metal, nonmetal,stone, etc.). . . . . . . . 97 0.07 8,121 5.82

Petroleum refining c . . 20 0.0011 8,799 5.30Chemical and allied

products c . . . . . . . . . 55 0.0025 78,700 7.20All industries . . . . . . . 4,950 0.0086 5,956,000 9.20aRate per 200,000 worker-howa (100 workef-yeara).bFor all companies.CFOr companies with 11 or more workers; fatallty data Include deaths due tojob-related accident and Illness.

SOURCE: Bureau of Labor Statlstlcs, personal communication, 19S1; and Staff,Mine Safety end Health Administration, personal communlcatlon, 19S1.


clinical name for a variety of respiratory illnessesaffecting underground miners of which coalworkers’ pneumoconiosis (CWP) is the mostprominent. Ten percent or more of working coalminers today show X-ray evidence of CWP, andperhaps twice that number show other black lungillnesses—including bronchitis, emphysema, andother impairments. so

To prevent CWP from disabling miners in thefuture, Congress mandated a 2-mg/m3 standardfor respirable dust (the small particles that causepneumoconiosis). However, critics now questionthe inherent safeness of this standard and thesoundness of the research on which it is based.Furthermore, other coal mine dust constituents—the large dust particles (that affect the upper res-piratory tract) and trace elements–as well asfumes from diesel equipment also represent con-tinued potential hazards to miners.

Mine safety–as distinct from mine health–hasshown a mixed record of improvement since the1969 Federal Coal Mine Health and Safety Actestablishing the Mining Enforcement and SafetyAdministration was passed. The frequency ofmining fatalities has decreased for both surfaceand underground mines, but no consistent im-provement has been seen in the frequency of dis-abling injuries, Coal worker fatalities numbered139 in 1977, and disabling injuries approached15,000.51 Each disabling injury resulted in an aver-age of 2 months or more of lost time. The numberof disabling injuries has been increasing as moreworkers are drawn to mining and accident fre-quency remains constant.

As shown in table 75, surface mining is severaltimes safer than underground mining. But someunderground mines show safety records equal toor better than some surface mines. Generally,western surface mines are safer than eastern sur-face mines. As western surface-mine productionassumes increasing prominence, accident fre-quency industrywide is likely to decline when ex-

SONational I r-rstitute for Occupational Safety and Health, NationalStudy of Coa/ Workers’ F%eurnoconiosis, unpublished reports onsecond round of examinations, 1975. Cited in The Direct Use ofCoal op. cit.

StMine !jafety and Health Administration, “1 njury Experience atAll coal Mines in the United States, by General Work Location,1977, ” 1978. Cited in The Direct Use of Coa[ op. cit.

pressed as accidents per ton of output. But thisstatistical trend may conceal a lack of improve-ment in safety in deep mines.

Liquefaction

Coal liquefaction plants transform a solid fuel,high in polluting compounds and mineral mat-ter, into liquid fuels containing low levels of sul-fur, nitrogen, trace elements, and other pollut-ants. In these processes, large volumes of gas-eous, liquid, and solid process streams must becontinuously and reliably handled and separatedinto end-products and waste streams. Simultane-ously, large quantities of fuel must be burned toprovide necessary heat and steam to the process,and large amounts of water are consumed forcooling and, in direct liquefaction processes, asraw material for hydrogen production. Theseprocesses, coupled with the general physicalpresence of the plants and their use of a largeconstruction (up to 7,000 men at the peak for asingle 50,000 bbl/d plant) and operating force (upto 1,000 workers per plant), lead to a variety ofpotential pathways for environmental damage.

As noted previously, the following discussiondivides impacts into “conventional” and “non-conventional” according to the extent to whichthe effects resemble those of conventional com-bustion systems. The discussion does not consid-er the various waste streams in detail because oftheir complexity. Appendix 10-A lists the gaseous,liquid, and solid waste streams, the residuals ofconcern, and the proposed control systems forgeneric indirect and direct liquefaction systems.DOE’s Energy Technologies and the Environmenthandbook, 52 from which appendix 1O-A is de-rived, describes these streams in more detail.

Conventional Impacts

An examination of the expected “convention-al” impacts reveals that, with a few exceptions,they are significant mainly because the individualplants are very large and national synfuels devel-opment conceivably could grow very rapidly—

5ZU. S. Depafirnent of Energy, Energy Technologies and the Envi-ronment, Enviromnenta/ Information Handbook DOE/EV/74010-l,December 1980.


not because the impacts per unit of productionare particularly large.

Air Quality.– Emissions of criteria air pollut-ants* from synfuels generally are expected to belower than similar emissions from a new coal-fired powerplant processing the same amount ofcoal. 53 A 50,000 bbl/d synfuels plant processesabout as much coal as a 3,000 MWe power-plant,** but (as shown in fig. 23) emits SO2 andNOX in quantities similar to those of a plant of onlya few hundred megawatts or less. For particulate,synfuels plant emissions may range as high asthose from a 2,200-MWe plant, but emissions formost synfuels plants should be much lower.

In any case, particulate standards for new plantsare quite stringent, so even a 2,200-MWe plant(or a “worst case” synfuels plant) will not havehigh particulate emissions. CO emissions fromsynfuels plants are expected to be extremely low,and are likely to be overwhelmed by a varietyof other sources such as urban concentrations ofautomobiles. HC emissions, on the other hand,conceivably could create a problem if fugitiveemissions—from valves, gaskets, and sources oth-er than smokestacks—are not carefully con-trolled. Although the level of fugitive HC emis-sions is highly uncertain, emissions from a 50,000bbl/d SRC II plant could be as high as 14 tons/day–equivalent to the emissions from severallarge coal-fired plants–if the plant’s valves andother equipment leaked at the same rate asequipment in existing refineries.54

The broad emission ranges shown in figure 23reflect very substantial differences in emissionprojections from developers of the various proc-

*“Criteria air pollutants” are pollutants that are explicitly regulatedby National Ambient Air Quality Standards under the Clean Air Act.Currently, there are seven criteria pollutants: SOl, CO, NO,, pho-tochemical oxidants measured as ozone (Oj), nonmethane HC, andlead.

53M. A. Chaflock, et al., Environment/ /sSues of Synthetic

Transportation Fue/s From Coa& Background Report, University ofOklahoma Science and Public Policy Program, report to OTA,forthcoming.

**The actual range is about 2,500 to 3,600 MW for synfuels proc-ess efficiencies of 45 to 65 percent, powerplant efficiency of 35 per-cent, synfuels load factor of 0.9, powerplant load factor of 0.7.

Sdoak Estimate of Fugitive Hydrocar-

bon Emissions for SRC II Demonstration P/ant, for U.S. Departmentof Energy, September 1980. Based on “unmitigated” fugitiveemissions.

Figure 23.-Size Ranges of New Coal-FiredPowerplants With Hourly Emissions Equal to

50,000 bbl/day Synfuels Plants

2200

2000

600

400 -

200 -

Sox NOX Particulate

SOURCE: M. A. Chartock, et al., “Environment Issues of Synthetic Tranporta-tlon Fuels From Coal,” contractor report to OTA, table revieed by OTA.

esses. OTA’s examination of the basis for theseprojections leads us to believe that the differencesare due less to any inherent differences amongthe technologies and more to differences in de-veloper control decisions, assumptions about theeffectiveness of controls, and coal characteristics.The current absence of definitive environmen-tal standards for synfuels plants will tend to aggra-vate these differences in emission projections,because developers have no emissions targets orapproved control devices to aim at. EPA has beenworking on a series of Pollution Control GuidanceDocuments (PCGDs) for the several synfuels tech-nologies in order to alleviate this problem. Theproposed PCGDs will describe the control sys-tems available for each waste stream and the level

Ch. 10—Enviroment, Health, and Safety Effects and Impacts ● 255

of control judged to be attainable. However, thePCGDs became embroiled in internal and in-teragency arguments and apparently may not becompleted and published in the foreseeablefuture.

The air quality effects of synfuels plants on theirsurrounding terrain vary because of differencesin local conditions—terrain and meteorology—as well as the considerable range of possible emis-sion rates. Some tentative generalizations can,however, be drawn from the variety of site-specif-ic analyses available in the literature. One impor-tant conclusion from these analyses is that indi-

vidual plants generally should be able to meetprevention of significant deterioration (PSD) ClassII limits* for particulate and SO2 with plannedemission controls, although in some cases (e.g.,the SRC II commercial-scale facility once plannedfor West Virginia) a major portion of the limitcould be used up.55 In addition, NOX and COemissions are unlikely to be a problem for indi-vidual pIants in most areas, while regulated HCemissions should remain within ambient air qual-ity guidelines if fugitive HC emissions are mini-mized. 56

Restrictions will exist, however, near PSD ClassI areas in the Rocky Mountain States and nonat-tainment areas in the eastern and interior coalregions. Several of the major coal-producingareas of Kentucky and Tennessee are currentlyin nonattainment status, and siting of synfuelsplants in those areas is virtually impossible with-out changes in current regulations or future airquality improvement.57 Finally, failure to controlfugitive HC emissions conceivably could lead toviolations of the Federal shot-t-term ambientstandards near the plant because, as noted above,the potential emission rate is quite high and

*PSD regulations limit the increases in pollution concentrationsallowed in areas whose air quality exceeds national ambient stand-ards. Class I areas, generally national parks and other areas wherepristine air quality is valued very highly, are allowed only minimalincreases. Class III areas are areas designated for industrial develop-ment and allowed substantial increases. Most parts of the countrypresently are designated Class II areas and allowed moderate in-creases in concentrations. PSD limits are under intense scrutiny byCongress and appear to be primary candidates for change underthe Clean Air Act reauthorization.

sscha~ock, et al., op. cit.

561 bid.571 bid.

because the emissions are released near groundlevel and will have a disproportionately large ef-fect on local air quality .58

Some potential restrictions on siting maybe ob-scured in current analyses by the failure to con-sider the short-term air quality effects of upsetsin the conversion processes. For example, underextreme upset conditions, the proposed (but nowcanceled) Morgantown SRC II plant would haveemitted as much S02 in 2 hours as it would haveemitted during 4 to 10 days of normal opera-tion.59 Unfortunately, most environmental analy-ses of synfuels development have tacitly assumedthat control devices always work properly andpIant operating conditions always are normal.These assumptions may be inappropriate, espe-cially for the first generation of plants and partic-ularly for the first few years of operating experi-ence.

On a wider geographic scale, most analysesshow that the emissions impact of a synfuels in-dustry will be moderate compared with totalemissions from all sources. For example, DOE hasestimated 1995 emissions from all major sourcesfor particulate, SO2, and NOX. Its calculationsshow that a 1.3 MM B/D synfuels industry (com-bining gasification, liquefaction, and oil shale)would represent less than 1 percent of nationalemissions for all three pollutants.60 A more inten-sive development—a 1 MM B/D liquefaction in-dustry concentrated in Wyoming, Montana, andNorth Dakota—would represent a 7.7 percent(particulate), 9.8 percent (SO2), 32 percent(Nox), and 1,7 percent (HC) increase over 1975emissions in a region where existing develop-ment—and thus the existing level of emissions—isquite Iow.61 These additional emissions are notinsignificant, and there has been speculation thathigh levels of development could cause someacid rain problems in the West, especially from

Sal. L. white, et al., Energy From the West, Impact Analysis ReportVolume /, /introduction arrd Summary, U.S. Environmental Protec-tion Agency report EPA-600/7-79-082a, March 1979.

59u .s, @partrnent of Energy, Drafi Appendix C of Fjna/ Environ-

mental Impact Statement: SRC-11 Demonstration Plant, Plant Designand Characterization of Effluents, 1980.

60U.S, Department of Energy, Synthetic Fuels and the Environ-

ment, An Envkonmentaland Regulatory Impacts Analysis, DOE/EV-0087, june 1980.

Glchafiock, et al., op. Cit.


NOX. * Nevertheless, if control systems work asplanned and facility siting is done intelligently,coal-based synfuels plants do not appear to repre-sent a severe threat to air quality.

Water Use. -Water consumption has also beensingled out as a significant impact of a large-scalesynfuels industry, especially in the arid West. Syn-fuels plants are, however, less intensive consum-ers of water than powerplants consuming similaramounts of coal. A 3,000-MWe plant—whichprocesses about as much coal annually as a50,000 bbl/d facility–will consume about 25,000acre-feet of water per year (AFY), whereas thesynfuels facility is unlikely to consume more than10,000 AFY and may consume considerably lessthan this if designed with water conservation inmind. According to current industry estimates,a standard 50,000 bbl/d facility will consumeabout as much water as a 640 to 1,300-MWeplant. Using stricter water conservation designs,the facility may consume as much water as a 400-to 700-MWe plant.62 Achieving an annual syn-fuels production of 2 MMB/D might require 0.3million AFY, or only about 0.2 percent of the pro-jected national freshwater consumptive use of151 million AFY in 2000.63

Environmental impacts associated with synfuelswater requirements are caused by the water con-sumption itself and by the wells, pipelines, dams,and other facilities required to divert, store, andtransport the necessary water.

The impacts associated with consumption de-pend on whether that consumption displaces oth-er offstream uses for the water (e.g., the devel-oper may buy a farmer’s water rights) or is addi-tive to existing uses. In the former case, the im-pact is caused by eliminating the offstream use;

*Current understanding of the transformation of NO X emissionsinto nitrates and into acid rain is not sufficient to allow a firm judg-ment to be made about the likelihood of encountering an acid rainproblem under these conditions.

6ZH. Gold, et al., Water Requirements for Stearn-~/ectriC powerGeneration and Synthetic Fuel Plants in the Western United States,U.S. Environmental Protection Agency report EPA-600/7-77-037,April 1977. Assumes powerplant load factor of 70 percent, synfuelsload factor 90 percent. Synthoil is used as a baseline liquid fuelsplant.

63U, S. Water Resources Council, ~ond Nat;ona/ Water Assess-

ment, The Nation’s Water Resources 1975-2MXI, Volume 11,December 1978.

in displacing farming, for example, the impactmay be a reduction in soil salinization that wasbeing caused by irrigation as well as a reductionin water contamination caused by runoff of fertil-izers and pesticides. Any calculation of impactsis complicated, however, by the probability thatlarge reductions in economic activities (such asfarming) in one area will result in compensatingincreases elsewhere as the market reacts to de-creases in production.

if the water consumption is additive to existinguses, it will reduce downstream flows. In surfacestreams or tributary ground waters connected tothese streams, the consumption may have ad-verse effects on the ability of the stream to dilutewastes and to support recreation, fishing, andother instream uses downstream of the withdraw-al. Also, consumption of ground water, if exces-sive, may lead to land subsidence and saltwaterintrusion into aquifers.

