nuclear power in an age of uncertainty

8/14/2019 Nuclear Power in an Age of Uncertainty

http://slidepdf.com/reader/full/nuclear-power-in-an-age-of-uncertainty 1/298

Nuclear Power in an Age of Uncertainty

February 1984

NTIS order #PB84-183953



Recommended Citation:

Nuclear Power in an Age of Uncertainty (Washington, D. C.: U.S. Congress, Office

of Tec hno log y Assessme nt, OTA- E-216, Feb ruary 1984).

Library of Congress Catalog Card Number 84-601006

For sale by the Supe rintend ent o f Docum ents,U.S. Government Printing Office, Washington, D.C. 20402



Foreword

This rep ort respond s to a req uest by the House Co mm ittee on Sc ience a nd Tec h-nolog y, with the endo rsem ent o f the Sena te Co mm ittee on Energy and Na tural Resource s,to a ssess the future of nuc lear powe r in this c ount ry, and ho w the tec hnolog y and in-

stitutions might be changed to reduce the problems now besetting the nuclear option.The repo rt builds on a n ea rlier OTA report, Nuclea r Powerplant Standa rdizat ion, a ndcomplements a current report, Managing Commerical High-Level Radioactive Waste.

The p resent nuc lear era is d raw ing to a c lose. The c over of th is report show UnitI of the Washington Public Pow er Sup p ly System, wh ich w as inde finite ly, perha ps per-manently, deferred even though it was 60 percent complete and $2.1 bil l ion had beeninvested . This p lant a nd oth ers suc h a s Zimme r and Ma rble Hill ep itom ize th e d iffic uItiesfac ing the nuc lear industry. It is impo rtant to rememb er, how ever, that other nuclea r

p la nts ha ve b ee n very suc c essful a nd p rod uc e relia ble, low c ost elec tric ity. The futureof nuclear power poses a complex dilemma of policy makers. It has advantages thatmay prove crucial to this Nation’s energy system in the coming decades, but at pres-ent it is an option that no electric utility would seriously consider.

OTA exam ined que stions of d ema nd growth, co sts, reg ulation, and pub lic ac c ep t-ance to evaluate how these factors affect nuclear power’s future. We reviewed researchdirections which could improve conventional l ight water reactor technology and op-portunities to develop other types of reactor concepts that might enhance safe and

reliab le o pe rat ion. In add ition, the c ruc ia l role of utility ma nage ment in constructingand op erating nuc lear pow erplants is examined at length. The c ontroversy ab out nuc learsafety regulation is also analyzed, and is presented with a review of current proposalsfor regulatory reform. Finally, the study discusses policy approaches that could assista revival of the nuclea r option should tha t b e a c hoice of Co ngress.

In the c ourse of this assessme nt, OTA d rew on the expe rience of ma ny o rga niza -

tions and individua ls. In p a rticular, we app rec iate the ge nerous assistanc e o f our d is-tinguished advisory panel and workshop participants, as well as the efforts of the proj-ect’s consuItants and contractors. We would also like to acknowledge the help of thenumerous reviewers who gave their time to ensure the accuracy and comprehensivenessof this rep ort. To a ll of the a bove go es the g ra titude o f OTA, and the p ersona l tha nksof the projec t staff.

Director



The Future of Conventional Nuclear Power Advisory Panel

Jan Beyea

National Audubon Society

Richa rd DeanGeneral Atomic s Corp.

George DilworthTenne ssee Va lley Authority

Linn DraperGulf States Utilities

Fritz Heimann

General Electric Co.

Leonard HymanMerrill Lync h, Pierce, Fenner & Smith

Robert KogerNorth Carolina Utilities Commission

Myron KratzerInternational Energy Associates, Ltd.

Byron Lee

Com mo nwea lth Edison

Jessica Tuchma n Ma thews

World Resources Institute

Reviewers

Institute for Nuclear Power Operations

Ralph Lapp, Lapp Inc.Yankee Atomic Electric Co.

Long Island Lighting Co.Public Service Co. of India na

Cincinnati Gas & ElectricPacific Gas & ElectricWashington Public Power Supply SystemOffice of the Governor, Washington State

Oregon Department of EnergyBonneville Power Administration

Portland General Electric

George Rathjens, ChairmanCenter for International Studies

Arthur Porter

Belfountain, Ontario

David RoseMa ssac husetts Institute of Tec hnolo gy

Lee Schipp erLawrenc e Berkely Lab s

James SweeneyEnergy Modeling Forum

Stanford University

Eric Van LoonUnion of Concerned

Observers

James K. Asselstine

Scientists

‘U.S. Nuclear Regulatory Commission

Thomas DillonU.S. Department of Energy

Victor GilinskyU.S. Nuclear Regulatory Commission

Arizona Public Service Co.William R. Freudenburg, Washington State

University

Paul Slovic, Decision Research

Steven Harod, Department of EnergyJohn Tay lor, Elec tric Power Research InstituteAtomic Industrial ForumStone & Webster, inc.Bechtel Power Corp.

Save the ValleyCongressional Research Service

iv



OTA Future of Nuclear Power Project Staff

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

Richard E. Rowberg, Energy and Materials Program Manager

Alan T. Crane , Project Director

Mary Proctor (Economic and Commercial Issues)

Patric ia S. Abe l (Tec hnic al, Ope rational, and Reg ulato ry Issues)

Margaret Hilton (Public Ac ce ptanc e)

Jenifer Robison (Regulations)

Administrative Staff

Lillian Quigg Edna Saunders

Contributors

Warren Donnelly, Congressional Research Service Thoma s Coc hran, Na tural Resourc es Defe nseBarbara G. Levi, Consultant, Colts Neck, N.J. Council

John Crowley, United Engineers & Contractors, Todd LaPorte, Universi ty of Cali fornia at BerkeleyInc. E. P. Wilkinson, Institute for Nuclear Power

Alfred Amerosi, Consultant, Downers Grove, Ill. Operations

James J. MacKenzie, Union of Concerned Arvo Niitenberg, Ontario HydroScientists

Ma rk Mills, Sc ienc e Conc ep tsJoanne Sed er, Offic e of Tec hnolog y A ssessme ntRichard Lester, Massachusetts Institute of

Tec hnolog y

Contractors

Institute fo r Energ y Ana lysis, Oa k Ridg e, Term. Terry Lash, Weehawk en , N.J.

Sanford Cohen & Associates, McLean, Va. NUS Corp., Gaithersburg, Md.

Pickard, Lowe & Garrick, Inc., Washington, D.C. ENSA, Inc., Rockville, Md.

Komanoff Energy Associates, New York MHB Tec hnic al Assoc iate s, San Jose, Calif.Scienc e Conc epts, Vienna, Va. Eliasanne Research, Inc., Washington, D.C.

William J. Lanouette, Washington, D.C.

OTA Publishing Staff

John C. Holmes, Publishing Officer

John Bergling Ka thie S. Bo ss De b ra M . Da tc he r Jo e He nso n

Glenda Lawing Lind a A. Le ahy Che ry l J. Ma nning



Workshop l—The Economic Context for Nuclear Power

Clark Bullard, Chairman University of Illinois

Jack BarkenbusInstitute for Energy Analysis

Charles BergBuckfield, Maine

Carleton D. BurttThe Equitable Life Assurance

United States

Gordon CoreyEvanston, Ill.

John Crowley

Mark MillsScience Concepts

David MoultonEnergy Conservation Coalition

Keith PaulsonWestinghouse Electric Corp.

Doan PhungSociety of the Oak Ridge Associated Universities

Paul Riegelhaupt

Stone & Webster, Inc.Marc RossUniversity of Michigan

United Engineers & Constructors, Inc.

James Edmonds

Institute for Energy Analysis

Victor GilinskyU.S. Nuclear Regulatory Commission

Eric HirstOak Ridge National Laboratory

Leonard HymanMerrill Lynch, Pierce, Fenner & Smith

Henry KellyOffice of Technology Assessment

Robert Koger

North Carolina UtilitiesCharles KomanoffKomanoff Energy Associates

Mark LevineLawrence Berkeley Laboratory

Lynn MaxwellTennessee Valley Authority

Philip SchmidtUniversity of Texas at Austin

Milton SearlElectric Power Research Institute

James SweeneyEnergy Modeling ForumStanford University

Vince TaylorRichford, Vt.

Jon VeigelAlternative Energy Corp.

James Walker

Office of the commissionersAlvin WeinbergOak Ridge Associated Universities

John WilliamsAnnapolis, Md.

Workshop n-Technology, Management and Regulation

Harold Lewis, Chairman University of California at Santa Barbara

James K. Asselstine

U.S. Nuclear Regulatory Commission

Lynne BernebeiGovernment Accountability Project

Dale BridenbaughMHB Technical Associates

Robert J. BudnitzFuture Resources Associates, Inc.

Sanford C. CohenSC&A, Inc.

John CrowleyUnited Engineers & Constructors, Inc.

Peter R, DavisIntermountain Technologies, Inc.

Thomas G. Dignan, Jr.Ropes & Gray

Richard EckertPublic Service Electric & Gas CO.

Colin R. FisherGeneral Atomics Corp.

Arthur FraasInstitute for Energy Analysis

Jerry GriffithU.S. Department of Energy

Saul LevineNUS Corp.

Fred LobbinSC&A, Inc.

James MacKenzieUnion of Concerned Scientists

Mark MillsScience Concepts


Daniel PrelewiczENSA, Inc.

George RathjensCenter for International Studies

Robert RenuartBechtel Power Corp.

David RoseMassachusetts Institute for Technology

vi



Alan RosenthalU.S. Nuclear Regulatory Commission

R. P. SchmitzBechtel Power Corp.

Irving SpiewakInstitute for Energy Analysis

Sharon ThompsonSC&A, Inc.

Alvin WeinbergInstitute for Energy Analysis

Abraham WeitzbergNUS Corp.

Bertram WolfeGeneral Electric

Edwin ZebroskiInstitute for Nuclear Power Operations

Robert E. UhrigFlorida Power & Light

Workshop Ill—Institutional Changes and Public AcceptancesStephen Gage, Chairman Northbrook, III.

James Asselstine

U.S. Nuclear Regulatory Commission

Jack BarkenbusInstitute for Energy Analysis

Edward A. BrownNew England Power Service Co.

Michael FadenUnion of Concerned Scientists

Eldon GreenbergGalloway & Greenberg

Jay Hakes

Office of the Governor of FloridaFritz HeimannGeneral Electric Co.