The impacts associated with wells, dams, andother infrastructure may also be significant. im-properly drilled wells, for example, can lead tocontamination of drinking water aquifers. Damsand other storage facilities will increase evapora-tive and other losses (e.g., Lake Powell is under-laid with porous rock and “loses” large amountsof water to deep aquifers). In many cases, thelands submerged by reservoirs have been valu-able recreational or scenic areas. In addition, insome circumstances dams can have substantialimpacts, including drastic changes in the natureof the stream, destruction of fish species, etc. Onthe other hand, the ability of dams to regulatedownstream flow may help avoid both floodingand extreme low-flow conditions and thus im-prove instream uses such as recreation andfishing.

Although consumptive water use by synfuelswill be small on a national basis, local and evenregional effects may be significant. Prediction ofthese effects is made difficult, however, by a num-ber of factors, including substantial uncertaintiesin water availability assessments, levels of disag-gregation in many assessments that are insuffi-cient to allow a prediction of local and subregion-al effects, and the variety of alternative supply op-tions available to developers. Water availability


considerations for the five major river basinswhere synfuels development is most likely to oc-cur are discussed in chapter 11.

Work Force and Population Impacts.-Thesize of the synfuels work force will be large com-pared with power generation; for a 50,000 bbl/dplant, it is equivalent to the work force that wouldbe needed for powerplants totaling 4,000 to 8,000Mw (during peak construction) and to plants to-taling at least 2,500 MW (during operation)64 (seech. 8 for detailed discussion). These high workforce values are particularly important for westernlocations, because significant population in-creases caused by energy development placeconsiderable stress on semiarid ecosystemsthrough hunting and recreational pressures, in-adequate municipal wastewater treatment sys-tems, and limited land use planning.

W. L. white, @ al., Energy From the West, Energy Resource @ve/-

opment Systems Reporf, Vo/urne //: Coa/, U.S. EPA report EPA-600/7-79 -060b, March 1979. Used for powerplant work force only (fora 3,000-MWe plant, construction peak is 2,545, operating force is436),

Summary of Conventional Impact Parameters.–Table 76 provides a capsule comparison of theconventional environmental impacts of synfuelsplants and coal-fired plants.

Nonconventional Impacts

The remaining, “nonconventional” impacts ofsynthetic fuel plants represent substantially differ-ent environmental and health risks than do coal-fired plants and other combustion facilities. Theconversion of coal to liquid fuels differs from coalcombustion in several environmentally importantways. Most importantly, the chemistries of thetwo processes are considerably different. Lique-faction is accomplished in a reducing (oxygenpoor) environment, whereas combustion occursin an oxidizing environment. Furthermore, theliquefaction reactions generally occur at lowertemperatures and usually higher pressures thanconventional combustion.

One major result of these chemical and physi-cal differences is that the heavier HCs originally

Table 76.—Two Comparisons of the Environmental Impacts of Coal-BasedSynfuels Production and Coal-Fired Electric Generation e

A. Coal-fired generating capacity B. Side-by-side Comparison of

that would produce the same environmental impact parameters

impact as a 50,000 bbl/d 3,000 MWe 50,000 bbl/dType of impact coal-based synfuels plant, MWe generator synfuels Units

Annual coal use . . . . . . . . . . . . . . . . . . . . 2,500-3,600 b 6.4-15.0 5.3-17.9 million tons/yrAnnual solid waste . . . . . . . . . . . . . . . . . (2,500-3,600)± c 0.9-2.0+ 0.6-1 .8+ million tons/yrAnnual water use: acre feet/yr

Current industry estimates. . . . . . . . . 640-1,300 25,000 5,400-10,800Conservation case . . . . . . . . . . . . . . . . 400-700 3,400-5,900

Annual emissions: tons/yrParticulate. . . . . . . . . . . . . . . . . . . . . . 120-2,800 2,700 100-2,500Sulfur oxides . . . . . . . . . . . . . . . . . . . . 90-500 27,000-108,000 1,600-9,900Nitrogen oxides . . . . . . . . . . . . . . . . . . 70-400 63,000 1,600-7,800

Hourly emissions: Ib/hrParticulate. . . . . . . . . . . . . . . . . . . . . . 90-2,200 30-800Sulfur oxides . . . . . . . . . . . . . . . . . . . . 70-40 8,800-35,200 500-3,200Nitrogen oxides . . . . . . . . . . . . . . . . . . 60-300 20,500 500-2,500

Peak labor. . . . . . . . . . . . . . . . . . . . . . . . . 4,100-8,000 2,550 3,500-6,800 personsOperating labor . . . . . . . . . . . . . . . . . . . . 2,500 440 360 personsaln ~xamPle-A, the ~werP~ant “~e~ the same coal as the Syntue}s plant, New source performance Standards (NSPS) apply, SOX emissions assumed to be 0.6 lb/10 9

Btu. In B, NSPS also apply but Sox emissions can range from 0.3 to 1.2 lbHOC Btu. In both cases, the sYnfuels Plant Parameters rePresent a ran9e of technologies,with a capacity factor of 90 percent and an efficiency range of 45 to 65 percent; the powerplant is a baseload plant, with a capacity factor of 70 percent, efficiencyof 35 percent.

b ln other words, the amount of coal—and thus the amount of mining—needed to fuel a 50,000 bbl/d synfuels plant is the same as that required fOr a 2,500 to 3,600

MWe powerplant.cA synfue[s plant will have about as much ash to dispose of as a coal-fired powerplant using the same amount of coal. It may have leSS scrubber slud9e, but it maY

have to dispose of spent catalyst material that has no analog in the powerplant . . . thus the +.

SOURCE: M. A. Chartock, et al., .5n4romrrerrtal Issues of Synthetic Transportation Fuels From Coal, Background Report, University of Oklahoma Science and PublicPolicy Program, contractor report to OTA, July 1981,


in the coal or formed during the reactions are notbroken down as effectively in the liquefactionprocess as in combustion processes, and thusthey appear in the process and waste streams.The direct processes (see ch. 6 for a brief descrip-tion of the various coal liquefaction processes)and those indirect processes using the lower tem-perature Lurgi gasifier are the major producersof these HCs; indirect processes using high-tem-perature gasifiers (e.g., Koppers-Totzek, Shell,Texaco) are relatively free of them.

The liquefaction conditions also favor the for-mation of metal carbonyls and hydrogen cyanide,which are hazardous and difficuIt to remove.Trace elements are less likely to totally volatilizeand may be more likely to combine with or dis-solve in the ash. The solids formed under theseconditions will have different mineralogical andchemical form than coal combustion ash, and thevolubility of the trace elements, which generallyis low in combustion ash, is likely to be different.Consequently, solid waste disposal is complicatedby the possibility that the wastes may be morehazardous than those associated with conven-tional combustion.

Finally, the high pressure of the processes, theirmultiplicity of valves and other vulnerable com-ponents, and, for the direct processes, their needto handle liquid streams containing large amountsof abrasive solids all increase the risk of accidentsand fugitive emissions.

The major concerns from the “nonconvention-al” waste streams are occupational hazards fromleaks of toxic materials, accidents, and handlingof process intermediates, and ground and surfacewater contamination (and subsequent health andecological damage) from inadequate solid wastedisposal, effluent discharges, and leaks and spills.

Occupational Hazards.–Coal synthetic fuelplants pose a range of occupational hazards fromboth normal operations and upset conditions.Aside from risks associated with most heavy in-dustry, including exposure to noise, dusts, andheat, and falls from elevated areas, synfuels work-ers will be exposed to gaseous and liquid fugitiveemissions of carcinogenic and other toxic materi-als. During upset conditions, contact with hot gasand liquid streams and exposure to fire and ex-

plosion is possible. Table 77 lists some of the po-tential exposures from coal gasification plantsdocumented by the National Institute for Occu-pational Safety and Health. Table 78 lists thepotential occupational health effects associatedwith the constituents of indirect liquefaction proc-ess streams. Similar exposure and health-effectpotentials would exist for any coal liquefactionprocess.

Although the precise design and operation ofthe individual plant is a critical factor in determin-ing occupational hazards, there are certain gener-ic differences in direct and indirect technologiesthat appear to give indirect technologies someadvantages in controlling health and safety risks.

The advantages of indirect technologies includethe need to separate only gases and liquids (thesolids are eliminated in the very first gasificationstep) in contrast with the gas/liquid/solid phaseseparation requirements of direct processes; few-er sites for fugitive emissions than the direct proc-esses; lower processing requirements for theprocess liquids produced (direct process liquidsrequire additional hydrogenation); the abrasive

Table 77.—Potential OccupationalExposures in Coal Gasification

Coal handling, feeding, and preparation.—Coal dust, noise,gaseous toxicants, asphyxia, and fire

Gasifier/reactor operation. —Coal dust, high-pressure hot gas,high-pressure oxygen, high-pressure steam and liquids, fire,and noise

Ash removal, —Heat stress, high-pressure steam, hot ash, anddust

Catalytic conversion. —High-pressure hot gases and liquids,fire, catalyst, and heat stress

Gas/liquids cooling. —High-pressure hot raw gas and liquidhot tar, hot tar oil, hot gas-liquor, fire, heat stress, and noise

Gas purification. —Sulfur-containing gases, methanol,naphtha, cryogenic temperature, high-pressure steam, andnoise

Methanol formation.—Catalyst dust, fire, and noiseSulfur removal — Hydrogen sulfide, molten sulfur, and sulfur

oxidesGas-liquor separation. —Tar oil, tar, gas-liquor with high con-

centrations of phenols, ammonia, hydrogen cyanide,hydrogen sulfide, carbon dioxide, trace elements, and noise

Phenol and ammonia recovery. -Phenols, ammonia, acidgases, ammonia recovery solvent, and fire

Byproduct storage. —Tar, oils, phenols, ammonia, methanol,phenol recovery solvent and fire

SOURCE: Adapted from U.S. Department of Health, Education and Welfare, “Cri-teria for a Recommended Standard . . . Occupational Exposures In coalGaslficatlon Plants” (Clncinnatl: National Institute of OccupationalSafety and Health, Center for Disease Control, 19S43).

Ch. 10—Environment, Health, and Safety Effects and Impacts Ž 259

Table 78.—Occupational Health Effects of Constituents of Indirect Liquefaction Process Streams

Constituents Toxic effects Stream or area

InorganicAmmonia

Carbon disulfide

Carbon monoxide

Carbonyl sulfideHydrogen sulfide

Hydrogen cyanide

Mineral dust and ash

Nickel carbonyl

Trace elements: arsenic, beryllium,cadmium, lead, manganese, mercury,selenium, vanadium

Sulfur oxides

OrganicAliphatic hydrocarbons

Aromatic amines

Single-ring aromatics

Aromatic nitrogen heterocyclics

Phenols

Polycyclic aromatic hydrocarbons (PAH)

Acute: respiratory edema, asphyxia,death

Chronic: no evidence of harm fromchronic subirritant levels

Acute: nausea, vomiting, convulsionsChronic: psychological disturbances,

mania with hallucinationsAcute: headache, dizziness, weakness,

vomiting, collapse, deathChronic: low-level chronic effects not

establishedLittle data on human toxicityAcute: collapse, coma, and death may

occur within a few seconds.Insidious, may not be detected bysmell

Chronic: possible cocarcinogenAcute: headache, vertigo, nausea,

paralysis, coma, convulsions, deathChronic: fatigue, weaknessChronic: possible vehicle for polycyclic

aromatic hydrocarbons andcocarcinogens

Acute: highly toxic, irritation, lungedema

Chronic: carcinogen to lungs andsinuses

(Complex)

Acute: intense irritation of respiratorytract

Chronic: possible cocarcinogen

Most not toxic. N-Dodecane potentatesskin tumors

Acute: cyanosis, methemoglobinemia,vertigo, headache, confusion

Chronic: anemia, skin lesions (aniline)Benzidine and beta-naphthylamine are

powerful carcinogensAcute: irritation, vomiting, convulsionsChronic: bone-marrow depression,

aplasiaAcute: skin and lung irritantsChronic: possible cocarcinogensChronic: possible carcinogens, skin

and lungsChronic: skin carcinogens, possible

respiratory carcinogens

Gas liquor

Concentrated acid gas

Coal-lockhopper vent gasRaw gas from gasifier

Concentrated acid gasCoal-lockhopper vent gasRaw gas from gasifierConcentrated acid gasCatalyst regeneration off-gas

Concentrated acid gasCoal-lockhopper vent gas

Ash or slag

Catalyst regeneration off-gas

Bottom ashFly ashGasifier ashSolid waste disposalCombustion flue gases

Evaporative emissions from productstorage

Coal-lockhopper vent gasGas liquor

Coal-lockhopper vent gasGas liquor

Gas liquorCoal-lockhopper vent gasGas liquor

Gas liquorCoal-lockhopper vent gasRaw gas

SOURCE: US. Department of Energy, Energy Technologies and the Environment. Environmental Information Handbook DOE/EV/74O1O-1, December 1980.

nature of the direct process stream (which con- streams. Lurgi gasifiers, however, produce atains entrained solids); and fewer dangerous aro- wider range of organic compounds than the high-matic compounds, including polynuclear aromat- er temperature gasifiers and as a result are moreics and aromatic amines, than in direct process comparable in health risk to direct processes.


In sum, however, the indirect processes appearto have a lower potential for occupational healthand safety problems than the direct processes.In actual practice other factors—such as differ-ences in the selection of control equipment andin plant design, maintenance procedures, andworker training—conceivably could outweighthese differences. In fact, developers of liquefac-tion processes appear to be aware of the poten-tial hazards and are taking preventive action suchas providing special clothing and providing fre-quent medical checkups. Nevertheless, the occu-pational health risk associated with synthetic fuelplants must be considered a major concern.

Ground and Surface Water Contamination.–A portion of the solid waste produced by liquefac-tion plants is ash-bottoms, fly ash, and scrubbersludge from the coal-fired boilers—materials thatare routinely handled in all coal-fired powerplantstoday. Much of the waste, however, is ash or slagfrom the gasifiers producing synthesis gas or hy-drogen and chars or “bottoms” from the directprocesses (although much of the latter materialis expected to be recycled to the gasifies), Asnoted previously, this material is produced in areducing atmosphere and thus contains organiccompounds as well as trace elements whose solu-bility may be different from that produced in theboiler.

Other solid wastes that may create disposalproblems more severe than those of powerplantwaste include spent catalysts and sludges fromwater treatment. Total solid wastes from a 50,000bbl/d plant range from 1,800 to 5,000 or moretons per day.65 At these rates, a 2 MM B/D industrywould have to dispose of between 26 million and72 million tons of wastes per year. The major con-cern from these materials is that water percolatingthrough landfill disposal areas may leach the toxicorganic compounds and trace elements out ofthe wastes and into the ground water. Current-ly, the extent of this risk is uncertain, althoughtests of EDS66 and SRC-II67liquefaction reactor

bscha~ock, et al., op. Cit.66R. C. Green, “Environmental Controls for the Exxon Donor SOl-

vent Liquefaction Process, ” Second DOE Environmental ControlSymposium, Reston, Va., Mar. 19, 1981.

b7Supra 59.

wastes and gasifier ash from several gasifiersbayielded Ieachates that would not have been ratedas “hazardous” under Resources Conservationand Recovery Act criteria. *

One major problem with permanent landfilldisposal, however, is that damage to ground wa-ter may occur at any future time when the land-fill liner may be breached–many of the toxic ma-terials in the wastes are either not degradable orwill degrade very slowly, and may last longer thanthe design life of the liner.

Liquid effluent streams from liquefaction plantsalso pose potential water pollution problems. Al-though there are a number of wastewater sourcesthat are essentially conventional in character—cooling tower and boiler blowdown, coal storagepile runoff, etc.—the major effluent streams, fromthe scrubbing of the gases from the gasifiers andfrom the water separation streams in the directprocesses, contain a variety of organics and tracemetals that will pose difficult removal problems.The direct processes are expected to have thedirtiest effluent streams, the indirect systemsbased on Lurgi gasifiers will also pose some prob-lems because of their high production of organics,and the systems based on high-temperature gasi-fiers should have only moderate treatment re-quirements. 69

Although total recycle of water is theoreticallypossible, in practice this is unlikely and “zero dis-charge” will only be achieved by using evapora-tion ponds. Aside from the obvious danger ofbreakdown of the pond liner and subsequentground water contamination (or overflows fromflooding), evaporation ponds may pose environ-mental problems through the formation of toxicgases or evaporation of volatile liquids. The com-plex mixture of active compounds in such a pondcreates a particular hazard of unforeseen reac-tions occurring.

ba/nside EPA, Sept. 26, 1980. As reported, researchers from TRWand Radian Corps. have tested ash from Lurgi, Wellman-Galusha,and Texaco gasifiers.

*Wastes are rated as “hazardous” and will require more secure

(and more expensive) disposal if concentrations of pollutants in theIeachates are greater than 100 times the drinking water standard.

‘9H. Gold, et al., “Fuel Conversion and Its Environmental Effects,”Chemical Engineering Progress, August 1979.


Although the use of ponds to achieve zero dis-charge is practical in the West because of the lowrainfall and rapid evaporation rates, zero dis-charge may be impractical at eastern sites with-out artificial evaporation, which is expensive andenergy-intensive. Consequently, it appears prob-able that continuous or intermittent effluentdischarges will occur at eastern plants, withadded risks from control system failures as oneresuIt.

Environmental Management

The likelihood of these very serious potentialenvironmental and health risks turning into actualimpacts depends on a variety of factors, and par-ticularly on the effectiveness and reliability of theproposed environmental controls for the plants,the effectiveness of environmental regulationsand scientists’ ability to detect damages and as-certain their cause.

In general, synfuels promoters appear to beconfident that the control systems proposed fortheir processes will work effectively and reliably.They tend to view synfuels processes as variationsof current chemical and refinery operations, al-beit variations that will require careful design andhandling. Consequently, the environmental con-trols planned for synthetic fuels plants are large-ly based on present engineering practices in thepetroleum refining, petrochemical, coal-tar proc-essing, and power generation industries.

There are reasons to be concerned about con-trol system effectiveness and reliability, however,especially for the first generation of commercialplants. First, few of the wastewater effluents fromeither direct or indirect processes have been sentthrough a complete environmental control sys-tem such as those designed for commercial units.Process waste streams from several U.S. pilotplants have been subjected only to laboratory andbench-scale cleanup tests or else have been com-bined with waste streams from neighboring refin-eries and treated, with a poorly understood levelof success, in the refinery control systems.

Second, scaling up from small-scale operationsis particuIarly difficult for the direct processes,because of the entrainment of solids in the liquidprocess streams. Engineering theory for the scale

up of solids and mixed solids/liquids processesis not well advanced. For the most part, the prob-lem of handling liquid streams containing largeamounts of entrained solids under high-tempera-ture and pressure conditions is outside of currentindustrial experience.

Third, currently available refinery and petro-chemical controls are not designed to capture thefull range of pollutants that will be present in syn-fuels process and waste streams. Several of thetrace elements as well as the polycyclic aromatichydrocarbons (PAHs) are included in this group,although techniques such as hydrocracking areexpected to help eliminate PAHs when they ap-pear in process streams. (As noted previously,problems with the trace organics generally arefocused on the direct and on low-temperatureindirect processes, because high-temperature gas-ifiers should effectively destroy most of thesecompounds.)

Fourth, in some cases, compounds that gener-ally are readily controlled when separately en-countered appear in synfuels process and wastestreams in combinations that complicate control.For example, current processes for removing hy-drogen sulfide, carbonyl sulfide, and combusti-bles tend to work against each other when thesecompounds appear in the same gas stream,70 asthey do in synfuels plants. Also, the high levelof toxics that appear in the waste streams maycreate reliability problems for the biological con-trol systems.

Fifth, as noted earlier, the high pressures, multi-plicity of valves and gaskets, and (for the directprocesses) the erosive process streams appear tocreate high risks of fugitive emissions. plans forcontrol of these emissions generally depend on“directed maintenance” programs that stress fre-quent monitoring and inspection of vulnerablecomponents. Although it appears reasonable toexpect that a directed maintenance program cansignificantly reduce fugitive emissions, rigorousspecifications for such a program have not beenpublished,71 and some doubts have been raised

Zocongressional ReSearCh Service, Synfue/s From COa/ and the

National Synfuels Production Program: Technical Environmentaland Economic Aspects, December 1980 (Committee Print 11-74No. 97-3, january 1981, U.S. Congress).

TI u .S. @panrnent of Energy, Fina/ Environrnenta/ /rnpaCt S~ate-

ment: Solvent Refined Coal-n Demonstration Project, Fort Martin,Monongalia County, W, Va., 2 VOIS., 1981.


about the adequacy of proposed monitoring forpioneer plants.

The significance of these technological con-cerns is uncertain. As noted previously, industryrepresentatives generally have dismissed the con-cerns as unimportant, at least with regard to theextent to which pollution control needs might becompromised. Government researchers at EPAand DOE72 have expressed some important reser-vations, however. On the one hand, they are con-fident that each of the synfuels waste streams isamenable to control, usually with approachesthat are not far different from existing approachesto control of refinery and chemical processwastes. On the other hand, they have reserva-tions about whether or not the industry’s con-trol program, as it is currently constituted, willachieve the high levels of control possible. Poten-tial problem areas (some of which are related)include wastewater treatment, ’J control systemreliability, and pollution control during processupsets.

Virtually all of the Government researchersOTA contacted were concerned that the industryprograms were not addressing currently unregu-lated pollutants but instead were focusing almostexclusively on meeting immediate regulatory re-quirements. Several expressed special concernabout the failure of some developers to exploitall available opportunities to test integrated con-trol systems; they expected these integrated sys-tems to behave differently from the way the indi-vidual devices behave in tests.

The above concerns, if well founded, imply thatenvironmental control problems could have seri-ous impacts on the operational schedules of thefirst generation of commercial plants. These im-pacts could range from extentions in the normalplant shakedown periods to extensive delays forredesign and retrofit of pollution controls.74 Be-

Zzpersonal Communications with headquarters and field person-

nel, EPA and DOE.zJThe draft of EPA’s Pollution Control Guidance Document on

indirect liquefaction also expressed strong concerns about waste-water treatment. /nside EPA, Sept. 12, 1980, “Indirect Liquids DraftSees Zero Wastewater Discharge, Laments Data Gap.”

z4Some of the architectural and engineering firms submitting syn-

fuels plant designs have incorporated certain control system flexibil-ities as well as extra physical space in their control systems designs.These features presumably would reduce schedule problems.Frederick Witmer, Department of Energy, Washington, D. C., per-sonal communication.

cause of the large capital costs of the plants, therewill likely be severe pressure on regulators tominimize delays and allow full-scale productionto proceed. The outcome of any future conflictsbetween regulatory requirements and plantschedules will depend strongly on the publicpressures exerted on the industry and Federal andState Governments.

There are reasons to believe that a great dealof public interest will be focused on the synfuelsindustry and its potential effects. For one, whenplant upsets do occur, the results can be visual-ly spectacular–for example, purging an SRC-llreactor vessel and flaring its contents can producea flame up to 100 ft wide and 600 ft Iong.75 It alsoseems likely that odor problems will accompanythese first plants, and in fact the sensitivity ofhuman smell may render it impossible to evercompletely eliminate this problem. Malodorouscompounds such as hydrogen sulfide, phenols,organic nitrogen compounds, mercaptans, andother substances that are present in the processand waste streams can be perceived at very lowconcentrations, sometimes below 1 part perbillion.76

In addition, the presence of highly carcinogenicmaterials in the process and waste streams ap-pears likely to sensitize the public to any prob-lems with these plants. This combination of po-tential hazards and perceptual problems, coupledwith the industry strategy of locating at least someof these plants quite close to populated areas(e.g., SRC-ll near Morgantown, W. Va., now can-celed, and the Tri-State Synthetic Fuels Projectnear Henderson, Ky.), appears likely to guaranteelively public interest.

The nature of the industry’s response to unex-pected environmental problems as well as its gen-eral environmental performance also will dependon the degree of regulatory surveillance and con-trol exerted by Federal and State environmentalagencies. Although the degree of surveillance andcontrol will in turn depend largely on the envi-ronmental philosophy of the Federal and StateGovernments at various stages in the lifetime ofthe industry-a factor that is unpredictable-it willalso depend on the legal framework of environ-

75 Supra 59.7%upra 58.


mental regulations, the scientific groundwork thatis now being laid by the environmental agencies,and the nature of the scientific problems facingthe regulatory system.

Existing Federal environmental legislation givesthe Occupational Safety and Health Administra-tion (OSHA) and EPA a powerful set of tools fordealing with the potential impacts of synfuels de-velopment. OSHA has the power to set occupa-tional exposure standards and define safety pro-cedures for all identified hazardous chemicals inthe workplace environment. EPA has a wide vari-ety of legal powers to deal with synfuels impacts,including:

●

●

●

●

●

●

●

setting National Emission Standards for Haz-ardous Air Pollutants (N ES HAPS) under theClean Air Act;setting New Source Performance Standards,also under the Clean Air Act;setting effluent standards for toxic pollutants(which, when ingested, cause “death, dis-ease, cancer, genetic mutations, physiologi-cal malfunctions or physical deformations”)under the Clean Water Act;setting water quality standards, also underthe Clean Water Act;defining acceptable disposal methods forhazardous wastes under the Resource Con-servation and Recovery Act;defining underground injection guidelinesunder the Safe Drinking Water Act; anda variety of other powers under the men-tioned acts and several others.77

The regulatory machinery gives the Federal en-vironmental agencies a strong potential meansof controlling synfuels plants’ hazardous emis-sions and effluents. In general, however, the ma-chinery is immature. Because there are no operat-ing commercial-scale synthetic fuels plants in theUnited States, EPA has not had the opportunityto collect the data necessary to set any technol-ogy-specific emission and effluent limitations forsynfuels plants. Aside from this inevitable prob-

TTSee table 4.1, synthetic Fue/s and the Environment: An Environ-

ment/ and Regulatory /mpacts Ana/ysis, Office of Technology im-pacts, U.S. Department of Energy, DOE/EV-0087, june 1980. Also,see ch. 5, The /rnpacts of Synthetic Fue/s Development, D. C.Masselli, and N. L. Dean, jr., National Wildlife Federation, Septem-ber 1981.

Iem, the environmental agencies have not fullyutilized some of their existing opportunities forenvironmental protection. For example, EPA hasallowed its authority to define standards for haz-ardous air pollutants to go virtually unused. Inaddition, in some areas, such as setting effluentguidelines and New Source Performance Stand-ards for air emissions, EPA has a substantial back-log of existing industries yet to be dealt with.

The environmental research programs con-ducted by various Federal agencies will lay thegroundwork for EPA’s and OSHA’s regulation ofthe synfuels industry. The key programs are thoseof EPA and OSHA themselves and those of DOE.DOE’s programs appear likely to be essentiallyeliminated if current plans to dismantle DOE aresuccessful. EPA and OSHA research budgets haveboth been reduced. In particular, EPA has essen-tially eliminated research activities aimed at de-veloping control systems for synfuels wastestreams, on the basis that such development isthe appropriate responsibility of industry. As men-tioned before, Federal researchers familiar withthe industry’s current environmental researchprograms perceive that the industry has little in-terest in developing control measures for poten-tial impacts that are not currently regulated, andthey believe that industry is unlikely to expandits programs to compensate for EPA’s reduc-tions.78

With or without budget cuts, EPA and OSHAface substantial scientific problems in setting ap-propriate standards for hazardous materials fromsynfuels technologies. Probably the worst of theseproblems is that current air pollution and occupa-tional exposure regulations focus on a relativelysmall number of compounds and treat each oneindividually or in well-defined groups, whereassynfuels plants may emit dozens or even hun-dreds of dangerous compounds with an extreme-ly wide range of toxicity (i.e., the threshhold ofharm may range from a few parts per billion toseveral parts per thousand or higher) and a varietyof effects.

The problem is further complicated by the ex-pected wide variations in the amounts and types

78 Supra 72,


of pollutants produced. The synfuels wastestreams are dependent on the type of technology,the control systems used, the product mix chosenby the operator (which determines the operatingconditions), and the coal characteristics. The im-plication is that uniform emission and worker ex-posure standards, such as a “pounds per hour”emission limit on total fugitive HC emissions ora “milligrams per cubic meter” limit on HC ex-posures, are unlikely to be practical because theywould have to be extraordinarily stringent to pro-vide adequate protection against all componentsof the emission streams. Consequently, EPA andOSHA may not be able to avoid the extremelydifficult task of setting multiple separate standardsfor toxic substances.

The regulatory problem represented by the tox-ic discharges is compounded by difficulties in de-tecting damages and tracing their cause. Becauselow-level fugitive emissions from process streamsand discharges or leaks from waste disposal oper-ations probably are inevitable, regulatory require-ments on the stringency of mitigation measureswill depend on our knowledge of the effects oflow-level chronic exposures to the chemical com-ponents of these effluents. Aside from the prob-lems of monitoring for the actual presence of pol-lution, problems may arise both from the longlag times associated with some critical potentialdamages (e.g., 5 to 10 years for some skin can-cers, longer for many soft-tissue cancers) andfrom the complex mixture of pollutants thatwould be present in any emission.

Transport and Use

As synthetic liquids are distributed and usedthroughout the economy, careful control of expo-sure to hazardous constituents becomes less andless feasible. This is especially true for liquid fuelsbecause of the multitude of small users and thegeneral lack of careful handling that is endemicto the petroleum distribution system. Conse-quently, the toxicity of synfuels final productsmay be critical to the environmental acceptabilityof the entire synfuel “fuel cycle.”

The pathways of exposure to hazardous sub-stances associated with synfuels distribution anduse include accidental spills and fugitive emis-

sions from pipelines, trucks, and other transportmodes and storage tanks; skin contact and fumeinhalation by motorists and distributors; and pub-lic worker exposure to waste products associatedwith combustion (including direct emissions andcollected wastes from control systems).