Roger Kasperson

Clark University

Terry LashWeehawken, N.J.

Utility Executives Attending Workshop Ill

E. F. CobbChief of Nuclear Planning and ControlGeorgia Power Co.

Austin J. Decker II

Manager, Operational Analysis andTraining

Power Authority of the State of New York

James W. DyeExecutive Vice PresidentLong Island Lighting Co.

Jack Fager

Vice President, Engineering andConstruction

Louisiana Power & Light Co.

Wendell KelleyChairman and CEOIllinois Power Co.

T. C. NicholsSenior Vice President, Power OperationsSouth Carolina Electric & Gas Co.

Warren OwenSenior Vice PresidentDuke Power Co.

John D. LongIndiana University

R. P. MacDonaldAlabama Power Co.

Roger MattsonU.S. Nuclear Regulatory Commission

Eugene W. MyerKidder Peabody & Co., Inc.


Irving SpiewakInstitute for Energy Analysis

Linda Stansfield

N.J. League of Women VotersLinda TaliaferroPennsylvania Public Utilities Commission

Alvin WeinbergInstitute for Energy Analysis

Hugh G. ParrisManager of PowerTennessee Valley Authority

Cordell Reid

Vice PresidentCommonwealth Edison Co

Sherwood H. Smith, Jr.Chairman and PresidentCarolina Power & Light Co.

Norris L. StampleySenior Vice President for NuclearMississippi Power & Light Co.

Richard F. WalkerPresident and CEOPublic Service Co. of Colorado

William O. WhittExecutive Vice presidentAlabama Power Co.

Russell C. YoungdallExecutive Vice President

Consumers Power Co.

vii



Chapter

Over

Contents

view . . . . , . . . , . . . . . . . . . . . . . . . . . . . . . . . ,

1. Introduc tion : The Seven-Side d Co in . . . . . . . . . . .

2. Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

3. The Unc erta in Financ ial and Ec onomic Future . . . . . . . . . . . . . . . . . .

4. Alternative Reactor Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

5. Management of the Nuclear Enterprise. . . . . . . . . . . . . . . . . . . . . . . . .

6. The Regula tion o f Nuc lea r Power . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

7. Surviva l of the Nuc lear Ind ustry in the United Sta tes and Ab road . . .

8. Pu

9. Po

lic Att itudes Tow ard Nuc lea r Pow er . . . . . . . . . . . . . . . . . . . . . . . .

icy Options. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

ossary . . . . . . . . . . . , . . . . . . . . . . . . . . . . . .Appendix: Acronyms and G

Index . . . . . . . . . . . . . . . . . .

Page

ix

3

13

29

83

113

143

179

211

251

283

289

ix



OVERVIEW AND FINDINGSWithout significant changes in the technology, management, and level of public acceptance,

nuclear power in the United States is unlikely to be expanded in this century beyond the reactorsalready under construction. Currently nuc lear p owe rplants present too ma ny financ ial risks as a resultof uncertainties in electric demand growth, very high capital costs, operating problems, increasing reg-

ulatory requirements, and growing public opposition.

If all these risks were inherent to nuclear power, there would be little concern over its demise.However, enough utilities have built nuclear reactors within acceptable cost limits, and operatedthem safely and reliably to demonstrate that the difficulties with this technology are not insurmount-able. Furthermore, there are na tional policy reasons why it c ould b e highly d esirab le to ha ve a nuclea r

op tion in the future if present problems ca n be overco me. Demand for elec tricity could grow to a levelthat would m and ate the co nstruction of ma ny new pow erplants. Unce rtainties over the long- term en-vironmental acceptability of coal and the adequacy of economical alternative energy sources are alsogreat and underscore the potential importance of nuclear power.

Some of the problems that have p lagued the present generation of reac tors are due to the imma turity

of the technology, and an underestimation by some utilities and their contractors of the difficulty ofmana ging it. A major com mitment was ma de to build large reac tors before any had b een c omp leted.Many of these p rob lems should not reoc cur if new reac tors are ordered. The c hange s that have bee napplied retroactively to existing reactors at great cost would be incorporated easily in new designs. Safety

and reliability should be better. It is also likely that only those utilities that have adequately managedtheir nuclear projects would consider a new plant.

While imp ortant a nd essential, these improvements by them selves are p robab ly not a deq uate tobreak the present impasse. Problems such as large cost overruns and subsequent rate increases, inade-

quate quality control, uneven reliability, operating mishaps, and accidents, have been numerous enoughthat the confidence of the public, investors, rate and safety regulators, and the utilities themselves istoo low to b e restored ea sily. Unless this trust is restored, nuclear power will not be a credible energyoption for this country.

It appe ars possible, however, that ad ditional improvements in technolog y a nd the wa y nuclea r power

is ma nag ed and regulated m ight be sufficient to restore the req uired co nfidenc e.Technological im-provements, while insufficient by themselves, can nevertheless be very important in that effort. Oneapproach would be to focus research and development (R&D) on improving current light water reactor

(LWR) designs. The g oa l would be standa rdized de signs rep resenting a n op timal b alance of c osts, safe-

ty, and op erab ility. Private industry is unlikely to und ertake all the R&D nee de d, so a Fed eral p resenc eis probably necessary.

It is also possible, however, that even greatly improved LWRs will not be viewed by the publicas acceptably safe. Therefore, R&D on alternative reactors could be essential in restoring the nuclearoption if they have inherently safe characteristics rather than relying on active, engineered systemsto protect against accidents. Several c onc epts app ear p romising, including the high tem perature ga s-

c oo led reac tor (HTGR), the PIUS reac tor, and hea vy w ate r rea c tors. Suc h R&D should also be direc tedtowa rd de sign a nd d eve loping sma ller reac tors suc h as the mod ular HTGR.

Improvements in areas outside the technology itself must start with the management of existingreactors. The Nuclear Regulatory Commission, as well as the Institute for Nuclear Power Operations,must ensure a commitment to excellence in construction and operation at the highest levels of nuclearutility management. Improved training programs, tightened proce dures, and heightened awa reness ofopportunities for improved safety and reliability would follow. If some utilities still prove unable to im-prove sufficiently, consideration could be given to the suspension of operating licenses until their nuclear

operations reflect the required competence, perhaps by employing other utilities or service companies,

Similarly, c ertifica tion of utilities or op erating c om pa nies c ould b e c onside red a s a p rerequisite for pe r-

mits for new p lants in order to guarantee that only q ulaified c ompa nies would have respo nsibility. Theseare drastic steps, but they m ay be warranted be ca use a ll nuclea r reac tors are hostag e,in a sense, tothe poorest performing units. Public acceptance, which is necessary if the nuclear option is to revive,depends in part on al l reactors performing reliably and safely.

xi



Nuclear safety regulation also can be improved even without substantial new legislation.Severalutilities recently have shown that current regulatory procedures need not preclude meeting construc-tion budgets and schedules. The regula tory proc ess, howev er, is mo re unpred icta ble tha n nec essary,and there is no assuranc e that safety and efficiency are b eing o ptimized. Encouraging preapp roved stand-

ardized designs and de veloping p roced ures and the requisite a nalytica l tools for evaluating proposedsafety backfits would help make licensing more efficient without sacrificing safety.

The improvements in technology and operations described above should produce gains in publicacceptance. Additional steps may be required, however, considering the current very low levels ofsupport for more reactors. Add ressing the c onc erns of the c ritics and providing a ssuranc e o f a co n-trolled rate o f nuclear exp ansion c ould e liminate m uch of the reason for public d isaffection. An impor-tant contribution to restoring public confidence could be made by a greater degree of openness byall parties concerned about the problems and benefits of nuclear power.

If progress ca n be ma de in all these a reas, nuclear powe r would be m uch m ore likely to be co n-sidered w hen new elec tric-g eneration ca pa city is needed . Such progress will be d ifficult, however,bec ause many divergent groups will have to wo rk together and substantial technical a nd institutional

change may be necessary.

xi i



Chapter 1

Introduction: The Seven-Sided Coin

A

Photo credit: Atom/c Industrial Forum

Steam generator for a pressurized water reactor (PWR) arrives by barge



Contents

Page

The Polic y Prob lem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3

Nuc lea r Disinc en tives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4

The impasse . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5

The Purpose o f This Stud y . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8

Figure

Figure No. Page

A. The Sev en Side s to the Nuc lea r Debate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7



Chapter 1

Introduction: The Seven-Sided Coin

THE POLICY PROBLEM

The nuc lea r powe r industry is fac ing a periodof e xtreme unce rtainty. No nuc lea r plant now

operating or still under active construction hasbeen ordered since 1974, and every year sincethen has seen a decrease in the total utility com-mitment to nuclear power. By the end of this dec-ade, a lmost a ll the p rojec ts still unde r co nstruc-tion will have been completed or canceled. Pros-pects for new domestic orders during the nextfew yea rs a re d im.

Suc h a b leak set of c ond itions has led som eob servers to c onc lude tha t the ind ustry has nofuture aside from op erating the existing p lants.

Som e c onc lude further tha t suc h an e nd is en-

tirely appropriate because they believe thatnuclea r rea c tors will not b e nee de d d ue to thelow g rowth in dem and for electric ity, and thatthe present p rob lems are large ly a result of theindustry’s own mistakes.

If nuclear power were irrelevant to futureenergy needs, it would not be of great interestto policy makers. However, several other factorsmust be taken into account. While electricgrowth has been very low over the last decade(in fact, it was negative in 1982), there is noassurance that this trend will continue. Evengrow th tha t is quite m od est b y historic a l stand -

ards would mandate new plants—that have notbee n orde red yet–co ming online in the 1990’s.Replacement of aging plants will call for still morenew gene rating c ap ac ity. The industrial ca pa bility

already exits to meet new demand with nuclearreactors even if high electric growth resumes. Inaddition, reactors use an abundant resource. Oilis not a realistic option for new electric-generatingplants because of already high costs and vulner-ability to import disruptions which are likely toincrease by the end of the century. Natural gasmay also be too costly or unavailable for gener-

ating large quantities of electricity.The use of c oal ca n and will be expand ed co n-

siderably. All the plausible growth projectionsconsidered in this study could be met entirely by

co al. Such a dep endenc e, however, would leavethe Nation’ s elec tric system vulnerab le to p rice

increases and disruptions of supply. Furthermore,c oa l ca rries signific ant liab ilities. The c ontinued

combustion of fossil fuels, especially coal, has thepo tential to relea se eno ugh c arbon d ioxide tocause serious climatic changes. We do not knowenough about this problem yet to say when itcould happen or how severe it might be, but thepossibility exists tha t e ven in the ea rly 21st c en-tury it ma y be c om e essential to reduc e sha rplythe use of fossil fuels especially coal. Another po-tentially serious problem with coal is pollutionin the form of acid rain, which already is caus-ing considerable concern. Even with the strictestcurrent control technology, a coal plant emitslarge quantities of the oxides of sulfur and nitro-ge n that a re b elieved to b e the p rima ry sourceof the prob lem. There a re g rea t unc ertainties inour unde rsta nd ing of this p rob lem a lso, b ut thepote ntial exists for la rge -sc a le c oa l co mb ustion

to bec ome unacc eptable or much more expen-sive due to tighter restrictions on emissions.