Evaluation of the relative danger of these expo-sure pathways and comparisons of synfuels totheir petroleum analogs are extremely difficult atthis time. Most environmental and health effectsdata on synfuels apply to process intermediates–“syncrudes”- rather than finished fuels. Combus-tion tests have generally been limited to fuel oilsin boilers rather than gasolines in automobiles. ’gThe tests that have been conducted focus moreon general combustion characteristics than onemissions, and those emission characterizationsthat have been done measure mainly particulateand SOX and NOX rather than the more danger-ous organics.80 Adding to the difficulty of deter-mining the relative dangers of synfuels use is aseries of surprising gaps in health effects data onanalogous petroleum products. Apparently, manyof these widely distributed products are assumedto be benign, and monitoring of their effects hasbeen limited.81

Table 79 presents a summary of the known dif-ferences in chemical, combustion, and health ef-fects characteristics of various synfuels productsand their petroleum analogs. The major charac-teristics of coal-derived liquid fuels are:

● The major concern about synthetic fuelsproducts is their potential to cause cancer,mutations, or birth defects in exposed per-sons or wildlife. (Petroleum-based productsalso are hazardous, but usually to a lesserextent than their synfuels counterparts.)82 Ingeneral, the heavier (high boiling point) liq-uids—especially heavy fuel oils—are the mostdangerous, whereas most of the lighter prod-ucts are expected to be relatively free ofthese effects. This distribution of effects maybe considered fortunate because the lighter

T9M. Ghassemi and R. Iyer, Environmental Aspects of SYnfue/

Utilization, U.S. Environmental Protection Agency report EPA-600/7-81-025, March 1981.

~lbid.81 Ibid.

Szlbid.

Ch. IO—Environment, Health, and Safety Effects and Impacts • 265

Table 79.—Reported Known Differences in Chemical, Combustion, and Health EffectsCharacteristics of Synfuels Products and Their Petroleum Analogs

Product Chemical characteristics Combustion characteristics Health effects characteristics

Shale 011Crude . . . . . . . . . . . . . . . . . . . . Higher aromatics, FBN, As, Higher emissions of NO x

particulate and (possibly)certain trace elements

Slightly higher NO X andsmoke emissions

Slightly higher NO x andsmoke emissions

More mutagenic, tumorigenic,cytotoxicHg, Mn

Higher aromatics

Higher aromatics

Higher aromatics

Higher aromatics

Higher aromaticsnitrogen

Higher aromaticsnitrogen

and

and

Gasoline . . . . . . . . . . . . . . . . .

Jet fuels . . . . . . . . . . . . . . . . . Eye/skin irritation, skinsensitization same as forpetroleum fuel

Eye/skin irritation, skinsensitization same as forpetroleum fuel

—

Slightly higher NO x andsmoke emissions

DFM . . . . . . . . . . . . . . . . . . . . .

Residuals. . . . . . . . . . . . . . . . .

Direct IiquefactionSyncrude (H-Coal, SRC ll,

EDS) . . . . . . . . . . . . . . . . . . .

SRC II fuel oil . . . . . . . . . . . . . Higher NO X emissions Middle distillates:nonmutagenic; cytotoxicitysimilar to but toxicitygreater than No. 2 dieselfuel; burns skin.

Heavy distillate: considerableskin carcinogenicity,cytotoxicity, mutagenicity,and cell transformation

Severely hydrotreated:nonmutagenic,nontumorigenic; lowcytotoxicity

—Nonmutagenic, extremely low

tumorigenicity cytotoxicityand fetotoxicity

Non mutagenic

H-Coal fuel oil . . . . . . . . . . . . Higher nitrogen content Higher NO X emissions

EDS fuel oil. . . . . . . . . . . . . . .SRC II naphtha. . . . . . . . . . . .

Higher NO X emissions—Higher nitrogen, aromatics

H-Coal naphtha. . . . . . . . . . . .EDS naphtha. . . . . . . . . . . . . .SRC II gasoline . . . . . . . . . . .H-Coal gasoline . . . . . . . . . . .EDS gasoline . . . . . . . . . . . . .

Indirect liquefact/onFT gasoline . . . . . . . . . . . . . . .FT byproduct chemical . . . . .Mobil-M gasoline. . . . . . . . . .

Higher nitrogen, aromaticsHigher nitrogen, aromaticsHigher aromaticsHigher aromaticsHigher aromatics

Lower aromatics; N and S nil—

(Gross characteristics similarto petroleum gasoline)

—

Noncarcinogenic—N/A

Methanol . . . . . . . . . . . . . . . . .

GasificationSNG . . . . . . . . . . . . . . . . . . . . .

Higher aldehyde emissions Affects optic nerve

Traces of metal carbonylsand higher CO

(Composition varies withcoal type and gasifier

design/operation)

Low/medium-Btu gas. . . . . . . (Emissions of a wide rangeof trace and minorelements and heterocyclicorganics)

—

Nonmutagenic, moderatelycytotoxic

Gasifier tars, oils, phenols . . (Composition varies withcoal and gasifier types;highly aromatic materials).

SOURCE: M. Ghasseml and R. Iyer, Environmental Aspects of Synfuel Utilization, EPA-600/7-81-025, March 1981.


●

●

products–such as gasoline–are more like-ly to be widely distributed.Products from direct liquefaction processesappear more likely to be cancer hazards thando indirect process products, because of thehigher levels of dangerous organic com-pounds produced in the direct processes.Coal-derived methanol fuel appears to besimilar to the methanol currently being us-ed, although there are potentials for con-tamination that must be carefully examined.Methanol is rated as a “moderate hazard”(“may involve both irreversible and revers-ible changes not severe enough to causedeath or permanent injury’’ )83 under chronic–long-term, low-level–exposure, althoughthe effects of multi-year exposures to verylow levels (as might occur to the public withwidespread use as a fuel) are not known.Methanol has been assigned a hazard ratingfor acute exposures similar to that forgasoline,84 but no comparison can be made

BJN. 1. sax, Dangerous properties of Industrial Materials, Fourth

Edition, Van Nostrand Reinhold Co., 1975.BJlbid,

●

for chronic exposures because data forgasoline exposure is inadequate.85 86 Inautomobiles, methanol use increases emis-sions of formaldehyde sufficiently to causeconcern, but lowers emissions of nitrogenoxides and polynuclear aromatics.87 De-pending on the potential health effects of lowlevels of formaldehyde, which are not nowsufficiently understood, and the emissioncontrols on automobiles, methanol use inautomobiles conceivably may provide a sig-nificant net pollution benefit to areas suffer-ing from auto-related air pollution problems.Many of the dangerous organics that are thesource of carcinogenic/mutagenic/teratogen-ic properties in synfuels should be control-lable by appropriate hydrotreating. Tradeoffsbetween environmental/health concerns andhydrotreating cost, energy consumption, andeffects on other product characteristics cur-rently are not known.

OIL SHALE

OaAn Assessment of Oil Shale Technologies, Op. Cit.

851 bid,

BsGhassemi and Iyer, Op. Cit.87Energy From Biological processes, OP. cit.

Production and use of synthetic oil from shaleraises many of the same concerns about limitedwater resources, toxic waste streams and massivepopulation impacts as coal-derived liquid fuels,but there are sufficient differences to demandseparate analysis and discussion, OTA has recent-ly published an extensive evaluation of oil shale;88

the discussion here primarily summarizes the keyenvironmental findings of that study.

U.S. deposits of high-quality oil shale (greaterthan 25 gal of oil yield per ton) generally are con-centrated in the Green River formation in north-western Colorado (Piceance Basin) and northeast-ern Utah (Uinta Basin), The geographic concen-tration of these economically viable reserves toan arid, sparsely populated area with complexterrain and relatively pristine air quality, and theimpossibility of transporting the shale (because

of its extremely low energy density) lead to apotential concentration of impacts that is (at leastin theory) easier to avoid with coal-derived syn-fuels. Thus, compliance with prevention of signifi-cant deterioration regulations for SO2 and particu-Iates may constrain total oil shale developmentto a million barrels per day or less unless currentstandards are changed or better control technol-ogies are developed.

Also, the lack of existing socioeconomic infra-structure implies that environmental impacts as-sociated with general development pressurescould be significant without massive mitigationprograms. Although coal development sharesthese concerns (especially in the West) and haswater and labor requirements as well as air emis-sions that are not dissimilar on a per-plant basis,it is unlikely to be necessary to concentrate coaldevelopment to the same extent as with oil shale.Thus, coal development should have fewer se-


vere physical limitations on its total level of devel-opment.

The geographic concentration of oil shale de-velopment should not automatically be inter-preted as environmentally inferior to a more dis-persed pattern of development, however. Al-though impacts will certainly be more severe inthe developed areas as a resuIt of this concentra-tion, these impacts must be balanced against thesmaller area affected, the resulting pressure onthe developers to improve environmental con-trols to allow higher levels of development, andthe possibility of being able to focus a major mon-itoring and enforcement effort on this develop-ment. Also, the major oil shale areas generallyare not near large population centers, whereasseveral proposed coal conversion plants are with-in a few miles of such centers and may conse-quently pose higher risks to the public.

The volume of the material processed and dis-carded by an oiI shale plant is a significant fac-tor in comparing oil shale with coal-derived fuels.A 50,000 bbl/d oil shale plant using abovegroundretorting (AGR) requires about 30 million tons peryear of raw shale* versus about 6 million to 18million tons of coal (the higher values apply onlyto low-quality Iignites converted in a relativelyinefficient process) for a similarly sized coal lique-faction plant. A modified in-situ (MIS) plant re-quires about the same tonnage of feedstock asdoes the coal plant, Consequently, although theunderground mining of shale thus far has had amuch better worker safety record than coal min-ing, underground mining of coal may be saferthan shale mining for an AGR plant on a “fueloutput” basis, especially when full-scale shalemining begins. Mining for an MIS plant, on theother hand, will be safer than that for the coalplant unless previous shale experience proves tobe misleading.

The very large amount of spent shale representsa difficult disposal problem. An AGR plant mustdispose of about 27 million tons/yr of spent shale,at least five times as much solid waste as that pro-duced by a similarly sized coal synfuels plant (MISplants may dispose of about 6 million tons/yr ofspent shale, one to three times the disposal re-

assuming 25 gal of oil per ton of shale.

quirements of a coal plant). At this rate, a 1MMB/D industry using AGRs will have to disposeof approximately 10 billion cubic feet of com-pacted shale each year.

This material cannot be fully returned to themines because it has expanded during process-ing, and it is a difficuIt material to stabilize andsecure from leaching dangerous compounds—cadmium, arsenic, and lead, as well as organicsfrom some retorts (for example TOSCO II andParajo Indirect)—into surface and ground waters.It also may cause a serious fugitive dust problem,especially with processes like TOSCO II that pro-duce a very fine waste. Even with secure disposal,it will fill scenic canyons and represents an esthet-ic and ecosystem loss. Current research on smallplots indicates that short-term (a few decades)stability of spent shale piles appears likely if suf-ficient topsoil is applied, but the long-term stabil-ity and the self-sustaining character of the vegeta-tion is unknown. For these reasons, solid wastedisposal may be oil shale’s major environmentalconcern.

As with coal liquefaction processes, the “reduc-ing environment” in the retorts produces bothreduced sulfur compounds and dangerous organ-ics that represent a potential occupational hazardfor workers from fugitive emissions and fuel han-dling. Crude shale oil appears to be more muta-genic, carcinogenic, and teratogenic than naturalcrude.

On the other hand, the refined products areless likely to be significantly different in effectfrom their counterparts produced from naturalcrude, and shale syncrude is less carcinogenicor mutagenic than syncrudes from direct coalliquefaction. Although comparisons of relativerisk must necessarily be tentative at this earlystage of development, it appears that the risksfrom these toxic substances–excluding problemswith spent shale—probably are somewhat com-parable to those of the cleanest coal-based lique-faction processes (indirect liquefaction with high-temperature gasifies).

Other oil shale environmental effects of partic-ular concern include:

● The mining of oil shale generates largeamounts of silica dust that is implicated invarious disabling lung diseases in miners.


● Aside from the reduced sulfur compoundsand organics, the crude shale oil contains rel-atively high levels of arsenic, and somewhathigher levels of fuel bound nitrogen thanmost natural crude does. These pollutantsas well as the organics can be reduced in therefining operation.

● In-situ production leaves large quantities ofspent shale underground and thus createsa substantial potential for leaching out tox-ic materials into valuable aquifers. Controlof such leaching has not been demonstrated.

● Although oil shale developers are proposingto use zero discharge of point-source watereffluents, it may be desirable in the futureto treat water and discharge it. The state ofwater pollution control in oil shale develop-ment is essentially the same as in coal-de-rived synfuels, however. Many of the con-trols proposed have not been tested with ac-tual oil shale wastewaters, and none havebeen tested in complete wastewater controlsystems.

BIOMASS

Production of liquid fuels from biomass willhave substantially different impacts from thoseof coal liquefaction and oil shale production.These are described in detail in OTA’s EnergyFrom Biological Processes 89 and summarizedbriefly here.

The liquid fuel that appears to have the mostpotential for large-scale production is methanolproduced from wood, perennial grasses and leg-umes, and crop residues. Ethanol from grains hasbeen vigorously promoted in the United States,but appears likely to be limited by problems offood/fuel competition to moderate productionlevels (a few billion gallons per year).

Obtaining the Resource

Environmental concerns associated with alco-hol fuel production focus on feedstock acquisi-tion to a greater extent than with coal liquefac-

agEnergy From Biological processes, OP. cit.

MIS production —whereby a moderateamount of mining is done to provide spacewith which to blast the shale into rubble andthen retort it underground—may present aspecial occupational hazard to workers fromexplosions, fire, and toxic gases as well asa potential danger to the public if toxic fumesescape from the mine to the surface.

To summarize, the environmental concerns ofoil shale production appear to be quite similarto those of coal-based synfuels production, butwith two important differences. First, the geo-graphic concentration of oil shale production willtend to concentrate and intensify its environmen-tal and socioeconomic impacts to a greater ex-tent than is likely to be experienced by coal de-velopment. Second, the problems of disposingof the huge quantities of spent shale associatedwith the AGR system appear to be substantiallygreater than those of coal wastes.

FUELS

tion. All of the credible alcohol fuel cycles requirevarious degrees of ecological alteration, replace-ment, or disruption on vast land areas. Takinginto account the expenditure of premium fuelsneeded to obtain and convert the biomass intousable fuels, replacing about 10 billion gal/yr ofgasoline with biomass substitutes would requireadding intensive cropping to a minimum of about25 million acres with a combination of sugar/starch crops (for ethanol) and grasses (for metha-nol).

If this savings were attempted strictly by the useof ethanol made from corn, the land requirementprobably would be at least 40 million acres. Ifmethanol from wood were the major source,much of the gasoline displacement theoreticallycould be obtained by collecting the logging resi-dues that are now left in the forest or burned.To replace 10 billion gal over and above theamount available from residues would involve in-creasing the scale and intensity of management(more acreage under intensive management,

Ch. 10—Environment, Health, and Safety Effects and Impacts . 269

shorter times between thinnings, more completeremoval of biomass, more conversion of low-quality stands) on upwards of so million acres ofcommercial forest. It might involve an increasedharvest of forestland with lower productive po-tential—so-called “marginal Iands’’—and it willalmost certainly mean that lands not now sub-ject to logging will be logged. Despite these diffi-culties, however, wood is the most likely sourceof large-scale biomass production.