There are o the r possible a lternat ives to c oa l,of course. Improving the performance of existingpowerplants would make more electricity avail-able without bui lding new capacity. Cogenera-tion and improved efficiency in the use of elec-tricity also are equivalent to adding new supply.These a pp roa c hes are likely to b e the bigg est c on-

tributors to meeting new electric service require-ments over the next few decades. Various formsof solar and geothermal energy also appear prom-

ising. Uncertainties of economics and applicabili-ty of these technologies, however, are too greatto d emo nstrate that the y will ob viate the nee dfor nuclear power over the next several decades.

Therefore, there ma y b e g ood national-po lic yreasons for wanting to see the nuclear option

preserved. However, the purpose of the preced-ing discussion is not to show that nuclear powernecessarily is vital to this Nation’s well-being, Itis, rather, to suggest that there are conditions

3



4 q Nuclear Power in an Age of Uncertainty

under which nuclear power would be the pre- ac c ident o r neg lec t. This rep ort ana lyzes the tec h-ferred c hoice, a nd that the se c ond itions might nical and institutional prospects for the future ofnot be recognized before the industry has lost nuclear power and addresses the question ofits ability to supply reactors efficiently and ex- what Congress could do to revitalize the nuclearseditiously. If the nuclear option is foreclosed, option if that should prove necessary as a nationalit should at least happen with foresight, not by policy objective.

NUCLEAR DISINCENTIVES

No efforts-whether by Government or the in-dustry itself–to resto re the vita lity o f the industrywill succeed without addressing the very real

p rob lems now fac ing the te c hnolog y. To illustratethis, consider a utility whose projections showa need for new ge nerat ing c ap ac ity by the mid-1990’s. in comparing coal and nuclear pIants,

current estimates of the cost of power over thepIant’s l ifetime give a small advantage–perhaps

10 percent–to nuclear. Fifteen years ago, thatadvantage would have been decisive. Now, how-

eve r, the utility manage rs c an see d iffic ulties a tsome current nuclear projects which, if repeatedat a new p Iant, would eliminate any p rojecte dc ost a dvanta ge and seriously strain the utility:

q

q

q

The c ost p rojections ma y b e inac c urate .Som e p lants a re being finished at m any time stheir originally estimated cost. Major portionsof a plant may have to b e rebuilt be ca useof d esign inad eq uac y, slop py wo rkmanship,or regulatory changes. Construction lead-

times can approach 15 years, leaving the util-ity da ngerously exposed financ ially. Thesevere cash flow shortages of the Washing-ton Public Power Supply System (WPPSS) arean extreme examp le o f this problem.

Demand growth may continue to fall belowprojections. A utility may commit large sumsof c ap ital to a p lant only to find p art waythroug h c onstruction tha t it is not ne ed ed .If the plant has to be canceled, the utility andits shareholders must absorb all the losseseven though it looked like a reasonable in-vestment at the b eg inning. The long c on-

struction schedules and great capital de-mands of nuclear pIants make them especial-ly risky in the light of such uncertainty.The Nuc lea r Reg ulato ry Com mission (NRC)continues to tighten restrictions and mandate

q

q

q

q

major changes in plant designs. Although therea sons for these c ha nge s often a re va lid,

they lead to increases in costs and schedulesthat are unpredictable when the plant isordered. In addition, the paperwork andtime demands on utility management aremuch greater burdens than for other gener-ating options.

Once a plant is completed, the high capital

c osts ofte n lead to ra te increases to utilitycustomers, at least until the plant has beenpartia lly am ortized . This c an c ause c onsid-erable difficulty with both the customers andthe public utility commission (PUC), If rateincreases are delayed to ease the shock, netpayba c k to the utility is po stp one d further.

Most of the money to pay for a plant has tobe raised from the financ ial market, whe renuclear reactors increasingly are viewed asrisky investments. The hug e dem and s forc ap ital to p ay c onstruction c osts (and the

high interest costs on this capital) make un-precedented financial demands on util it iesat a time when capital is costly.

There are ma ny op po rtunities for op po nentsof a p lant to vo ice the ir c onc erns. Som epIants have been the focus of suits over spe-cific environmental or safety issues. in thelicensing process, critics may raise a widevariety of issues to which the utility has to

be prep ared to respo nd . These responses ca llfor a signific ant leg al and tec hnic al effort a swell as long delays, regardless of the ultimate

disposition of the issue.

Plant operation may not meet expectations,Some reactors have suffered chronic reliabili-ty problems, operating less than 50 percentof the time. Others have had to replac e ma -

jor components, such as steam generators,



Ch. 1—Introduction: The Seven-Sided Coin q5

at a cost of tens of millions of dollars becauseof unexpectedly rapid deterioration. Whilethere is no specific reason to think a newplant would not operate its full life expect-

ancy without major repairs, no reactor is yetold eno ugh to ha ve d em onstrate d it. Therealso is the possibility of long-term shutdowns

bec ause o f ac c idents suc h a s Three MileIsland. Furthermore, a nuclear utility is

vuInerable to shutdowns and major modifi-cations not only from accidents at its ownfac ility, but also from a c c ide nts at any o therreactor.

qPublic support for nuclear power has beenslipping, largely due to concerns about safetyand costs. Public concerns can manifestthe mselves in po litica l opposition . Seve ral

states have held referenda banning nuclearpower or restricting future construction.None has passed that would mandate shut-

ting down operating reactors, but some havec om e c lose. Furthermore, Sta te a nd localgovernments have considerable control overthe plant through rate regulation, permitting,transportation of waste, and approval ofemergency plans. If the public does not wantthe plant, all these levers are likely to be usedagainst it.

Given all these uncertainties and risks, fewutilities wo uld no w c onsider nuc lear rea c tors tobe a rea sona ble c hoice. Mo reove r, the p ressuresarising from virtually continuous interactions with

contractors, NRC, the PUCs, financial institutions,and perhaps lawsuits by opponents, make nucle-ar po wer far more b urde nsome to a utility than

any o ther choice . The future o f nuclear po werwould appear to be bleak.

Yet there is more to nuclear power than the

well-publicized problems affecting some reactors.In fact, many have been constructed expeditious-ly, and are operating with acceptable reliabil ity.

Some have enjoyed spectacular success. For in-stance, the McGuire unit 2 of Duke Power inNorth Carolina was completed in 1982 at a costof $900/kW, less tha n a third of t he c ost o f theShoreham p lant in New York. The Vermont Yan-kee plant operated in 1982 at 93 percent avail-ability, one of the best records in the world forany kind of g ene ra ting p lant. Ca lvert Cliffs sup -

p lies elec tric ity to Ba ltimo re Ga s & Elec tric c us-to me rs a t 1.7¢/ kWh. Finally, sa fety a na lyses a reimproving steadily, and none has indicated thatnuc lear plants pose a leve l of risk to the pub licas high as that accepted readily from other tech-

nolog ies. These w ell-ma na ge d plant s ha ve o pe r-ated safely whiIe providing substantial econom-ic benefits for their customers.

Suc h exam ples, howe ver, a re insufficien t toc ounterba lanc e the p rob lems others have e n-countered. Nuclear power has become entangledin a complex web of such conflicting interests andem otions that m atte rs are at an imp asse. The util-ity viewp oint d isc ussed above show s tha t the reis l it t le advantage and a great many disadvantagesto the selection of a nuclear plant when new ca-pac ity is nee ded . Therefore, there will be fe w–

if any—more orders for reactors in this centurywithout significant changes i n the way the indus-try and the Government handle nuclear power.

THE IMPASSE

Consider now the perspective of those Federalenergy policy makers who believe the nuclear op-tion should be maintained in the national interest.It is unlikely tha t the U.S. Government w ill hea v-iIy sub sidize the purchase o f reac to rs by utilities

or that it will build and operate reactors itself.Therefo re, new o rders will be stimulated on ly byalleviating those concerns and problems that nowprec lude suc h orde rs. Any po lic y initiat ive tha tis p rop osed , how eve r, is likely to b e c ontrove r-

sial, because there are at least seven parties withdistinct–and often conflicting–interests:

q utilities,q nuc lear sa fety reg uIa tors,

q critics of nuclear power,q the public,q the nuclear supply industry,q investors and the financ ia l c om munity, and

q State public utility commissions.

25-450 0 - 84 - 2 : QL 3




To illustrate how these interests pull in differentdirections for different reasons, consider just oneissue. Changes in plant licensing and safety reg-ulation often are cited as necessary elements ofany strategy to revitalize the option, but there islittle a greeme nt on e ither the type or extent o freform that should be instituted.

q

q

q

Before utilities will ma ke a c omm itment toinvest several billion dollars in a nuclearplant, they want assurances that extensivemodifications will not be necessary and thatthe regu lations will rema in rela tively sta b le.Utilities contend that such regulatorychanges delay construction and add greatlyto costs without a clear demonstration of asignific an t risk to p ub lic he a lth a nd sa fety.To the utilities, suc h a ssuranc es do not a p-

pe a r to be impo ssib le to grant . They p ointout that NRC has licensed 80 plants and

should know what is necessary to ensureop erating sa fety. Therefore, they w ould sup -port revisions to the regulatory process thatwould make it more predictable and stable,

However, there is another side to this coin.No plant design has been analyzed exhaus-tively for every possible serious accident se-quence, and operating experience is still toolimited for all the potential problems to havebeen ident ified . Acc idents a t Three MileIsland and at the Browns Ferry reactor in-volved sequences of events that were not un-derstood clearly enough until they occurred.

if they had been, both could have been pre-ven ted ea sily. As the NRC and the industryrecognize different accident sequences,backfits are needed to prevent future occur-

rences. Proposals to reduce NRC’s ability toimpose changes in accordance with its engi-

neering judgment will be seen by safety reg-uIators as hampering their mission of ensur-ing safety.