If handled with care, a “wood-for-methanol”strategy could have a number of benefits. Theseinclude upgrading of poorly managed forests, bet-ter forest fire and pest control through slashremoval, and reduced pressure on the few re-maining unprotected stands of scenic, old-growthtimber because of the added yields of high-qualitytimber that are expected in the long run from in-creased management.

Nevertheless, there is substantial potential fordamage to the forests if they are mismanaged.High rates of biomass removal coupled with shortrotations could cause a depletion of nutrients andorganic matter from the more vulnerable forestsoils. The impacts of poor logging practices—ero-sion, degraded water quality, esthetic damage,and damage to valuable ecosystems—may be ag-gravated by the lessening of recovery time (be-cause of the shorter rotations) and any lingeringeffects of soil depletion on the forests’ ability torebound. The intensified management may fur-ther degrade ecological values if it incorporateswidespread use of mechanical and chemicalbrush controls, very large area clearcuts and elim-ination of “undesirable” tree species, and if itneglects to spare large pockets of forest to main-tain diversity.

Finally, the incentive to “mine” wood frommarginal lands with nutrient deficiencies, thinsoils, and poor climatic conditions risks the de-struction of forests that, although “poor” fromthe standpoint of commercial productivity, arerich in esthetic, recreational, and ecological val-ues. Because the economic and regulatory incen-tives for good management are powerful in somecircumstances but weak in others, a strong in-crease in wood energy use is likely to yield a verymixed pattern of benefits and damages unless theexisting incentives are strengthened.

The potential effects of obtaining other feed-stocks for methanol or ethanol production mayalso be significant. Obtaining crop residues, forexample, must be handled with extreme care toavoid removing those residues that are critical tosoil erosion protection. Large-scale productionof corn or other grains for ethanol is likely to oc-cur on land that is, on the average, 20 percentmore erosive than present cropland. Aside fromcreating substantial increases in erosion, cornproduction will require large amounts of agricul-tural chemicals, which along with sediment fromerosion can pollute the water, and will displacepresent ecosystems.

Equivalent production levels of perennialgrasses and legumes, on the other hand, couldbe relatively benign because of these crops’ resist-ance to erosion as well as their potential to beobtained by improving the productivity of pres-ent grasslands rather than displacing other ecosys-tems. Although large quantities of agriculturalchemicals would be used, the potential for dam-age will be reduced by the low levels of runofffrom grasslands.

Conversion

Production of alcohol fuels will pose a varietyof air and water pollution problems. Methanolsynthesis plants, for example, are small indirectliquefaction plants that may have problems simi-lar to those of coal plants discussed previously.The gasification process will generate a varietyof toxic compounds including hydrogen sulfideand cyanide, carbonyl sulfide, a multitude of oxy-genated organic compounds (organic acids, alde-hydes, ketones, etc.), phenols, and particulatematter. As with coal plants, raw gas leakage orimproper handling of tars and oils would posea significant hazard to plant personnel, and goodplant housekeeping will be essential. Because oflow levels of sulfur and other pollutants in bio-mass, however, these problems may be some-what less severe than in an equivalent-size coalplant.

Ethanol distilleries use substantial amounts offuel–and therefore can create air pollution prob-lems. An efficient 50-million-gal/yr distillery willconsume slightly more fuel than a 30-Mw power-


plant. There are no Federal emissions standardsfor these plants, and the prevailing local standardsmay be weak in some cases, especially for smallonfarm operations.

The plants also produce large amounts ofsludge wastes, called stillage, that are high in bio-logical and chemical oxygen demand and mustbe kept out of surface waters. Although the still-

age from grains is a valuable animal feed productand will presumably be recovered without theneed for any further incentives, the stillage fromsugar crops is less valuable and will require strictregulation to avoid damage to aquatic ecosys-tems, EPA has had a history of pollution controlproblems with rum and other distilleries, and eth-anol plants will be similar to these.

Ch. 10—Environment, Health, and Safety Effects and Impacts . 271

APPENDIX 10A.– DETAILED DESCRIPTIONS OF WASTE STREAMS,RESIDUALS OF CONCERN, AND PROPOSED CONTROL SYSTEMS

FOR GENERIC INDIRECT AND DIRECT COAL LIQUEFACTION SYSTEMS

Table 10A-1 .-Gaseous Emissions and Controls (indirect liquefaction)

Stream componentsGaseous stream Source of concern Controls Comments

Fugitive emissionsVent gases

Coal-lockhoppervent gas

Ash-lockhoppervent gas

Concentratedacid gas

Off-gases fromcatalystregeneration

Evaporativeemissionsfrom storedproducts

Auxillary plantFlue gases

Coal gasification Carbon monoxide,hydrogen sulfide, tars,oils, naphtha, cyanide,carbon disulfide

Coal gasification Particulate, trace elements

Gas purification Hydrogen sulfide, carbonylsulfide, carbon disulfide,hydrogen cyanide, carbonmonoxide, carbon dioxide,light hydrocarbons,mercaptans

Catalytic synthesis Nickel and other metalcarbonyls, carbonmonoxide, sulfurcompounds, organics

Product storage Aromatic hydrocarbons,C 5- C12 aliphatichydrocarbons, ammonia

emissions

Cooling-tower driftand evaporation

Treated wastegases

Power/steam Sulfur and nitrogen oxides,generation particulate trace

elements, coal fines

Power/steam Ammonia, sodium, calcium,generation, sulfides/sulfates,process cooling chlorine, phenols,

fluorine, trace elements,water treatmentchemicals

Gaseous emission Hydrogen sulfide, carbonylcontrols (e.g., sulfide, carbon disulfide,sulfur recovery hydrogen cyanide, carbon

monoxide, carbon dioxide,light hydrocarbons

Compression and recycleof pressurization gas,incineration of waste gas

Scrubber

Stretford or ADIP/Clausprocesses followed by asulfur recovery tail gasprocess, e.g., Beavon,and incineration of theBeavon off-gas in a boiler

Incineration in a flare,incinerator, or controlledcombustion

Vapor recovery systems,use of floating roofstorage tanks,conservation vents.Incinerate

Electrostatic precipitators,fabric filters, flue-gasdesulfurization systems,combustion modification

Proper design and sitingcan mitigate impacts

Essentially the same as forthe concentrated acid gas

The need for and the effectiveness ofincineration/particulate control havenot been defined

The acid gases will be concentrated bythe gas purification process. Thecontrol choice is dependent on thesulfur content of the gases; acombination of Stretford and ADIP/Claus may have the lowest overallcosts.

Other control technology requirementsnot established

Control technologies used in petroleumrefinery and other industries should beapplicable to Lurgi plants; standardspromulgated for the petroleum refiningindustry would probably be extendedto cover the synthetic fuel industry.

Controls applicable to utility andindustrial boilers would generally beapplicable. Established emissionsregulations would cover boilers atLurgi plants

Recycled process water is used forcooling-tower makeup. If cooling-towerdrift becomes a problem then therecycled water will receive additionaltreatment or makeup water will comefrom another source.

—

SOURCE’ U.S. Department of Energy, Energy Technologies and the Environment, Environmenfa/ Information Handbook, DOEIEVI74O1O-1, December 1980


Table 10A-2.—Liquid Waste Stream Sources, Components, and Controls (indirect liquefaction)

Liquid waste Stream componentsstream Source of concern Controls CommentsAsh quench water Gasification Dissolved and suspended Gravity settling of solids: the See table 10A-3

Gas Iiquor

Boiler blowdown

Spent reagentsand sorbents

Acid wastewater

Leachates

Treated aqueouswastes

solids, trace elements, overflow from the settlingsulfides, thiocyanate, basin is recycled back toammonia, dissolved organics, the ash quenching operationphenols, cyanides

Gas purification Sulfides, thiocyanate, ammonia, Lurgi tar/oil separatorcyanides, mono- andpolycyclic organics, tracemetals, mercaptans

Phenosolvan process

Phosam W or Chemi-Linz

Power/steam Dissolved and suspendedgeneration solids

Gaseous emission Sulfides, sulfates, tracecontrols, wastewater elements, dissolved andtreatment suspended solids, ammonia,

phenols, tar oils, hydrogensulfide, carbon dioxide

Product separation Dissolved organics,and purification thiocyanate, trace elements

Gasifier ash, boiler Trace elements, organicsash, FGD sludge,biosludge, spentcatalysts

Wastewater treatment Dissolved and suspendedsolids, trace elements

Bio-oxidation and reverseosmosis

Use as cooling-tower makeupor as ash quench watermakeup

Recovery of reagents from airpollution control processes,addition to ash quenchslurry

Oxidation, use as cooling-toweror quench water makeupLandfill should have

impervious clay liner and aIeachate collection system.If buried in the mine, themine should be dry and ofimpervious rock or clay.

Streams, for final dispositionof ash solids. Capabilities oftechnology in terms ofclarified ash slurry water notknown

Capabilities of tar/oilseparation, Phenosolvan,and ammonia recovery wellestablished in terms ofremoval of major constituents.Capabilities for removal ofminor constituents notestablished. Limited costdata available on processes.

Removes dissolved phenolsfrom water

Removes dissolved ammonia,produces saleable anhydrousammonia.

Removes dissolved organicsand inorganic.

Impacts on the quench systemand subsequent treatmentof clarified water notestablished.

Applicable controls (e.g.,resource recovery disposalin lined pond, dissolvedsolids removal, etc. arewaste- and site-specific; costand performance data shouldbe developed on a case-by-case basis.

—

—

Forced or natural evaporation The effectiveness and costs ofvarious applicable controls(e.g., solar or forcedevaporation, physical-chemical treatment for waterreuse, etc.) not determined.

SOURCE: U.S. Department of Energy, Energy Technologies and the Environment, Emdronrnental Information Handbook, DCN2EVI7401O-1, December 1980.

Ch. IO—Environment, Health, and Safety Effects and Impacts . 273

Table 10A-3.—Solid Waste Stream Sources, Components, and Controls (Indirect liquefaction)

Stream componentsSolid waste stream Source of major concern Controls Comments

Ash or slag Gasification

Scrubber sludge Power/steamgeneration

Boiler ash Power/steamgeneration

Sludge Waste treatment

Spent catalysts Gas shiftconversion,catalytic

Trace elements, sulfides,thiocyanate, ammonia,organics, phenols,cyanides, minerals

Calcium sulfate, calciumsulfite, trace metals,limestone, alkali metalcarbonates/sulfates

Trace elements, minerals

Trace elements,polycyclic aromatichydrocarbons

Metalic compounds,organics, sulfurcompounds

synthesis, sulfurrecovery(gaseousemission centrol)

Tarry and oily sludges Product/byproduct Mono- and polycyclic

Combined with boilerash and flue gasdesulfurization sludgeand disposed of in alined landfill or pond,or buried in the mine

Disposed of with thegasifier ash

Disposed of with thegasifier ash

Combined with gasifierash, boiler ash andflue gas desulfurizationsludge and disposedof in a lined landfiii orburied in the mine.May also be incinerated

Process for materialrecovery, or fixation/encapsulation anddisposal in landiil ormine

injection into the gasifier,separation aromatic hydrocarbons, disposal in a secure

trace elements landfiil, return to themine for burial,incineration

Ash is more than 90percent of the solidwastes generated at aLurgi plant. The choiceand design of disposalsystem depend on theash content of coaland plant/mine sitecharacteristics.

—

Because of lack of dataon waste quantitiesand characteristics,optimum control(s)cannot be established.

The technical andeconomic feasibility ofresource recovery havenot been established

Because of lack of dataon waste quantitiesand characteristics,optimum control(s)cannot be established

SOURCE: U.S. Department of Energy, Energy Technologies and the Env)ronrnent, Environmental Information Handbook, DOI3EVI74O1O1, December 19S0.


Table 10A-4.—Gaseous Streams, Components, and Controls (direct liquefaction)

Operation/auxiliary process Air emissions discharged Components of concern Control methods

Coal storage and pretreatment

Liquefaction

Separation:Gas separation

Solids/liquids separation

Purif ication and upgrading:Fract ionat ion

Hydrotreating

Water cooling

Steam and power generation

Hydrogen generation

Acid gas removal

Sulfur recovery

Hydrogen/hydrocarbon recoveryProduct/byproduct storage

Coal dust

Particulate-laden fluefrom coal dryers

Preheater flue gas

gas

Pressure letdown releases


Preheater flue gas

Particulate-laden vaporsfrom residue cooling(SRC-ll)


Preheater flue gas

Particulate-laden vaporsfrom product cooling(SRC-l)


Preheater flue gas


Drift and evaporation

Boiler flue gas

Preheater f lue gas


Flue gas

Low-sulfur effluent gas b

Pressure letdown releasesSRC dust (SRC-l)Sulfur dustHydrocarbon vapors

Respirable dust, particulate, trace Spray storage piles with water orelements

Respirable dust, particulate, tracemetals, sulfur and nitrogen oxides

Particulate, sulfur and nitrogenoxides

Hydrocarbons, hydrogen sulfide,hydrogen cyanide, ammonia, PAH,hydrogen, phenols, cresylics

Same as for liquefaction letdownreleases

Same as the liquefaction preheater

Particulate, hydrocarbons, traceelements


Same as for liquefaction preheater

Same as for SRCll residue cooling




Ammonia, sodium, calciumsulfides/sulfates, chlorine, phenols,fluorine, trace elements, watertreatment chemicals

Sulfur and nitrogen oxides,particulate


Hydrogen sulfide, hydrogen cyanide,carbon oxides, light hydrocarbons


Hydrogen sulfide, hydrogen cyanide,sulfur dioxide

Hydrogen, hydrocarbonsRespirable dust, particulateElemental sulfurPhenols, cresylics, hydrocarbons,

PAH

polymer. cyclones and baghousefilters for control of dust due tocoal sizing.

Cyclones and baghouse filters. Wetscrubbers such as venturi.

If other than clean gas, scrub forsulfur, nitrogen, and particulatecomponents.

Flaring a

Flaring a


Cyclone and baghouse filter. Wetscrubbers.

Flaring a


Cyclone and baghouse filter. Wetscrubbers

Flaring a

If other than clean gas, scrub forsulfur, nitrogen and particulatecomponents.

Flaringa

No controls available—good designof water management system canminimize losses.

Sulfur dioxide scrubbing, combustionmodifications.


Flaringa


Carbon absorption. Direct-flameincineration. Secondary sulfurrecovery (Beavon).

Direct-fired afterburnerSpray storage piles with water.Store in enclosed area.Spills/leaks prevention.

%ollection, recovery of useful products and incineration may be more appropriate.bA seconda~ sulfur recovev process may be necessary to meet specified air emission standards.

SOURCE: U.S. Department of Energy, Energy Technologies and the Environment, Environmental Information Handbook, DOE/EV174010-1, December 1980.