But there is a third side to this coin. Not onlydo the industry and NRC see reg ulatory re-form ve ry d ifferently, but critics of nuclear

po wer find muc h to fa ult with both the util-ities and the NRC. In particular, they feel thatthe NRC d oe s not eve n en force its p resentrules fully when such enforcement would betoo costly to the industry. Furthermore, they

q

q

q

q

believe that the technology has so many

uncertainties that much greater margins ofsa fety are wa rrante d. Thus, nuc lear c riticsstrenuously oppose any changes in the NRCreg ulations tha t m ight limit the ir ac c ess tothe reg ulatory p roc ess or c onstra in the im-plementation of potential improvements in

reactor safety.The public is yet a fourth side. Public opin-ion polls show a long-term trend against nu-c lea r powe r. The p ublic de ma nds that nu-clear reactors pose no significant risks, isfrustrated by the confusing controversy sur-rounding them, and is growing increasinglyskeptica l abo ut any b enefits from nuclea rpow er. These c ond itions do not give rise t oa clear mandate for regulatory reform inorder to facilitate more reactor orders. Sucha mandate will depend largely on improvedpublic confidence in the management ability

of utilities and their contractors, in the safe-ty of the technology, in the effectiveness ofthe regulatory proc ess, and on a pe rc ep tiontha t nuc lear ene rgy o ffers rea l bene fits.The nuclear supply industry’s interests are

not synonym ous with the utilities’ and thusrepresent a fifth side of the c oin. The util-it ies need to mee t de mand with whateverop tion ap pe ars lea st expe nsive. If that o p -tion is not nuc lear p ow er, som ething else w illsuffice . The sup p ly ind ustry, however, hasa large vested interest in promoting nuclearreactors, and the careers of thousands of

industry employees may hinge on policychanges to revitalize the nuclear option, in-cluding regulatory reform.Investors may be ambivalent about l icens-ing reform. Lengthy and uncertain licensingmakes nuclear power a riskier investmentduring c onstruction, but a ny ac c ide nt dur-ing operation can have the same, if not great-er, effect. Insofar as more stringent licens-ing makes accidents less likely, it reduces thefinancial risk. However, investors probablywill be more concerned with the near-termrisks involved in g etting a plant online andwould be more supportive of streamlinedlicensing if it reduced those risks.As representatives of consumers’ economicinterest, public utility commissions’ share



Ch. 1—introduction: The Seven-Sided Coin q 7

the investors’ am bivalence , but they m ightgive more weight to operating safety because

an accident that shuts down a reactor for aprolonged period usually will mean the sub-stitution of more expensive sources of elec-tricity.

Thus, there a re at lea st seven d ifferent pa rtiesin each policy debate on nuclear power: sevensides to the coin of each issue. No doubt otherscould be added, but those described above rep-resent the major positions. Each party is a col-

lection of somewhat differing interests, and eachwill look for different things in any policy initia-tive. Given such a multiplicity of interests, it isnot surprising that the present impasse hasdeveloped.

Figure 1 illustra tes the se c onc ep ts. Utilities a reat the center because they make the ultimate de-cision about whether to order a nuclear plant or

som ething e lse. The o the r pa rties ha ve c onsid-erable, sometimes decisive, influence over

whether a nuclear plant will be built, how muchit will cost, and how well it will work. Each of

these parties has its own agenda of conditions thatmust be met before it would support a decisionby a utility to orde r a reac tor. These c ond itions

a re listed with ea c h p arty. Those c ond itions thatare c omm on to a ll are listed at the b ottom of the

figure. For instance, nuclear power must be verysa fe, with a very low risk of c ore m eltdo wns ormajor releases of radioactivity. Disputes over thispo int relate to the de gree o f safety required , thead equa cy o f the method ology in determining

safety, the assumptions of the analyses, and theactual degree of compliance with regulations. inany case, however, existing reactors must bedemonstrably safe, and future reactors probablywill be held to even higher standards.

A c losely rela ted issue is reliab ility. A smo othly

operating reactor is more productive for its own-

ers, and it also is likely to b e sa fer tha n one tha tfrequently suffers mishaps, even if those mishaps

Figure A.— The Seven Sides to the Nuclear Debate

Public Utility Commissions Nuclear critics

Stable construction costs Confidence in the technologyMinimal operating risks Confidence in regulators and utilitiesAdequate financing Economic advantagePublic support Liabilities of other fuels proved

Utilitiesinvestors

.Adequate return on Investment

Nuclear industry

Healthy utility Adequate financing Stable IicensingStable construction costs Minimal opposition National policyMinimal operating risks predictable construction costs Public acceptanceMinimal political risk Public and political acceptance Favorable risk/reward

. Predictable regulationq

Nuclear RegulatoryCommision / /

Confidence in technologyConfidence in utilitiesPublic support I

Public

Confidence in safetyconfidence in regulators and utilitiesLees controversyEconomic advantageNational policy -

Noncontroversial, necessary conditions

No major accidents

Reactors prove reasonably reliableAdditional generating units neededCost advantage for nuclear powerConvincing waste disposal program

SOURCE Off Ice of Technology Assessment



8 qNuclear Power in an Age of Uncertainty

ha ve no imm ed ia te sa fety c onseq uenc es. Thus,it also will be c onsiderab ly m ore rea ssuring to thepubl ic.

Other common criteria are that there must bea clear need for new generating capacity and a

significant cost advantage for nuclear power. Inaddition, a credible waste disposal program is aprerequisite for any more orders.

Other conditions are especially important tosome groups but less important to others. Someof these conditions already are met to somede gree. The a rrow s in figure A draw n to the c on-ditions under utilities indicate the major areas thatare related to the other parties.

THE PURPOSE

This rep ort responds to req uests from theHouse Com mittee o n Sc ienc e and Tec hnologyand the Senate Com mittee on Energy and NaturalResource s asking OTA to “ assess how nuc lea r

technology could evolve if the option is to bema de more attrac tive to a ll the p arties of c on-

c ern” and to identify possible tec hnic al and in-stitutional approaches for the Congress “thatcould contribute to the maintenance of this im-portant industry.” The rep ort de sc ribes the ma -

jor impediments to nuclear power relative toother types of generating capacity, identifies op-t ions that might be considered to remove those

imp ed iments in light of the problems and c on-flicts discussed above, and explores the conse-quences of not maintaining the nuclear option.

Changes could be made in the technology andin the institutions that manage it. If a reactor were

to be developed that physically could not suffera major accident or pose health and safety risksfor the public, it might allay some of the concernsof the regulators, the interveners, and the public.Suc h a rea c tor might not require the e ver morestringent standards of quality required for currentlight water reactors (LWRs), thus reducing the

economic risks. Improvements also could be con-sidered in management of the construction, op-eration, and reg ulation of reac tors. If a ll reac torswere to match the experiences of the best man-

Ma ny of the c ond itions in figure 1 are neces-sary before enough of the participants in thedeb a te w ill be sa tisfied tha t nuc lear po wer is aviable energy source for the future. It is muchmore diffic ult to know ho w ma ny must be met

to be sufficient. All the groups discussed abovehave c onside rab le influenc e o ver the future o f

nuclear power. Efforts to revive the option—whether initiated legislatively, administratively,or by industry—are unlikely to be successful ifsom e of the interests find them unac c ep tab le. Thetask of breaking the impasse therefore is formid-able.

OF THIS STUDY

aged plants, there would be much less concernover the future prospects for the nuclear option.

It is the intent of this study to explore these

possibilities in the light of the different interestsand d ifferent c onc erns d isc ussed ab ove . Therep ort d eta ils the va rious d iffic ulties fac ing thefuture o f nuclea r powe r and the m ea sures thatwould be useful and practical in overcomingthese difficulties if the Nation wishes nuclearpower to once again be a well accepted, viableene rgy op tion. The tec hnolog ic a l op tions a re re-stric ted to c onve rter reac tors simila r to tho se n ow

av ailable o n the internationa l market. These a rethe reactors that could be deployed in the United

Stat es by the end of the c entury. Bree de r rea c -tors are not included because their developmentprogram will not make them commercially avail-ab le until som etime in the next c entury. Theother elements of the fuel cycle—uranium re-sources and enrichment, reprocessing and wastedisposal–are not included either. Waste hasbe en c onside red in g rea t d eta il in a rec ent O TArep ort. The o ther elements nee d no t po se c on-

straints to reactor orders, which is the key issueaddressed in this report.

This assessme nt w as c a rried out w ith the a s-

sistance of a large number of experts from all sidesof the nuc lear de ba te—utilities, nuc lea r critics,



Ch. 1—Introduction: The Seven-Sided Coin q 9

reactor vendors, consumer groups, NRC, aca-

demics, State PUCs, nuclear insurers, executive

branch agencies, the financial community, archi-

tect-engineering (AE) firms, and interested mem-

bers of the pub lic , As in all OTA stud ies, an ad -

visory panel representing most of these interestsmet periodicalIy during the course of the assess-

ment to review and c rit ique interim p rod ucts andthis report. Contractors supplied analyses and

background papers in support of the assessment(these are compiled in vol. II). In a dd ition, O TAheld three workshops to review and expand onthe contractors’ reports and to ensure that all thereleva nt interests on e ac h issue w ould b e c on-

side red . The first wo rkshop e xam ined t he e ne rgyand economic context for nuclear power, includ-ing projections of electricity demand, capital costsfor pow erplant c onstruction, and the financ inga nd rate regulation of e lec tricity gene ration. Thesecond workshop focused on the technological,

managerial, and regulatory context for nuclearpower, identifying the problems with currentLWRs and the licensing process for them, and as-sessing alternative reactor technologies and pro-posals for licensing revision. The third wo rkshop

examined institutional changes, public accept-ance, and policy options for revitalizing nuclearpower. Based on these and other discussions, theOTA staff d eve lop ed a set of po licy op tions. Ad-visory panel members, contractors, and workshopparticipa nts a re listed a t the front of the rep ort.

The nuc lea r de ba te long ha s be en c harac ter-

ized by inflexible, polarized positions. We seesome evidence that this polarization is softeningFor the mo st p art, the O TA w orkshop p artic ipa ntsand a dvisory pa nel mem bers show ed a w illing -

ness to compromise, including admissions by in-dustry representatives that many mistakes hadbeen made, and by nuclear crit ics that nuclearpo wer co uld be a viab le source o f electric ity ifmanaged properly.

Volume I of this report is orga nized as follows:

• Cha pte r 2 presents a summa ry of the rep ort.q Chapter 3 sets the context for decisions on

the future role of nuclear power–factors af-fecting electricity demand, f inancial consid-erations including rate regulation and the

q

q

q

q

q

q

costs of nuclear plants, and other elementsin utility planning.

Chapter 4 considers the technological alter-natives to today’s light water reactor: im-proved LWRs; the high-temperature gas re-actor as evolving from the demonstrationplant at Fort St. Vrain, Colo.; the heavy water

rea c tor as de veloped in Ca nad a; the PIUSconcept—an LWR redesigned to make catas-trophic accidents essentially impossible; theeffects of standardization and sealing downreac tor size.

Chapter 5 examines the human element inbuilding and operating reactors and ways toimprove the q ua lity of the se e fforts; it a na -lyzes the wide range of experiences in con-struction c osts and sc hed ules and in rea c -tor operation, new measures that may im-p rove qua lity co ntrol (e.g., the Institute forNuclear Power Operations), and further

steps that could be implemented if existingefforts prove inadequate.

Chapter 6 de sc ribes the p resent reg ulato ryprocess and the various concerns with it, andevaluates the major proposals for revision.

Chapter 7 reviews the long term viab ility ofthe nuclear industry if no new orders areforthc oming to see if the op tion wo uld beforeclosed without stimulation, and how theop eration o f existing rea c tors would be af-fected; and examines the management ofnuclear power in other countries to see what

lessons can be learned from alternativeapproaches.

Chapter 8 foc uses on trends and influentia lfactors in public acceptance as one of thekey elements in a revival of nuclear power,and evaluates measures designed to improvepublic acceptance.

Chapter 9 analyzes a series of policy optionsthat Congress might consider. Depending onone’s views of the desirability of and necessi-ty for nuc lea r powe r, a p olic y ma ker mightsee little nee d to d o a nything, wa nt to im-p rove the o pe ra tion of existing rea c tors, or

make the option more attractive so that it canplay an expanded role in the Nation’s energyfuture. The o pt ions a re a na lyzed for effec -



10 • Nuclear Power in an Age of Uncertainty

t iveness and for ac c ep tab ility by the va rious in support of the assessment, will be available

parties to the debate. Packages of options are through the Nationa l Tec hnica l Informa tion Serv-considered to see if compromises might be ice, 5285 Port Royal Rd., Springfield, Va. 22161.

possible.

Volume II of the report, which includes con-tractor reports and background papers prepared



Chapter 2

Summary

Photo credit: Department of Energy

Oconee nuclear units owned by Duke Power Co.



Contents

Page

The Unc erta in Financ ial a nd Ec onom ic Future . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13

Nuc lea r Rea c tor Tec hno log y.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15

Management of Nuclear Powerplants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18Reg ula to ry Co nside rat ions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21

Survival of the Nuclear industry in the United States and Abroad . . . . . . . . . . . . . . . . . . 22

Public Attitudes on Nuclear Power . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23

Polic y Options . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25

Tables

Tab/e No. Page

I. Additional Capacity Required by 2000. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132. Comparison of Construction and Reliability Records for Selected

U.S. Light Water Rea c tors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19

Figures

Figure No. Page

l. Concep tua l Design of a Modu lar, Peb ble -Bed HTGR . . . . . . . . . . . . . . . . . . . . . . . . . . . 172. Trend s in Pub lic Op inion on Nuc lea r Power . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24



Chapter 2

Summary

THE UNCERTAIN FINANCIAL AND ECONOMIC FUTURE

Future orders for nuclear plants depend inpart on electricity demand and on the financial

comparisons that utilities will make with alter-natives to nuclear power. Utilities ordered farmore generating capacity in the early 1970’s thanthey turned out to need, and have canceled manyof their planned plants. Nuclear plants haveborne the brunt of the slowdown in construction.

There has been a pronounced decline in thegrowth rate of electricity demand. D e m a n dgrowth has averaged about 2.5 percent annual-ly sinc e 1973, c omp ared to a bo ut 7.0 perce ntfrom 1960 to 1972. Utility executives contem-plating the construction of long Ieadtime coal

or nuclear powerplants must contend with con-siderable uncertainty about the probable futuregrowth rates in electricity demand. With certainassumptions about the future, it is reasonable toexpect fairly slow growth rates of 1 to 2 percentpe r yea r. Very few large new po we rplants wo uldbe required to meet this demand. With otherplausible assumptions, electricity load growthc ould resume at rate s of 3 to 4 p ercent p er year,which would require the construction of severalhundred gigawatts* of new powerplants by theyear 2000. The a c tual need for new p ow erplantswill depend on the g rowth rate of the ec onomy,

the rate of increase in the e ffic iency o f use o f elec -tricity, price increases for electricity vis a vis otherenergy sources, new uses for electricity, and therate of retirement of existing plants. None of thesevariab les c an be predicte d with ce rtainty. The e f-fects of the electric growth rate and the replace-ment rate on the capacity that would have to be

ordered in time to be completed by 2000 areshown in table 1.

in a dd ition to the slowd own in elec tric loa d

growth, powerplants have also been canceledand de ferred due to d eterioration in the finan-

c ia l co nd ition o f utilities. Although the industry’s

*One gigawatt equals 1000 MW (1 ,000,000 kW) or slightly lessthan the typical large nuclear powerplant of 1100 to 1300 MW.

Table 1 .–Additional Capacity Required by 2000(gigawatts)

Levels of replacement Electricity demand growthof existing plants 1.5%/yr 2.5%/yr 3.5%/yr

Low: 50 GW to replace all plantsover 50 years old . . . . . . . . . 9 144 303

Moderate: 125 GW to replace allplants over 40 years old, plus20 GW of oil and gascapacity . . . . . . . . . . . . . . . . 84 219 379

High: 200 GW to replace allplants over 40 years plus twothirds of the oil and gascapacity . . . . . . . . . . . . . . . . 159 294 454

NOTES: 1. Planned generating capacity for 1991 is 740 GW, 158 GW more than1982 generating sources of 582 GW. Starting point for demand calcula-

tions Is 1982 summer peak demand of 428 GW.2. The calculations assume a 20-percent reserve margin, excess of

planned generating resources over peak demand.

SOURCE Office of Technology Assessment

financial picture is improving as external financ-ing needs decline and allowed rates of return in-crease, current rate structures still may not pro-vide adequate returns for new investment in largenuc lea r projec ts. Without changes in rate regu-lation, utilities may not be able to attract capitalwhen they need it for construction, because in-vestment advisers associate construction witha deterioration in financial health.

The primary targets for rate reform includethe current lag between allowed and earnedrates of return, the “rate shock” which resultsin the first few years after a large, capital-intensive plant is added to the rate base, andthe absence of explicit incentives to reduce fuelcosts. Options for resolving these problemsassume tha t the inve stors a nd Sta te p ub lic utilitycommissions will take a long-term perspectiveand will ma intain a p artic ular method of d eter-mining revenue requirements for several dec-ades. Yet when commissioners may only remainin office for a few years, or when State legislaturesadopt a short-term perspective, methods that takea long-term view of rate regulation are difficult

to achieve.

13




Although ratemaking changes to increase theattractiveness of capital investment would

eliminate some disincentives, utilities and theirinvestors and ratepayers would still face sub-

stantial financial risks from nuclear power.These risks includ e the unp redic ta b ility of the

capital costs of a nuclear plant at the beginningof c onstruction, the d iffic ulty of predicting c on-

struction Ieadtimes, the very high costs of cleanup

and replacement power in the event of a majoraccident, and the possibility of future regulatorychanges.

Nuclear plant average construction costs more

than doubled in constant dollars during the

1970’s and are expec ted to increase b y another80 percent for plants now under construction.

Som e o f this increase ha s c om e from ne w reg u-latory requirements which are applied to allplants, whether operating or under construction.

Some utilities, howe ver, have ad ap ted be tter tothese new regulatory conditions, as shown by the

increasing variability in capital costs. Of the group

of plants now under construction, the most ex-pe nsive is expec ted to co st mo re tha n four times

the least expensive. The variation in c ost ha s be endue in part to regional differences in the cost oflab or and ma terials and the we ather, but more

to differences in the experience and ability ofutility and construction managers. Only the best

Photo credit: United Engineers & Constructors

Managing a multibillion dollar nuclear construction project is difficult, complex, and subject to uncertainty.The most expensive nuclear plant under construction is estimated to cost about four times the least expensive

(per unit of generating capacity)



Ch. 2—Summary q 15

managed construction projects are now compet-itive with new coal plants.

Average nuclear plant construction Ieadtimesdoubled over the decade (from about 60 to about120 months) and are now about 40 percent long-er than coal plant Ieadtimes. Very long Ieadtimes

increase interest costs and the difficulty of match-ing ca pa city with demand . Average plant con-struction costs and leadtimes could be reducedin the future in several ways: 1) Plants could bebuilt only by experienc ed and c om pe tent utilit iesand contractors, who would work under con-tracts with incentives to control costs and use in-nova tive co nstruction tec hniques. 2) Standa rdiza-tion of design and licensing could bring the low-est U.S. co nstruc tion c osts dow n by anothe r 20to 25 percent. 3) Further reductions in plant car-rying costs could come about if Ieadtimes werec ut b y 25 to 30 pe rc ent a nd interest ra tes we re

reduced. It should also be recognized, however,that there are circumstances under which costsmight increase. In particular, further serious ac-cidents or resolution of important safety issuesc ould lea d to a new round o f co stly cha nge s.

Utility executives are also aware that singleevents couId occur causing the loss of the entire

value of a nuclea r plant. The a c c ide nt a t ThreeMile island will have cost the owner $1 billionin cleanup costs alone, plus the cost of replace-ment power, the carrying costs and amortizationof the original capital used to build the plant, andthe cost of restarting the plant (if possible). Only$300 million of the cleanup cost was covered by

prop erty insuranc e. Nuclea r plants c an also b eclosed by referenda such as the narrowly de-feated vote in 1982 that would have closed MaineYankee.

Utility exec utives have o ther op tions to m ee tfuture load growth than c onstruc ting new ge n-erating plants including: converting oil or gasplants to coal, building transmission lines to fa-cilitate purchase of bulk power, developing smallhydro, wind or cogeneration sources, or loadmanagement and energy conservation programs.Som e o f these a lterna tives ma y prove m ore a t-

tractive to utilities than nuclear plants given theuncertain demand and financial situation. Evenif rate regulatory policies across the country wereto shift to favoring longer Ieadtime capital-inten-sive technologies, smaller coal-fired powerplantswould b e p referred be c ause they ha ve shorterIea d times, low er fina nc ia l risk, and grea ter p ub licacceptance than current nuclear designs.