Ch. IO—Environment, Health, and Safety Effects and Impacts ● 275

Table 10A-5.—Liquid Stream Sources, Components, and Controls (direct liquefaction)

Operation/auxiliary process Wake effluents discharged Components of concern Control methods

Coal pretreatment Coal pile runoff Particulate, trace metals Route to sedimentation pond.Thickener underflow Same as above Route to sedimentation pond.

Water cooling Cooling tower blowdown Dissolved and suspended solids Sidestream treatment(electrodialysis, ion exchangeor reverse osmosis) permitsdischarge to receiving waters.

Hydrogen generation Process wastewater Sour and foul wastewater; Route to wastewater treatmentspent amine scrubbing facility.solution

Acid gas removal Process wastewater Dissolved hydrogen sulfides, Route to wastewater treatmenthydrogen cyanide, phenols, facility.cresylics

Ammonia recovery Process wastewater Dissolved ammonia Route to wastewater treatmentfacility.

Phenol recovery Process wastewater Dissolved phenols, cresylics Route to wastewater treatmentfacility.

SOURCE: U.S. Department of Energy, Energy Technologies and the Environment, Environmental Information Handbook, DOEiEV174010-1, December 1980.

Table 10A.6.—Soiid Waste Sources, Components, and Controls (direct liquefaction)

Operation/auxiliary process Solid waste discharged Components of concern Control methods

Coal pretreatment RefuseSolids/liquids separation Excess residue

or filter cake

Hydrotreating Spent catalyst

Steam and power generation AshHydrogen generation Ash or slag

Mineral matter, trace elements(SRC-ll) Mineral matter, trace elements,(SRC-l) absorbed heavy hydrocarbons

Metallic compounds, absorbedheavy organics, sulfurcompounds

Trace elements, mineral matterTrace elements, sulfides,

ammonia, organics, phenols,mineral matter

Landfill, minefillGasification to recover energy

content followed by disposal(landfill or minefill

Return to manufacturer forregeneration

Landfill, minefillLandfill, minefill

SOURCE: U.S. Department of Energy, Energy Technologies and the Errvironment, Environmental Information Handbook, DOEIEVI74O1O-1, December 1980

Chapter 11

Water Availability for SyntheticFuels Development

Contents

Page

Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 279

Water Requirements for Synfuels Plants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 280

lmpacts of Synfuels Development on Water Availability . . . . . . . . . . . . . . . . . . . 281

Water Availability at the Regional Level . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 283Eastern River Basins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 283The Western River Basins. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 285

TABLES

Table No. Page

80. Estimates of Net Consumptive Use Requirements of GenericSynfuels Plants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 280

81. Regional Streamflow Characteristics 1975 . . . . . . . . . . . . . . . . . . . . . . . . . . . 28182. Preliminary Quantification of Uncertainties With Respect to Water

Availability in the Upper Colorado River Basin . . . . . . . . . . . . . . . . . . . . . . . 288

FIGURES

Figure No. Page

24. Water Resources Regions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27925.The Nation’s Surface Water Laws . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28426.The Nation’s Ground Water Laws . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28427. Streamflows and Projected Increased Incremental Water Depletions,

Yellowstone River at Sidney, Mont. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 286

Chapter 11

Water Availability for Synthetic Fuels Development

INTRODUCTION

Operation of a synthetic fuels plant requires asteady supply of water throughout the year forboth plant and site activities. Availability of waterwill be determined not only by hydrology andphysical development potential, but also by insti-tutional, legal, political, and economic factorswhich govern and/or constrain water allocationsand use among all sectors. This chapter expandsthe environmental discussion of the role of water

Figure 24.—Water

in synfuels development and examines the ma-jor issues that will determine both water availabil-ity for synfuels and the impacts of procuring watersupplies for synfuels on other water users. Thereare five river basin areas where oil shale and coalresources are principally located: in the easternbasins of the Ohio, Tennessee and the UpperMississippi, and inper Colorado and

Resources Regions

the western basins of the Up-the Missouri (see fig. 24).

yCalifornia

Region

/ 0, . , - - -

Souris. Red-Rainy

Pacific NorthwestGreat Lakes

R e g i o n

I Upper M ISSISSI PPI A

Region Texas-Gulf Region

Hawai iRegion

i c

AlaskaRegion

Caribbean Region

SOURCE: U.S. Water Resources Council, The fVatIon’S Water Resources 197S2000, vol. 2, pt. 1, p. 3,

279

98-281 0 - 82 - 19 : QL 3


WATER REQUIREMENTS FOR SYNFUELS PLANTS

Estimates of the consumptive use requirementsof generic synthetic fuels plants producing 50,000barrels per day oil equivalent (B/DOE) of productare shown in table 80. In general, the actualamount of water consumed will vary accordingto the nature of the products produced, processmethods, plant design, and site conditions. Incoal conversion, the largest single component oftotal water consumption is typically for cooling, *with other major components being for hydrogenproduction, waste disposal, and revegetation. Inproducing synfuels from oil shale, retorting andupgrading require the most water; other majoruses are for the handling and disposal of spentshale, and for revegetation.

*The amount of water consumed in cooling will depend on manyfactors, includifig the degree to which evaporative or “wet” cool-ing, or dry cooling, are used. Air or “dry” cooling is an alternativeto wet cooling but is less efficient and generally more expensive.

Table 80.—Estimates of Net Consumptive UseRequirements of Generic Synfuels Plants

(50,000 B/DOE)A

Barrels wate/Acre-feet/year barrel product

Gasification . . . . . . . . . . . . . 4,500-8,000 1.9-3.4Liquefaction. . . . . . . . . . . . . 5,500-12,000 2.3-5.1Oil shale. . . . . . . . . . . . . . . . 5,000-12,000 2.1-5.1aAVall~le estimates me baaed on theoretical Calculations, conceptual desirms,

small-scale experimental facllltles, etc. A range is shown for each generic proc-esa In order to reflect differences among process technologies (e.g., Indirectliquefaction will generally consume more water than direct liquefaction; modl-fled-in-situ will generally consume less water than aboveground 011 shale pro-cesses), plant design options (e.g., alternative methods of water reuse, con-servation, and cooling), and sites. Estimates also vary with the level of detailand state of development of the engineering designs. There are also at leasttwo major elements of uncertainty surrounding these estimates. Firat, both therefinement and optimization of operational requirements are Ilmlted by the lackof commercial experience. Secondly, eatlmates commonly assume zerowastewater discharge, which Is to be achieved vla the treatment and reuse ofplant wastewater for cooling water makeup and boiler feed; however, the treat-ment processes to be used genereJly have yet to be demonstrated on a commer.clal scale. Although the estimates shown In table 80 may thus not be represen.tative of actual consumptive use requirements In specific cases, the magnitudeof the other uncertainties concerning water availability in general, as discuss-ed [n this chapter, will likely overshadow the question of how much water willbe required for expected synfuels development. The following references pro-vide additional details:1. Office of Technology Assessment, An Assessment of 0!1 Shale Technologies,

June 1980, ch. 9.2. Ronald F. Probsteln and Harris Gold, Water In Synthetic Fuel F’roducflorr,

MIT Press, Cambridge, Mess., 1978.3. R. M. Wham, et al., Llquefactlon Technology Assessment—Phase 1: Indirect

Llquefection of Coal to Methanol and Gasoline Using Availabie Technology,Oak Ridge National Laborato~, Oak Ridge, Term., February 1981.

4. Exxon Research and Engineering Co., EDS Coal Liquefaction ProcessIX?velopment, phase V, VOIS. 1, 11, and Ill, March 1881.

5. Harris Gold and David J. Goldstein, “Water Requirements for Synthetic FuelPlants;” and Harris Gold, J. A. Nardella, and C. A. Vogel (ads.), “Fuel Conver-sion and Its Environmental Effects,” Chemical Engineering Progress, August1979, pp. 58-84.


Synfuels plants will also generally require waterfor other process-related activities such as envi-ronmental control (e.g., dust control) and for as-sociated growth in population, commerce, andindustry (e.g., for water supply and sewerage).Plant activities will not all require water of similarqualities. As examples, high-quality water is re-quired for processing; intermediate-quality wateris required for cooling; mining, materials prepara-tion, and disposal activities are the least sensitiveto water quality characteristics.

Procuring water supplies for synfuels plants willrepresent a small fraction of total plant investmentand operations costs (typically less than 1 per-cent). * * Thus, assuming that the overall econom-ic feasibility of the plant has been established, themore critical industrial considerations in select-ing a water source will be the ease of acquiringwater of appropriate quality and the certainty ofthe yield. Major water sources for synfuels wouldinclude the direct diversion of surface water, thepurchase or transferring of existing water rights,the use of existing or the construction of newstorage, the use of tributary and nontributaryground water,*** savings from improved efficien-cy, reuse, and conservation by all users, and inter-basin diversions.

The feasibility and attractiveness of sources willvary among sites according to environmental,social, legal, political, and economic criteria, and

**Obtaining reliable and comparable cost data on the procure-ment of water to the synfuels industry is difficult because of varia-tion in the conditions surrounding each sale (e.g. water rights varyaccording to their seniority, historic use, point of diversion, etc.).As examples, annual costs per acre-foot of consumption vary be-tween $50 to $300; water rights have sold for as high as $2,000/acre-foot (in perpetuity). Assuming a cost of $2,000/acre-foot, waterrights costs would still represent a maximum of only 0.8 percentof the cost of a $2 billion plant with an average annual consump-tion of 8,000 acre-feet. Note that what is bought is the right to usewater, not the water per se.

Costs are, nevertheless, important industrial criteria for evaluatingalternative sources of water supply. Costs will also be importantfor water resources planning efforts, as they will help to determinethe nature and extent of impacts on other water users from syn-fuels development.

***The development of deep, nontributary ground water, whichis hydrologically unconnected to the surface flow, can be consideredas an “additional” source of water. Development of tributaryground water, which is hydrologically connected to streamflow,does not represent an increase in supply and may alter the surfaceflow regime,

Ch. 1 l—Water Availability for Synthetic Fuels Development . 281

it is therefore difficult a priori to predict how andwhich water “packages” will be assembled. Evi-dence suggests that the industry is conservativein planning for a plant’s water resource needs inorder to ensure (both hydrologically and legal-ly) that the plant obtains its minimum operatingrequirements. As examples, developers can se-cure several different sources of supplies; esti-

mates of resource needs will include a margin ofsafety; and sources can be “guaranteed” by ob-taining agreements not only with rights holdersbut also with upstream appropriators and/or po-tential downstream claimants. Synfuels technol-ogy modifications should also be forthcomingfrom the industry, if needed to reduce waterneeds.

IMPACTS OF SYNFUELS DEVELOPMENT ON WATER AVAILABILITY

In the aggregate, water consumption require-ments for synfuels development are small.Achieving a synfuels production capability of 2MMB/DOE would require on the order of 0.3 mil-lion acre-feet/year (AFY), which will be distributedamong all of the Nation’s major oil shale and coalregions. This compares with an estimated (1975)total national freshwater consumptive use of119million AFY, of which about 83 percent is foragriculture. ’ Table 81 shows the general hydro-logic characteristics of the principal river basinsto be affected.

Although in the aggregate synfuels water re-quirements are small, each synfuels pIant, never-theless, is individually a relatively large water con-sumer. Depending on both the water supplysources chosen for a synfuels plant and the sizeand timing of water demands from other users,synfuels development could create conflictsamong users for an increasingly scarce water sup-

1 U.S. Water Resources Council, The Nation Water Resources—1975-2000, December 1978. The assessment projects a total na-tional freshwater consumption of 151 million AFY in 2000, of whichabout 70 percent would be for agriculture.

ply or exacerbate conflicts in areas where wateris already limited or fully allocated. Sectors thatwill be competing for water will vary among theregions and will include both offstream uses (e.g.,agriculture, industry, municipalities) and instreamuses (e.g., navigation, recreation, water qualitycontrol, fish and wildlife, hydropower). Becauseenergy developers can afford to pay a relativelyhigh price for water, nonenergy sectors are notlikely to be able to compete economically againstsynfuels for water. However, it is speculative toidentify which sectors may be the most vulner-able to synfuels development.

Public reactions to proposed water use changeand nonmarket mechanisms can be used to allo-cate and protect water for use by certain sectorsdepending on the region and State. Examples ofnonmarket mechanisms include the assertion ofFederal reserved water rights, water quality legis-lation, and State water allocation laws. Whilesuch mechanisms may prevent developers fromalways obtaining all the water they need, the syn-fuels industry is expected to obtain the major por-tion of its water requirements.

Table 81 .—Regional Streamflow Characteristics 1975 a (millions acre-feet/year)

Consumption c

Mean annual streamflow b 1975 2000 Low flow ratio d Low flow monthOhio . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199 2.0 4.9 0.15 SeptemberTennessee . . . . . . . . . . . . . . . . . . . . . . . . . 46 0.5 1.2 0.38 SeptemberUpper Mississippi. . . . . . . . . . . . . . . . . . . 136 1.3 3.0 0.23 JanuaryUpper Colorado. . . . . . . . . . . . . . . . . . . . . 11 2.7 3.6 0.12 JulyMissouri . . . . . . . . . . . . . . . . . . . . . . . . . . . 49 17.3 22.3 0.19 Januaryau,s, water ROSOUrCOS council (wRc), The P&Won Water ROSOUrCOS— 1975-.

%RC, table IV-1. Note that all these outflows are inflows to a downstream river basin.CWRCj table 111.3,dRatio of the annual flow of a Vew dw year (that flow which will be exceeded with a 95-percent probability in any Year) to the mean annual flow. WRC, table ‘V-2

SOURCE: US. Water Resources Council as tabulated by OTA.


Photo credit: Office of Technology Assessment

Competing uses will increase pressures on the Nation’s water resources, especially in the arid West

The nature and extent of the impacts of syn-fuels development on water availability in gener-al, and on competing water users, are controver-sial. The controversy arises in large part becauseof the many hydrologic, institutional, legal, andpolitical constraints and uncertainties that will ulti-mately determine when, how, and if users willbe able to obtain the water they need. Further-more, analyzing these constraints and uncertain-ties is difficult because of many additional com-plex factors: the lack of dependable and consist-ent data, limitations of demand-forecasting meth-ods, time and budget constraints, and the unpre-dictability of future administrative decisions andlegal interpretations. In some cases, the uncer-tainties about water availability in general appearto be so large that they overshadow the question

of how much water will be required for synfuelsdevelopment.

OTA’s study2 found that there was considerablevariation in the quality, detail, and scope of thewater availability assessments that have beencompleted related to synfuels development. Fewstudies take into account all of the issues that willdetermine resource allocations and use; and stud-ies rarely try to address the likely, cumulativewater resource impacts of alternative decisionson reducing uncertainties and resolving conflictamong competing water users. Decision makersneed to be better informed about the assump-

Zwright Water Engineers, Inc., “Water Availability for SyntheticFuels,” prepared for the Office of Technology Assessment, June1981.