NUCLEAR REACTOR TECHNOLOGY

Virtually all nuclear powerplants in this coun-try and most in other countries are light waterreac tors (LWRs). This c onc ep t w as de velop ed forthe nuc lea r-powe red subm arine p rog ram , andwa s ad ap ted to e lectric utility nee ds. Since the n,many questions have been raised over the safe-ty and reliability of LWRs in utility service, costshave risen dramatically and regulatory require-me nts ha ve p rolifera ted . There is no spec ific in-dication that LWRs cannot operate safely for theirexpec ted lifetimes, but it a pp ea rs that c urrentLWR designs are unlikely to be viable choices inthe future unless concerns over costs, regulatory

uncertainties and safety can be alleviated. EitherLWR designs will have to be upgraded, or alter-native reactor concepts will have to be consid-ered.

There is no sta nd ardized LWR design in theUnited Sta tes. This is due to two m a jor fac to rs.First, the different combinations of vendors, archi-tect-engineers (AEs), constructors, and utilitiesproduced custom-built plants for each site. Inadd ition, the rea c tor designs them selves have

changed greatly since introduction of LWR tech-nology. The pa c e of de velopme nt from p roto -type to nearly 100 commercial reactors was veryrapid. Large, new reactors were designed andconstruction started prior to significant operatingexpe rience of their prede c essors. As ha rdwa reproblems developed or new safety issues sur-

faced, changes had to b e mad e to existing reactors, rather than integrating them into new de-signs, As regulatory agencies improved theirunderstanding of nuclear power safety, criteria



16 . Nuclear Power in an Age of uncertainty

changed, and many features had to be mandatedas retrofits. Thus the light w a ter rea c tors und erconstruction and in operation today do not rep-resent an optimized LWR design.

Utilities’ experience with the LWR range fromexcellent to poor. Some rea c tors have op eratedat up to a n 80-perce nt ca pa city fac tor for years

with no significant problems, while others havebeen plagued with continual hardware problemsthat lead to low-capacity utilization. While thesafety record to date is very good, the accidentat Three M ile Island (TMI) and other potent iallysevere incidents raise concerns about the abilityof all utilities to maintain that record.

Many of the concerns over safety and reliabilityhave been fueled by the seemingly constantstream of hardware problems and backfits asso-c iated with LWRs. Many of tho se in the nuc learindustry fee l that suc h p rob lems reflec t no rma l

progress along the learning c urve o f a very com -plex technology, and they assert that the reac-tors are nea ring a platea u on tha t c urve. Nuclea rcritics observe that there are still many unresolved

safety issues associated with LWRs, and the tech-nology must continue to change until these areaddressed adequately.

The design and operation of LWRs has un-questionably improved over time. The tra iningof operators has been upgraded, human factorsconsiderations have been incorporated into con-trol room design, information on operating ex-perience is shared, and numerous retrofits have

been made to existing reactors.

Whether these steps have made LWRs safeenough cannot be demonstrated unambiguous-ly, however. There is no c onsensus on ho w todetermine the present level of safety, nor on themagnitude of risk represented by particular prob-lems or the cost-benefit criteria for assessing possi-

ble solutions. In some cases, retrofits in one areacan possibly reduce safety in other areas, eitherbecause of unanticipated system interactions, orsimply because the additional hardware makesit difficult to get into part of the plant for

maintenance or repairs. Even if all the parties tothe debate could agree that the risks are accept-

ably small, the public still might not perceivenuclear power as safe.

It is clear that, before they order new nuclearplants, utilities will want assurances that theplants will operate reliably and will not requireexpensive retrofits or repairs due to unantici-pated design problems or new Nuclear Regula-tory Commission (NRC) regulations which maybe needed to solve such problems, and that theywill not run an unacceptable risk of a TMI-typeaccident that could bankrupt them.

Many of the nuclear industry’s concernsabout the current generation of LWRs are be-ing addressed in designs for advanced reactors.An advanced pressurized water reactor is beingdesigned to be safer and easier to operate thanthe p resent ge neration, and to ha ve improvedfuel burnup and higher availability (90 percentis the goal) through resolution of some of themore crit ical hardware problems. An advancedboiling water reactor is being designed to oper-ate at a relatively high capacity factor and to in-corporate advanced safety features that willreduce the risk of core-melting even if the primary

c oo ling system fa ils.

If the utilities and the public cannot be con-vinced that new LWRs would be acceptably safeand reliable, however, renewed interest may de-velop in using alternative reactor technologies.Among the more promising near-term possibili-ties are high temperature gas-cooled reactors,LWRs with inherently safe features, and the heavywater reactor.

The high temperature gas-cooled reactor(HTGR) has attracted considerable interestbecause of its high thermal efficiency (nearly 40percent—compared to 33 percent for an LWR)and its inherent safety features. The c ore o f theHTGR is slow to hea t up even if c oo lant flow isinterrupted; this reduces the urgency of the ac-tions that m ust be taken to respo nd to an a c c i-de nt. In a dd ition, the entire c ore and the p rima rycooling loops are enclosed by a vessel whichwouId prevent the release of radioactive materi-als even after an accident. Lessons learned on theon ly U.S. op era ting HTGR are being app lied to

the design of a 900-megawatt (MW) prototype.Sma ll, modular versions (fig. 1) also ha ve bee nproposed that might have very attractive safetyc ha rac teristics and be e spe c ially suitab le as a



Ch. 2—Summary q 1 7

Photo credit: Gas-cooled Reactor Associates (GCRA)

The fuel in a high temperature gas-cooled reactoris inserted into graphite blocks like the one shownabove. The fuel form is one of the key features ofthe HTGR, since graphite can absorb a great deal

of heat before melting

source of p roc ess hea t. While HTGRs appea r to

be potentially safer than LWRs, there are stillma ny q uestions c onc erning HTGR reliab ility andeconomics. Continued research and develop-ment (R&D) of the HTGR is necessary if thesequestions are to be resolved.

The heavy water reactor (HWR) has attractivesafety and reliability features, but there areseveral roadblocks to its adoption in this coun-try. The HWR ha s pe rformed we ll in Ca na d a, b utthe process of adapting it to the American envi-ronment might introduce modifications whichwould lower its performance. In addition, muchof its go od p erforma nc e m a y b e the result of skill-ful management and not a consequence of thereactor design. Without significant evidence thatthe reac tor is inherently sup erior to othe r op tions,the HWR is no t a strong c and ida te fo r the U.S.

Figure 1 .—Conceptual Design of a Modular,Pebble-Bed HTGR

Top-mounted helium circulator

Steam Feedwateroutlet inlet

Reflectorcontrol rod

200-MWtpebble-bed rreactor ,core

I

Pebbletransportsystem

Steam

/ “ ’ g e n e r a t o r

fuel \ /

Spent fuelSOURCE: Office of Technology Assessment.

market, unless the Canadian technology can beea sily a dap ted , or the U.S. experienc e w ithHWRs in the we ap ons p rog ram c an b e ut ilized .

The process inherent ultimately safe (PIUS) re-actor, a new LWR concept being developed inSweden, is designed with safety as the primaryobjective. protective against large releases of

radioactivity would be provided by passive meansthat are independent of operator intervention and




mechanical or electrical components. Becausethe PIUS is designe d so tha t a me ltdo wn is virtu-ally impossible, it might be the reactor most suitedto restoring public confidence in nuclear power.The PIUS reactor is still in the initial design phase,however, and has not yet been tested, althoughcomputer simulations have been init iated to ad-

dress questions about operational stability. Exten-sive R&D is nee ded to na rrow the unc ertaintiesab out c ost, ope ration, and ma intena nce . ThisR&D a nd eve ntua l deploym ent of t he PIUS,would be expedited by its similarity, in somerespects, to conventional LWRs.

Features that might be applied to any reactortechnology include smaller sizes and stand-ardized designs. Smaller nuclear plants wouldprovide greater flexibility in utility planning—especially in times of uncertain demand growth —and less extreme economic consequencesfrom an unscheduled outage. The shorter c on-

struction periods and lower interest costs duringconstruction would reduce the utilities’ financialexposure. The a b ility to build m ore o f the sub -systems in the factory rather than onsite mightreduce some construction costs, offsetting the losteconomies of scale. Moreover, smaller reactorsmight be easier to understand, more manageable

to construct, and safer to operate. Federal R&D

would probably be required to achieve designsthat exploit the favorable characteristics of smallreactors.

The p ote ntial bene fit of a stan da rd ized d esignappears to be especially high in view of the prob-lems of today’s nuclear industry. Many of theproblems with construction and operation stem

from mismanagement and inexperience, and astandardized plant would help all utilities learnfrom those who have been successful. Franceand Cana da seem to have done wel l with build-ing many plants of one basic design. Still, the im-plementation of standardized plants in the UnitedStates faces many obstacles. Reactor system de-signs differ from vendor to vendor and grow fur-ther apart when coupled with the different bal-ance of plant designs supplied by the numerousAEs. They are a dd itiona lly mod ified by the re-quirements of NRC, the utilities, and the specificsites. Despite these obstacles, the industry may

be motivated to converge on one or two stand-ardized designs if that path seemed to offerstreamlined licensing, stabilized regulation,faster construction, and better management.The he lp of the Fed eral Government ma y b e re-quired to develop and approve of a common de-sign, e spec ially if it is signific antly d ifferent fromthe LWR.

MANAGEMENT OF NUCLEAR POWERPLANTS

The management of commercial nuclear pow-erplants has proven to be a more difficult taskthan originally anticipated by the early pro-ponents of nuclear technology. While the overallsa fety rec ord of U.S. plants is very good , therehas be en g rea t va riab ility in c onstruc tion time sand capacity factors (see table 2). Some utilitieshave demonstrated that nuclear power can bewell managed, but many util it ies have encoun-tered d iffic ulties, Som e of these p rob lems havebe en serious enoug h to have safety a nd finan-cial implications. Since the entire industry isoften judged by the worst cases, it is importantthat all nuclear utilities be able to demonstratethe capability to manage their powerplants safe-ly and reliably.

There a re ma ny spe c ia l prob lem s assoc ia tedwith managing a nuclear powerplant. Nuclear reac-tors are typica lly half ag ain as large and c on-siderably more complex than coal plants. The job

of building and operating a nuclear powerplanthas been further complicated by the rapid paceof development. As new lessons were learnedfrom the ma turing tec hnology, they had to b e in-corporated as retrofits rather than integrated intothe origina l de sign. The regulato ry struc ture w asevolving along with the plants, and the additionalengineering associated with changing NRC reg-

ulations and with retrofits strained the alreadysc a rce resources of ma ny ut ilities. Som e utilitieshave also had diff iculty coordinating the variousparticipants in a construction project.




Table 2.—Comparison of Construction and Reliability Records for Selected U.S. Light Water Reactors

Construction a “ --- Reliabilityb

— .Lifetime

Date of commercial Years to Date of commercial capacityoperation construct operation factor (0 /0)

Best:St. Lucie 2 . . . . . . . . . . . . 1983Hatch 2 . . . . . . . . . . . . . . . 1979

Arkansas Nuclear One 2 1980Perry 1 . . . . . . . . . . . . . . . 1985Palo Verde 1 . . . . . . . . . . 1984Byron 1 . . . . . . . . . . . . . . . 1984Callaway 1 . . . . . . . . . . . . 1984

Worst:Watts Bar 1 . . . . . . . . . . . 1984Sequoyah 2 . . . . . . . . . . . 1982Midland 1 . . . . . . . . . . . . . 1985Zimmer 1 . . . . . . . . . . . . . 1985Salem 2. . . . . . . . . . . . . . . 1981Diablo Canyon 2 . . . . . . . 1984Diablo Canyon 1 . . . . . . . 1984

Best:6 Point Beach 2 . . . . . . . . . 1972 797 Connecticut Yankee . . . . 1968 76

7 Kewaunee . . . . . . . . . . . . 1974 768 Prairie Island 2 . . . . . . . . 1974 768 Calvert Cliffs 2 . . . . . . . . 1977 758 St. Lucie 1 . . . . . . . . . . . . 1976 748

Worst:12 Brunswick 1 . . . . . . . . . . . 1977 4812 Indian Point 3 . . . . . . . . . 1976 4613 Salem 1 . . . . . . . . . . . . . . . 1976 4613 Brunswick 2 . . . . . . . . . . . 1975 4113 Davis Besse 1 . . . . . . . . . 1977 4014 Palisades . . . . . . . . . . . . . 1971 3916 Beaver Valley 1 . . . . . . . . 1977 34

alnclude~ only ~lant~ licensed t. operate after the accident at Three Mile Island in March of 1979Dlnclude~ only ~lant~ greater than IOIJ MWe in Opf3C3tiOfI for k3n9er than 3 years.

SOURCE Nuc/ear News, February 1983 and U S Nuclear Regulatory Commwslon

Both technical and institutional changes are

needed to improve the management of the nu-

clear enterprise. Technical modifications wouldbe useful insofar as they reduce the complexityand sophistication of nuclear plants and theirsensitivity to system interactions and humanerror. More substantial design changes, such asthe PIUS reac tor conc ep t, mig ht be c onside red

as an option since they have the potential to ad-ditionally decrease the sensitivity of nuclear plantsto variations in management ability.

Tec hnolog ic a l c ha nge s, b y themselves, how -ever, cannot eliminate all the difficulties involvedin building and operating nuclear units since theycannot replace commitment to quality and safe-ty. It is important that design changes be sup-plemented with institutional measures to im-prove the management of the nuclear enter-prise. One example is the Institute of Nuclear

Power Operations (INPO), * which is a ttem pt ingto improve the quality of nuclear powerplantoperation, and to enhance communication

among the various segments of the industry.

*The Institute for Nuclear Power Operations is a self-reguIatorynonprofit organization organized by the electric utilities to establish

industrywide standards for the operation of nuclear powerplants,including personnel and training standards, and to ensure that util-ities meet those standards.

The m ost impo rtant improvement required isin the interna l mana ge me nt of nuc lea r utilities.Top utility executives must become aware of the

unique demands of nuclear technology. Theynot only must develop the commitment andskills to meet those demands, but they mustbecome directly involved in their nuclear proj-ects and they must impress on their projectmanagers and contractors a commitment tosafety that goes beyond the need to meet regu-latory requirements. They also need to establish

clear lines of authority and specific respon-sibilities to ensure that their objectives will bemet. INPO could be instrumental in stimulating

an awareness of the unique management needsof nuclear pow er and in providing guida nc e tothe utilities.

It is also important that utilities be evaluatedobjectively to assure that they are performingwell. Both NRC a nd INPO ha ve rec og nized theneed for such evaluations, and currently are en-gaged in assessment activities. INPO attempts toassess the performance of utility management inorder to identify the roo t c auses of the problemsas we ll as the ir conseq uences. The NRC conductsseveral inspection programs with the purpose ofidentifying severe or recurrent deficiencies.

NRC’ s prog ra m is mo re fra gm ente d tha n IN PO’ s,




and the relationships among its various inspec-

tion activities appear to be uncoordinated.

Enforcement activities also can be importantin encouraging better management. Both NRC

and INPO can take actions to encourage utilitiesto make changes or penalize them if their per-formanc e is be low sta nda rd. If mea sures taken

by NRC and INPO prove to be ineffective in pro-moting quality construction and safe operation

of nuclear powerplants, however, more ag-gressive action might be required. A future for

nuclear power could depend on institutionalc hang es that d emo nstrate the a bility of all util-ities with nuclea r pow erplants to ope ra te them

safely and reliably. It is not yet clear whetherthese effo rts will p rove a deq ua te.

Another approach might be for the NRC to re-

quire tha t a utility be c ertified as to its fitness to

build and operate nuclear powerplants. Certifica-

tion c ould force the p oor pe rformers to e ither im-prove their management capabilities, obtain the

expertise from outside, or choose other types ofgenerating capacity.

Many of the c urrent ma nage ment problemscan also be traced to the overlapping and con-

flicting authority of the the utility, the reactor ven-dor, the AE, and the constructor. Centralized

responsibility for overall design and, in some

cases, actual construction could alleviate thisproblem. Increased vendor responsibility might

encourage fixed-price contracts for nuclearplant s, but it co uld d etrac t from utilities’ ability

to ma nage the plant if they are not involved ac -tively in all stages of construction.

A second means of centralizing responsibilityis through nuclear service companies, which

already offer a broad range of regulatory, engi-neering, and other services to utilities. Nuclear

service companies could help strengthen the ca-

pabilities of the weaker utilities by providing allthe services needed to build and/or operate anuclear plant. However, utilities may be reluc-

tant to forego their responsibility for safety andquality while retaining financial liability. Also,without some mechanism that required weaker

utilities to hire service companies, their existence

might ha ve little effec t o n the overall qua lity ofnuclear management.

Privately owned regional or national nuclear

power companies would extend the service com-

pa ny conc ep t into the ac tual ownership o f nu-c lea r powe rplants. Suc h c omp anies c ould b eowned by a consortium of utilities, vendors, AEs,

and / or c onstructo rs and would be c rea ted e x-

pressly to build and operate nuclear powerplants.




REGULATORY CONSIDERATIONS

Nuclea r powe r is one of t he mo st intensivelyregulate d industries in the United Sta tes, and

the sc op e a nd prac tic e o f reg ulation is a volatileissue, Strong —and usua lly c onflic ting—opinions

ab ound a mong the ac tors in the nuclear deb ate

on the ad equa cy a nd efficiency of the c urrent reg-ulatory system.

The utilities and the nuc lear industry ha ve be enoutspoken critics of the current system of nuclearplant regulation, claiming that neither the criterianor the sc hedules for siting , designing, building,and operating nuclear plants are predictableund er the c urrent lic ensing sc hem e. They a rgu ethat public participation has been misused to pro-long lic ensing hea rings unne c essarily. They be lieve

that these factors have been the primary causeof higher costs and longer construction Ieadtimes

and may have been detrimental to safety.

Nuclear crit ics, on the other hand, have beenless critical of Federal regulation of nuclearpowerplants than of the industry that designs,

c onstruc ts, and o pe rates them . They a rgue tha tthe lack of predictability and the increase in lead-t imes were due to the immaturity of the technol-ogy and growing pains due to rapid escalation.They a ttrib ute m an y safe ty c onc erns to utility andconstructor inattention to quality assurance, andinc onsistent interpreta tion and enforcem ent ofregulations within the NRC. While some critics

feel that nuclear plants will never be safe enough,others believe that the current regulatory process

could ensure safety if it were interpreted consist-ently and enforced adequately, but that l imitingthe op portunities for interested me mb ers of thepublic to participate in licensing will detract fromsafety.

As a result of these concerns, a number of mod-ific ations in reac tor reg ulation ha ve b een pro-

posed, either through legislation, rulemaking, orbe tter ma nag eme nt of the regulatory proc ess. Theprimary targets of the various packages are back-fitting, the hearing process, siting, and the licens-ing of designs and plants. The evaluation of pro-posals for regulatory revision must depend firston whether they will ensure adequate protec-

tion of public health and safety and national se-curity, and only secondarily on additional ben-efits, such as reducing the cost of nuclear plants.It is also important to recognize that licensingchanges alone cannot resolve the problems of

the nuclear industry. All parties to nuclearregulation must commit themselves to excel-lence in the management of licensing, construc-tion, and operation, as well as to resolving out-standing safety and reliability issues.

Ma ny nuclear utilities are a da ma nt that theywill not order another reactor until licensing is

more p red ic tab le and c onsistent. These c harac -teristics should a lso b e w elco med by the c ritic ssince they are prerequisites for uniformly highsa fety sta nd a rds. The p rima ry source of c urrentuncertainty is the potential for imposition of back-

fits. Backfits serve an important safety function,since unanticipated safety problems do arise afterconstruction permits are granted. But carefulrevision of the backfit rule could make the proc-ess more rational and ensure that plant safetyis not inadvertently decreased by installation ormaintenance problems or by unexpected inter-actions with other systems. Proposed changesto the backfit rule focus on making the criteriafor orde ring bac kfits mo re explic it, suc h a s theuse of cost-benefit analysis. While a cost-bene-fit approach would “improve” consistency, itshould not be used as the sole criterion since

the available methodologies are inadequate tofully quantify safety improvements. A processto review proposed backfits could also involvea c entralized group either within the NRC o r asan independent panel drawn from utilities, thepublic, and the nuclear industry to ensure thatc riteria and sta nda rds a re c onsistently ap plied.