Ch. n-Water Availability for Synthetic Fuels Development ● 2 8 3

tions and uncertainties upon which reports arepredicated, so that estimates can be properly in-terpreted and tradeoffs can be evaluated.

Some of the major uncertainties about wateravailability for synfuels are discussed below. Moreinformed decisions on water availability ques-tions, however, can only partially be achievedby “improving” studies themselves; more in-formed decisions also depend on greatly im-proved water planning practices in general in theNation. The present fragmentation of responsibil-

ities for water policy, planning, and managementeffectively prevents an assessment of the cumula-tive impacts of water resource use on an ongoingand comprehensive basis. *

*The fragmentation of water-related responsibilities among agen-cies, States, and levels of governments arises in large part becauseriver system boundaries rarely coincide with political boundaries.As a result, there can be major inconsistencies in water manage-ment practices across the country (e. g., inconsistent criteria forevaluation; the lack of integrated planning—including datamanagement—for ground and surface waters, water quality andquantity, and instream and offstream uses).

WATER AVAILABILITY AT THE REGIONAL LEVEL

Eastern River Basins

In the principal eastern basins where energyresources are located (i.e. Ohio, Tennessee, andthe Upper Mississippi), water should be adequateon the mainstems and larger tributaries, withoutnew storage, to support planned synfuel develop-ment.3 However, localized water scarcity prob-lems could arise during abnormally dry periodsor due to conflicts in use on smaller tributaries.The severity and extent of local problems can-not be fully ascertained from existing data andhave not yet been examined comprehensively, *but, with appropriate water planning and man-agement, these problems should be reduced ifnot eliminated.

There are, nevertheless, various uncertaintiesin the eastern basins that will influence wateravailability for synfuels development, and difficultlocal situations could arise.4 For example, 7-day,10-year minimum low flows are used to esti-mate water availability. * * These estimates areessentially based on recorded streamflow data

— . —31 bid.“For example, available reports related to ynfuels for the Tennes-

see River Basin generally deal with specific project sites; the spar-sity of comprehensive information with respect to cumulative im-pacts and possible water use conflicts is presumably because ofthe large quantities of water available at the regional level. The OhioRiver Basin Commission study focuses on water availability for plantslocated on the mainstem, even though there are facilities being pro-posed for tributaries.

* *The use of the 7-day, 10-year minimum flow in the East is alsothe basis for water quality regulations and for estimating critical con-ditions for navigation in rivers with limited storage.

which can be of varying quality. Furthermore, byusing historical streamflow records directly,reports on water availability in the eastern basinscharacteristically underestimate the frequency offuture critical low flows; i.e., as flow depletionsincrease in the future, the critical flow associatedwith the 7-day, 10-year frequency will actuallyoccur more often in the future than the historicaldata would indicate.

The political, institutional, and legal factors thatwill determine water availability for synfuels inthe eastern basins differ in type and complexityfrom those in the western basins. For example,the East and West have different regional hydro-logic characteristics, with the East being relativelyhumid. There are also varying legal and adminis-trative structures as shown in figures 25 and 26:riparian water law is generally applied in the Eastwhereby riparian landowners are entitled to anequal, “reasonable” use of adjacent streamflow;the prior appropriation doctrine is generally ap-plied in the West whereby water rights are basedon “beneficial” use with priorities assigned ac-cording to “first in time, first in right. ” Further-more, in the East there is a general lack of treatiesand compacts, and there are no major Federal(including Indian) reserved water rights questions.

Although water may thus appear to be morereadily available for synfuels development in theEast (e.g., through the transfer of ownership ofriparian land), eastern water law can result insignificant uncertainty concerning the depend-ability of the supply: because all users have equal

284 . Increased Automobile Fuel Efficiency and Synthetic fuels: Alternatives for Reducing Oil Imports

Figure 25.– The Nation’s Surface Water Laws

“Ripar ian only appl ies to lakes: Washington ““Riparian with permit system: Delaware

I

Arkansas

Texas

—

Louisiana

~ / Florlda

Combinationappropriation/riparian

Riparian with permit system

Appropriation doctrine

Riparian doctrine

SOURCE: U.S. Water Resources Council, (The Ncrtiorr’s Water Resources 7975-2000, vol. 2, pt. iv, P. 118).

Figure 26.— The Nation’s Ground Water Laws

Nebraska

SOURCE: U.S. Water Resources Council, (The Afafiort’s Water Resources 197%?000, vol. 2, pt. iv, p. 118).

Ch. n-Water Availability for Synthetic Fuels Development Ž 285

rights under riparian law, the law does not pro-tect given users against upstream diversions oragainst pumping by adjacent wells. * Uncertain-ty also arises because eastern water law has notbeen as well advanced through court tests as inthe West. There are also questions in the Eastconcerning the availability of water from Feder-al storage (i.e., in the Ohio River Basin) becauseof uncertainties regarding who has responsibilityfor marketing and reservoir operation.

The Western River Basins

Competition for water in the West already ex-ists and is expected to intensify with or withoutsynfuels development. There are potentialsources of supply in both the Upper Coloradoand the Missouri River basins that could supportsynfuels development. However, the issues deter-mining whether and the extent to which thesesources will be available for use differ betweenthe two basins. These issues concern complexState water allocation laws, compacts and treat-ies, Federal including Indian reserved water rightsclaims, and the use of Federal storage. In addi-tion, the use of “mean annual virgin flows” inboth regions to characterize the hydrology resultsin the masking of important elements of hydro-logic uncertainty.** However, and in contrastwith the situation in the East, although the com-plex water setting in the West will probably make

*For example, Federal storage has not yet been utilized in Illinoisbecause delivery of the water from the reservoirs (e.g., to the syn-fuels plants) cannot be guaranteed along the river; riparian land-owners along the way could intercept the released water. Energycompanies are thus faced with having to build private pipelines.

**The accuracy of mean annual virgin flows is uncertain due topossible inaccuracies in the underlying data both on streamflowsand on depletions. (Depletions are usually not measured directlyfor practical reasons.) Furthermore, virgin flow estimates are treatedas both deterministic and stationary, rather than as time-varying,which prevents the variability of streamflows from being addressedaccurately in areas lacking sufficient storage. Estimates of the meanannual virgin flow for the Colorado River at Lees Ferry vary from12.5 million to 15.2 million acre-feet depending on the assump-tions (in this case, the period of the historical record) used.

In general, the use of aggregated data, in the form of regionaland basinwide averages, will mask the local and cumulative down-stream effects of development on water availability. Such data donot provide information about either the seasonal variability ofstreamflows and demands or the relative positioning and henceinterrelationships among users. These factors are important for iden-tifying potential competition for water, especially in areas wherewater is scarce and subject to development pressures, as will oftencharacterize locations for synfuels development.

obtaining water difficult, the user will be moreassured of a certain supply once a right is ob-tained. s

Missouri River Basin

The magnitude of the institutional, legal, politi-cal, and economic uncertainties in the MissouriRiver Basin, together with the need for major newwater storage projects to average-out seasonaland yearly streamflow variations, preclude an un-qualified conclusion as to the availability of sur-face water resources for synfuels development.bGround water resources are not well understoodin the basin, but are not likely to be a primarysource of water for synfuels.

Major coal deposits for synfuels developmentin the Missouri River Basin lie within and adja-cent to the Yellowstone River subbasin. The avail-ability of water for synfuels from the Yellowstonesubbasin, however, could be constrained by theprovisions of interstate compacts, i.e., theYellowstone River Compact. For example, atpresent all signatory States must approve anywater exports from the basin (e.g., to the coal-rich Belle Fourche/Gillette area where water isscarce). Although export approval procedures arenow being challenged in court and States havebegun to modify approval procedures, such ap-provals are likely to take some time. Furthermore,additional storage would likely be required todevelop fully the compact allocations.

Federal reserved water rights are often seniorrights and have the potential of preempting cur-rent and future uses. These rights, however, haveyet to be quantified and are a major source ofuncertainty for water planning. The largest singlecomponent of Federal reserved rights are Indianwater rights. There is a general lack of quantitativedata concerning Indian water rights because ofpolitical controversy over which jurisdictionsshould be adjudicating the claims, varying inter-pretations of the purposes for which water rightsreservations can be applied, and ongoing litiga-tion.7

‘Ibid.61~id.

The only “official” Government estimates of Indian reserved wa-ter rights project depletions (i.e. requirements) of 1.9 million acre-

(continued on next page)


Other major uncertainties that could effect theavailability of water for synfuels concern State wa-ter allocation laws. For example, Montana hasestablished instream flow reservations in thelower-Yellowstone River of 5.5 million AFY toprotect future water quality and wildlife. Over500,000 AFY have also been reserved in the basinfor future municipal and irrigation use. Additionalstorage would be required to meet these reserva-tions during years of low flow, but Montana Stateofficials generally do not advocate the construc-tion of new mainstem storage, even if instreamflow shortages were to occur otherwise, as thiswould interfere with the free-flowing nature ofthe river.8 9 No determination has yet been madeas to how these instream flow reservations wouldbe accommodated under the Yellowstone Com-pact.

The transferring of water rights from existing(e.g., agricultural) to new (e.g., synfuels) uses inMontana is subject to administrative restrictionsunder primarily the 1973 Water Use Act, andState environmental and facility siting acts.10 Be-cause of these restrictions, water rights are notfreely transferable from existing users, and, in ef-fect, there is presently no economic market forrights transfers.

State water laws and statutory provisions inother Upper Basin States similarly could constrainwater rights transfers to synfuels.11 As examples,water for irrigation takes precedence in theseStates over water for energy development, andthe “public interest” is to be explicitly considered

feet for the year 2020 in the Yellowstone. (U.S. Department of in-terior, Water for Energy Management Team, Report on Water forEnergy in the Northern Great Plains With Emphasis on theYe//owstone River Basin, January 1975.) A lower estimated valueof 0.5 million acre-feet appeared in a 1960 background paper (fora larger framework study of the Missouri River Basin) by the Bureauof Reclamation. For a detailed discussion of Indian reserved waterrights, the reader is referred to Constance M. Boris and John V.Krutilla, Water Rights and Energy Development in the YellowstoneRiver Basin, Resources for the Future, 1980.

BWright Water Engineers, Inc., op. cit.9Personal communications, Department of Natural Resources and

Conservation, State of Montana.IOFor a detailed discussion of State water allocation laws See Grant

Gould, State Water Law in the West: /mp/ications for Energy DeveLopment, Los Alamos Scientific Laboratory, Los Alamos, N. Mex.,January 1979.

I I Ibid.

in approving water allocations. Alternatively,other laws could work to the disadvantage ofnonenergy sectors, such as navigation in the Mis-souri region under the Federal Flood Control Actof 1944 (33 USC 701 -(b)).

Many of the water availability issues in the Mis-souri River Basin cannot be adequately evaluatedbecause of a lack of supporting data and case lawinterpretations. Figure 27 illustrates the possiblemagnitude of uncertainty by superimposing themajor projected consumptive uses (excludingsynfuels) onto the availability of water in theYellowstone River. As can be seen, assuming alow total estimated demand growth scenario, de-mands would not be met in a dry year withoutadditional storage. Assuming a high-growth sce-nario, not only would demands not be met in adry year without storage, but they would also ex-ceed the average annual flow with additionalstorage.

Upper Colorado River Basin

Although water may not be available in certaintributaries and at specific sites, sources of water

Figure 27.–Streamflows and Projected IncreasedIncremental Water Depletions, Yellowstone River

at Sidney, Mont.

1975 1985 2000

Year

SOURCE: “WaterAvailability for Synthetic Fuels,” Wright Water Engineers, Inc.,contractor report to OTA, June 5, 1981.

Ch. n-Water Availability for Synthetic Fuels Development . 287

generally exist in the Upper Colorado River Basinthat could be made available to support OTA’slow and high estimates of oil shale developmentthrough at least 1990. * However, the institution-al, political, and legal uncertainties in the basinmake it difficuIt to determine which sourceswould be used, the actual amount of water thatwouId in fact be made available from any sourceto support synfuels development, and thus thewater resource impacts of using any source forsynfuels on other water users. Until major com-ponents of these uncertainties are analyzed quan-titatively and start to become resolved, the ex-tent to which synfuels production can be ex-panded beyond a level of several hundred thou-sand barrels/day (i. e., about 125,000 AFY) can-not be estimated with confidence.12

One potential source of water supply for syn-fuels is storage from Federal reservoirs. For exam-ple, approximately 100,000 AFY could be madeavailable for synfuels from two Federal reservoirson the western slope of Colorado (Ruedi andGreen Mountain). However, the amount of wateravailable is uncertain because of questions re-garding firm yields, contract terms for water sales,which purposes are to be served by the reser-voirs, competing demands, the marketing agent,and operating policy.

Under State water laws, water rights throughoutthe basin—in Colorado, Utah, and Wyoming—can generally be transferred (e. g., from agricul-ture) via the marketplace (i. e., sold) to synfuelsdevelopers who can afford to pay a relatively highprice for water.13 The degree to which developersrely on such transfers will determine the subse-quent economic and social impacts on the usersbeing displaced and, in turn, on the region. * Thetransfer process, however, is time-consuming and

*The low estimate for shale oil production in 1990 (see ch. 6)

implies a range of annual water use of 20,000 to 48,000 acre-feet;

the high estimate implies a range of 40,000 to 96,000 acre-feet. By

2000, annual water requirements would be, respectively, 50,000

to 120,000 acre-feet, and 90,000 to 216,000 acre-feet.I Zwright Water Eng ineers , ! nc., OP. cit.

‘JGould, op. cit.

“ irr igation requirements are determined by many factors, includ-

ing climate, crop, irrigation methods, etc. Assuming that agriculture

consumes 1.5 to 2.5 acre-feet/acre in the Rocky Mountain area,

an average oil shale plant consuming 8,500 AFY would need toacquire water rights applicable to about 3,4oo to 5,7oo irrigatedareas.

legally cumbersome, is constrained under Statewater law by the nature of the original right, andis subject to political and legal challenge.

Some provisions of the laws and compacts gov-erning water availability to the States within thebasin will not be tested and interpreted untilwater rights in the basin are fully developed. Forexample, procedures and priorities have not yetbeen developed for limiting diversions among theUpper Basin States when downstream com-mitments to the Lower Basin, under the ColoradoRiver Basin Compact, cannot otherwise be met.There is also controversy about whether the Up-per Basin States as a whole will be responsiblefor providing any of the 1.5 million AFY commit-ment to Mexico under the Mexican Water Trea-ty of 1944-45. Individual States within the basin,such as Colorado, have generally not yet devel-oped procedures and priorities for internally ad-ministering their downstream delivery com-mitments for when the basin becomes fullydeveloped; thus, the impacts of a State’s alloca-tion of available water to individual subbasins andusers within that State, such as synfuels, cannotyet be determined. State water law also general-ly evolves through individual court cases, so thatthe cumulative effects of development are notknown.

There are generally no institutional or financialmechanisms for obtaining water for synfuels, ei-ther through conservation or through increasedefficiency in water use in other sectors, as in otherparts of the country. In Colorado, for example,changes in agricultural practices to increase waterefficiency are likely to be challenged legally, sincedownstream water rights appropriators are en-titled to return flows resulting from existing albeitinefficient practices. It has been reported thatbasin exports for municipal uses could be re-duced by as much as 200,000 to 300,000 AFYwith improved water use efficiency .14

Other uncertainties that affect water availabilityfor synfuels in the area include: Federal reservedwater rights (e. g., for the Naval Oil Shale Reserve

IAoffice of the Executive Director, Colorado Department of Natu-ral Resources, The Availability of Water for Oil Shale and Coal Gasi-fication Development in the Upper Colorado River Basin, Upper

Colorado River Basin 13(a) Assessment, October 1979.


at Anvil Points, Colo.) have not yet been quanti-fied; storage would have to be provided in theWhite River Basin (where the Uinta and PiceanceCreek oil shale reserves are located) but primereservoir sites are located in designated wilder-ness areas; there is as yet no compact betweenColorado and Utah apportioning the flows of theWhite River; and in Colorado, in order to developmuch of the deep ground water in the PiceanceBasin, oil shale developers must prove that theground water is nontributary, for which data areoften lacking and difficult to obtain. The resolu-tion of the uncertainties in the Upper Coloradocould limit large-scale synfuels growth as illus-trated in table 82, but “even at these highly ag-gregated levels for the entire Upper ColoradoRiver Basin, the confidence limits or ranges thatare placed on estimates of water availability areso broad that they tend to (overshadow) theamount of water needed for synfuelsdevelopment.” 15

Wright Water Engineers, Inc., op. cit., p. IV-38.

Table 82.—Preliminary Quantification ofUncertainties With Respect to Water Availability

in the Upper Colorado River Basin

Annual amount available for consumption(millIons of acre-feet) a

12.5 -15.2

Subtract 7.5

5.0 -7.7Subtract 0.75

4,25-6.95Subtract .65

Estimates of mean annual flowof the Colorado River atLees Ferry

Required delivery to the LowerBasin

Estimate of the Upper Basin’sMexican Treaty obligation

Estimated annual reservoirevaporation from FlamingGorge, Lake Powell, and theCurecanti Unit Reservoirs

Total 3.60-6.30

Annual projected consumptive demands(millions of acre-feet) In 2000 b

Total 4.10-4.78 (excluding synfuels)Total 4.15-4.90 (including OTA low estimates for oil

shale c)Total 4.19-5.00 (including OTA high estimates for oil

shale c)ams not nl~e ~lowmces for the quantlflcatlon of Federal reserved water rights

cleims (the Naval Oil Shale Reserve at Anvil Points has claimed, for exempie,200,01Xl AFM. the effect of Dotentlai environmental constraints (e.ci.. saiinityconiroi, prot~tion of endangered species), or the availability of F~er~’storage.

bEstlmates ~ for 2ocm and exclude synfuels development (@iorado @Partment

of Natural Resources, Section 13(a) Assessment of the Upper Coiorado RiverBasin; 1975 estimate = 3.12 maf). instream uses are not inciuded.

cThe low estimate for shale oil production in 1990 (see ch. 6) imPlies a ran9eof annuai water use of 20,000 to 48,000 acre-feet; the high estimate implies arange of 40,000 to 98,tXXl acre-feet. By 2000, annual water requirements wouidbe, respectively, 50,000 to 120,000 acre-feet, and S,000 to 216,000 acre-feet.

SOURCE: OTA based on Wright Water Engineers, Inc.

Index

Aamco Transmissions, 202alcohols, 99, 269-270American Motors (AMC), 107, 202

Arizona Public Service Co., 249

Arthur Anderson & Co., 199

asphalt, 78

automobile industry

auto repair facilities (table), 201capital intensiveness of, 42, 48, 60conduct of manufacturers, 197-198economic impacts of change in, 195-203employment in, 220-223financing of capital programs in, 130-131links to foreign companies, 196-197

location of parts-supplier plants (table), 224

manufacturing structure of, 196-197

occupational and regional issues in, 223-225

problems of, 40-41prospects for, 202-203rate of change in, 133

sales and service segment of, 200-202structural readjustments in, 7, 40suppliers for, 198-200vertical integration in, 133

Booz-Allen & Hamilton, 199Brayton engine, 144-145

Bureau of Labor Statistics, 220

Bureau of the Census, 220

CAFE (see Corporate Average Fuel Efficiency)Carter administration, 50Chrysler Corp., 12, 53, 130, 196, 197, 198, 202, 221coal

acid drainage from mining, 251-252aquifer disruption from mining, 252direct liquefaction of, 163-165indirect liquefaction of, 161-163

mining for synfuels, 204-205

occupational hazards of mining, 252-253

subsidence from mining, 252

Colony Oil Shale Project, 232

Colorado Department of Local Affairs, 232Colorado River Basin Compact, 287

Colorado West Area Council of Governments, 228Congress

acts of (see legislation)Congressional Budget Office (CBO), 140

Congressional Research Service (CRS), 63, 159

policy options for (see policy options)proposals on environmental and safety regulation, 56Senate Committee on Commerce, Science, and Transpor-

tation, 31water resources and, 58

conservation, 83, 184, 188-198Corporate Average Fuel Efficiency (CAFE), 3 ( 9, 13, 30, 31,

36, 41, 50, 55-56, 60, 61, 105 { 106, 120

Data Resources, Inc., 202Delphi survey, 199Department of Commerce, 223Department of Energy (DOE), 4, 9, 10, 43, 44, 54, 57, 58,

96, 183, 253, 255, 262, 263Department of Housing and Urban Development (HUD),

59, 62Department of Labor, 222Department of Transportation, 129, 135, 196, 199, 220, 221

222, 223, 243Department of the Treasury, 43

Eaton Corp., 199Economic Development Administration, 59electricity, 186electric vehicles, 3, 41, 71, 74, 219, 250

battery-electric and hybrid, 117-120cost compared to synfuels (table), 77cost to consumers, 5, 141-142effects on fuel consumption, 128-129, 150, 153and electric utilities, 149-153, 245-246environmental effects of, 7, 24, 84, 85, 245-248impact on oil imports, 5occupational and public health risks of, 247-248prospects for, 24, 94-95resources requirements for, 246-247safety of, 24-25, 247

Energy information Administration (EIA), 22, 30, 37,69, 73,184, 185, 186, 189

Energy Mobilization Board, 57Engineering Societies’ Commission on Energy (ESCOE), 170Environmental Protection Agency (EPA), 10,41, 57, 58,93,

96, 121, 124, 239, 254, 262, 263, 264, 270environment, health, and safety, 58, 60, 84

air quality, 244-245, 254-256coal liquefaction and, 253-261effects of auto fuel conservation on, 237-245effects of diesel fuel on, 84effects of energy alternatives on, 84-85effect of fuel efficiency on, 15effect of policy options on, 56-58electric vehicles and, 7, 24, 84, 85, 247-248emissions characteristics of alternative engines (table), 245future trends in auto safety, 240-242gaseous emissions and controls (table), 271gaseous streams, components, and controls (table), 274impacts of fuel efficiency and synfuels compared, 83-85liquid stream sources, components, and controls (table),

275liquid waste stream sources, components, and controls

(table), 272small v. large cars, 7, 92-93, 237-242solid waste sources, components, and controls (table), 275solid waste stream sources, components, and controls

(table), synfuels and, 7, 248-270environment (see environment, health, and safety)EPA (see Environmental Protection Agency)

291


ESA (see legislation, Energy Security Act)ethanol, 179, 263-269EVs (see electric vehicles)Exxon, 30, 186, 208Exxon Donor-Solvent (EDS) process, 20, 164

Fatal Accident Reporting System (FARS), 238, 241, 242Fischer-Tropsch process, 163Ford administration, 46Ford Motors, 12, 15, 130, 132, 196, 197, 202, 221fuel consumption

estimates for 1985-2000, 126-130of passenger cars, 126-128of vehicles other than passenger cars, 129-130

fuel efficiencyaccessories and, 149advantages of, 60aerodynamic drag and, 148-149automobile technologies and, 108-120by car class size, 93, 123-126costs to consumers, 79-80, 139-141costs of improving, 12-14, 75-77, 130-142demand for, 93-94economic impacts of, 14-15, 71, 87-89effects on communities, 86effects on employment, 86, 87-88effects on environment, health, and safety, 15features of, 38-40Federal role in, 37-38growth of, 9, 10, 11-12, 36, 40-41, 105-107, 120-126,

142-149and health of the auto industry, 89-90investments in, 131-135policy background of, 41-42potential in 1995 (table), 56potential for a 75-mpg car, 90-91rolling resistance and, 149social impacts of, 15, 86, 219-225stimulation of, 3tradeoffs with safety and emissions, 113-115vehicle weight and, 146-148

fuel gasesfrom biomass, 168

fuel switching, 83, 184, 185-188cost estimates of (table), 78

gasifiers, 163gasoline

from methanol, 163taxes on, 47

gas turbine, 144-145, 244General Agreement on Tariffs and Trade (GATT), 49, 51General Motors, 12, 15, 90, 130, 133, 139, 196, 197, 198,

199, 202, 221, 223, 242

hazards (see environment, health, and safety)health (see environment, health, and safety)Honda, 53

Insurance Institute for Highway Safety (IIHS), 238, 239,240

legislationClean Air Act, 58, 244, 263Clean Water Act, 263Economic Recovery Tax Act, 131Energy Policy and Conservation Act, 30, 35,41, 105, 106,

108Energy Security Act, 31, 35, 42, 53EPCA (see Energy Policy and Conservation Act)Federal Coal Mine Health and Safety Act, 253Federal Flood Control Act, 286Federal Nonnuclear Research and Development Act, 43Fuel Efficiency Act, 41Fuel Use Act, 35, 62National Energy Act, 42Powerplant and Industrial Fuel Use Act, 30-31Resources Conservation and Recovery Act, 260, 263Safe Drinking Water Act, 263Surface Mine Control and Reclamation Act (SMCRA), 250,

251, 252Trade Act, 59, 221Trade Expansion Act, 221Water Use Act, 286

liquefied petroleum gas (LPG), 78

methanol, 163, 179, 209, 269Mexican Water Treaty, 287Michigan Manufacturers Association, 199Midas Muffler, 202Middle East War, 67Moody’s Investors Service, Inc., 224

National Accident Sampling System, 239National Crash Severity Study, 239National Electric Reliability Council, 187National Highway Traffic Safety Administration (NHTSA),

15, 41, 84, 92, 237, 238, 240, 243National Petroleum Council, 186national security, 37National Synfuels Production Program (NSPP), 42, 43, 56New England Electric Co., 62Nissan Motors, 223NSPP (see National Synfuels Production Program), 42

Occupational Safety and Health Administration (OSHA), 10,41, 97, 263, 264

Office of Surface Mining, 10Office of Technology Assessment (OTA), 3, 4, 5, 6, 8, 35,

36, 37, 42, 43, 45, 47, 53, 60, 69, 70, 71, 74, 75, 77,78, 88, 94,98, 113, 121, 122, 123, 124, 131, 136, 140,170, 179, 185, 189, 190, 233, 237, 238, 240, 241, 250,254, 262, 266, 268, 282, 287

oilcost of reducing consumption of, 23-24, 74-79decline in U.S. production of, 30, 35depletion of reserves of, 67displacement of imported, 8-10, 35-37, 44-61and foreign policy, 69

Index . 293

imported, 3, 5, 67-69import levels in absence of synfuels and fuel-efficiency

increases (table), 70and military outlays, 69production of gas and, 74-75quotas on imported, 49-50reductions in consumption of (table), 70refining capacity (table), 159refining processes, 157-158from shale, 99, 209stationary uses of, 5, 22-24, 78, 183-191supply interruptions, 68-69taxes on imported, 46-47in transportation (table), 184

Organization of Petroleum Exporting Countries (OPEC), 46OTA (see Office of Technology Assessment)

petrochemicals, 78petroleum (see oil)policy options, 35-63

bounties for higher fuel economy, 51conservation, 62dimensions of, 44-45economywide, 45, 48-49effect on environment, health, and safety, 56-58fuel conversion, 62general, 62loan guarantees and grants, 53-54methanol promotion, 52, 54-55purchase and price guarantees, 54purchase taxes and subsidies, 50-51registration taxes on automobiles, 51-52regulation, 55-56research and development, 48-49sector-specific, 45-46social costs and benefits of, 8, 58-60subsidies and guarantees, 52-53taxation of fuels, 8, 46-48trade protection, 49-50

Rand Corp., 171research and development, 48-49, 130-131, 137, 139, 144,

200costs in 1980, 130

Robogate Systems, Inc., 221

safety (see environment, health, and safety)SFC (see Synthetic Fuels Corp.)shale, 165-167, 204-205, 266-268Solar and Conservation Bank, 62Solar Energy Research Institute (SERI), 187, 189, 190still gas, 78Stirling engine, 144-145synfuels

advantages of, 60alcohols, 99from biomass, 85, 167-168, 179-180, 268-270

capital intensiveness of, 42, 48, 57, 60characteristics of (table), 265from coal, 99-100, 161-165, 204-205commercial-scale process units (CSPUs), 43, 44compatibility with existing end uses, 99-101constraints on development of, 42, 176-177, 209-216costs of, 17-18, 81-83, 168-175definition of, 157demonstration projects, 43-44development of an industry for, 175-180economic impacts of, 7, 18-19, 87-89, 203-216effects on communities, 86and employment, 86, 87-88, 226-227, 257environment, health, and safety effects, 19-21, 85, 95-98,

254-256, 258-261environmental management of, 261-264ethanol, 167, 268-269from fossil sources, 177-179fuel gases, 165growth management, 233impacts on the private sector, 230-231impacts on the public sector, 231-233industrial structure of, 204-205investment costs, 77-78and local population growth, 227-230, 257manufacturing processes, 16, 160-168methanol, 115-117, 142, 167, 269oil import reductions and, 4-sfrom oil shale, 165-167, 266-268policy background of, 42-44principal products (table), 160production rate, 9, 16-17, 36-40, 42projects on Indian lands, 233project types, 43-44regional distribution of income from, 89social impacts of, 7, 19, 86, 225-233transportation, wholesaling, and retailing, 207water for, 7, 31, 58, 98-99, 256-257, 279-288

Synthetic Fuels Corp. (SFC), 3, 8, 18, 31, 43, 44, 53, 58synthetic fuels (see synfuels)

tax credits, 48transmissions, 145

United Auto Workers, 221

Volkswagen of America, 53, 107, 221, 241

water availability, 21-22, 31Eastern river basins, 22, 283-285Indian water rights, 285Missouri River Basin, 22, 285-286Upper Colorado River Basin, 22, 286-288Western river basins, 285

Yellowstone River Compact, 285, 286

U . S , GOVERNMENT PRINTING OFFICE : 1982 0 - 98-281 : QL 3

Increased Automobile Fuel Efficiency and Synthetic Fuels

Documents

Transcript of Increased Automobile Fuel Efficiency and Synthetic Fuels