Legislative amendment of the Atomic EnergyAct is not necessary to reform the backfit regu-lations, since the changes discussed above canbe accomplished through rulemaking. Moreover,

legislative definitions and standards may actual-ly red uce flexibility need ed to a djust to c hang -ing construction and operating experience. Leg-islative action would be more likely, however,to ensure predictability.

25-450 0 - 84 - 3 : QL 3




Another issue in regulatory reform is the useof formal trial-type hearings in reactor licensing.Because adjudicatory hearings can be long andco stly, prop osals have be en ma de to replac ethem with hybrid hearings, which would be morerestricted. A hybrid hearing format might be at-tractive to the owners of nuclear powerplants,but it might also limit the opportunities for public

inquiry and foreclose debate on safety issues. Thehearings could be made more efficient withoutchanging the format if they were managed bet-ter. They could also be improved by makinggreater use of rulemaking to resolve generic is-sues and by eliminating issues not germane tosafety. Only the last of these changes would re-quire legislative amendment of the AtomicEnergy Act.

It has also been suggested that constructionpermits and operating licenses in the currentsystem be combined into a single step to improve

predictability and efficiency. One-step licensing,however, raises questions on how to manage out-standing safety issues and backfits during con-struction without any guarantee that the licens-

ing process would not be even lengthier andmore uncertain.

Two othe r propo sals for cha nge s to the c urrentreg ula tory system would a llow for binding p re-approval of reactor designs and sites. Proapprovalof standardized plant designs could make the li-censing process more predictable and efficient

by removing m ost d esign q uestions from lic ens-ing. It also raises new issues, such as the degreeof specificity required for proapproval and theconditions under which a utility and its contrac-tors c ould de viate from a preap proved de sign.Proap prova l of rea c tor sites is a less c ontrover-

sial proposal. As long as safety issues related tothe combination of a site and a proposed plantare considered in subsequent licensing process,binding site approval would not detract fromplant safety. Moreover, it would contribute toshorter co nstruction Iead times since it w ould ta kesiting off the c ritica l pa th for licensing. This pro-

cedure, which is followed in France, Great Brit-ain, and Japan, could even enhance public par-ticipation by encouraging in-depth analysis at anearlier phase.

SURVIVAL OF THE NUCLEAR INDUSTRYIN THE UNITED STATES AND ABROAD

The bleak outlook for nuclear power, at leastin the near future, raises concern about thelong-term viability of the nuclear industry in the

United States and its ability to compete inter-nationally.

Reactor vendors may remain busy for manyyears by providing operating plant services andfuel loa ding. These c om pa nies are a lso e xpan d-ing their scope and competing with the servicecontractors for jobs. However, in the absence ofat least a few new-plant orders each year, thevendors will not survive in their present form.

The AEs will also have substantial work fin-ishing construction on plants now in progress,installing retrofits and dealing with special prob-lems such as replacement of steam generators.The AEs ma y find a dd itiona l ac tivity by “ rec om -missioning,” or extending the useful life of olderplants.

The c omp anies that supp ly nuc lea r com po -

nents ma y keep going by supp lying pa rts for bac k-fits and repairs, but their numbers are expected

to shrink by two-thirds in the next 3 to 5 years.Utilities will have increasing difficulty purchas-ing parts when needed and at expected costs.The c essa tion of ne w-p lant o rders ha s a lrea dycaused some shortages in parts and servicesneeded by operating plants.

Shortages are also developing in some person-nel areas. The industry has vacancies for healthphysicists and for reactor and radiation-protec-tion eng ineers, bu t it ha s a surplus of d esign en-gineers. Enrollment for nuclear engineering de-grees has de c lined since the mid-1970’s, and the

grad uate levels will barely be e noug h to fill theanticipated need for 6,000 engineers for opera-tion of plants by 1991, even if enrollments dropno mo re. With no fresh orders, the industry isnot likely to attract the best students.



Ch. 2—Summary q23

The a b ility of the nuc lea r industry to respond

to an influx of new orders depends on the lengthof t ime b efore those orde rs arrive. If utilities re-quest new powerplants within 5 years, the in-dustry could supply them, although perhapswith delays of a year or so to restart designteams and manufacturing processes. If the

hiatus in plant orders lasts 10 years, the recov-ery would be slow and not at all certain-espe-cially if U.S. vendors have not been selling reac-tors abroad in that period. In that case, U.S. util-ities may have to buy components, if not entirepowerplants, from foreign suppliers.

Many of the problems that have beset the U.S.nuclear industry have hampered the nuclear in-dustries abroad, but with less severe impact inge neral. Worldwide forec asts of the future roleof nuclear power have experienced the sameboom and bust that they have in the UnitedStates. West Germany, France, Japan, and Can-

ad a a ll are intending to c omp ete w ith the UnitedStates in wha t is expec ted to b e a very com pe ti-tive international market for nuclear plants.

In most nations with nuclear power programs,the public has expressed some opposition. Inseveral cases (e.g., Sweden), this has been strong

enough to stop new plants from being ordered.All nuclear industries have experienced delaysin building plants, but the costs have typicallybee n low er. The lic ensing p roc ess in West Ger-

many is as complex as that in the United States,but licensing in nations such as France is stream-lined by strong government control and supportand the use o f sta nda rd ized designs.

The e fforts by the m a jor rea c tor vend or in WestGerma ny to standa rd ize its plants might proveto be a useful model for the U.S. nuclear industry.The G erma n vend or plans to p roduc e a series ofpowerplants in groups of five or six whose stand-ardized features will reduce delays, engineeringworkhours, and paperwork. Each series of stand-ardized designs wo uld build o n the e xperienc eof the previous group of plants.

PUBLIC ATTITUDES ON NUCLEAR POWER

public attitudes towards nuclear power havebecome increasingly negative over the pastdecade, largely because of growing concern oversafety and economics. The most recent polls in-dicate that only a third of the public supports

construction of nuclear plants, while over sopercent are opposed (see fig. 2).

Public support is an essential ingredient in anystrategy for recovery of the nuclear power op-tion. Negative public attitudes are most directlymanifested through referenda. Although all bind-ing referenda that would have shut existing plantshave b een rejected to d ate, some ha ve bee n

close. Referenda and legislation have been ap-proved in 11 States that will prevent constructionof any new nuclear plants unless prescribed con-ditions are met. Indirectly, public worry over

nuclear risks has been a principal reason forNRC’s imposition of safety backfits to existing re-actors. State public utility commissions are unlike-Iy to adopt rate structures favoring nuclear proj-ects unless a majority of the public is in favor.

A central factor in public concern is the fearof a nuclear accident with severe consequences.Surveys indica te tha t mo st p eop le view de ath d ueto a nuclear accident as no worse than othercauses of death, but they fear nuclear plants

be c ause the tec hnology is unfamiliar and fore-bo ding. Much o f the loss of p ublic c onfide nc eis a result of a series of safety-related incidents

at several rea c tors, espe c ially the a c c ident a t TMI,and the e vide nt mishand ling of these incidentsby u tilities. The likelihood of a catastrophic ac-cident is perceived as greater than that estimatedby safety analysts in industry and government,creating a credibility gap.

Another factor in public concern about nucle-ar power is the ongoing debate about nuclearplant safety among scientists and other experts.

As the public has listened to the experts debate,they have grown increasingly dubious about plantsafety. If the experts cannot agree, the public con-cludes, then there must be a serious questionab out the safety of nuclea r po wer.




Figure 2.—Trends in Public Opinion on Nuclear Power

100

90

80

10

0

100

90

20

01975 1976 1977 1978 1979 1980 1981 1982 1983

Year of survey

Question asked” “Do you favor or oppose the construction of more nuclear powerplants?"SOURCE: Cambridge Reports, Inc.

Concerns other than accidents have causedsome people to turn away from nuclear power.Perhaps the largest concern after the possibilityof an accident is the disposal of high-level wastesge nerated by nuclear rea c tors. In ad d ition, thepo tential esthetic and environmenta l da ma gecaused by nuclear plant construction also raisesobjections. Some groups see a link between themilitary and commercial applications of atomicpower. Finally, distrust of large government andinstitutions has carried over, to some degree, toboth the nuclear industry and NRC.

People are prepared to accept some risk ifthey see a compensating benefit. The high costof some nuclear plants and current excess gen-erating capacity, however, lead many to ques-tion if there is any advantage.

While media coverage of nuclear power hasbecome more extensive in recent years, thereis no evidence of overall bias against nuclearpower. The spe c trum of op inion am ong rep orters

is the sa me as tha t for the p op ulat ion a s a w hole.Their coverage is mo re likely to reflec t tha n todetermine society’s concerns.

The credibility of both the industry and NRCis low, so words and studies alone will have lit-tle impact on the public. Steps to improve publicattitudes towards nuclear power must rely on anactual demonstration of the safety, economics,and reliability of nuclear power. If the reactorscurrently under construction experience contin-ued cost escalation, the next generation will haveto be much more economic to gain publ ic sup-port. Alternative reactor concepts that have in-herent safety features, and studies of other energysources, including analysis of the environmen-ta l costs and be nefits, also m ight help c hangepublic attitudes, though other concerns such asover waste disposal would still remain.

One of the most important steps in reducingpublic fears of a nuclear accident would be toimprove utility management of the technology.Improved management could greatly reduce thelikelihood of accidents which the public viewsas precursors to a catastrophe. While makingevery effort to minimize both minor incidents

and more serious accidents, however, the nu-clear industry should be more open about thepossibility of accidents.



Ch. 2—Summary q25

Improved communications with nuclear respond to the substance of critic’s concernscritics might also alleviate public concerns about could reduce acrimonious debate which con-reactor safety. A concerted effort to identify and tributes to negative public opinion.

POLICY

Further orders for nuclear powerplants are un-likely without some government action and sup-port. If Congress chooses to improve the chances

of nuclear reactors being ordered in the future,Federal initiatives could be directed to the fol-lowing goals:

q red uc e c ap ital c osts and unce rtainties,q improve reactor operations and economics,q reduce the risks of accidents that have public

sa fety o r utility financ ial imp ac ts, andq alleviate public concerns and reduce polit i-

c a l risks.These g ene ra l goa ls are ne ither new nor as con -troversial as the specific steps designed to achieve

the goa ls. The initiat ives d isc ussed in this rep ortthat-are likely to be the m ost effec tive a re:

1.

2.

Support a design effort to re-optimize re-actor designs for safety, reliability, andoverall economy. This initia tive wou ld e x-tend the efforts of the reactor vendors. De-signs would incorporate the backfits thathave occurred in existing plants and addressthe outstanding safety issues, thereby signif-

icantly reducing the possibility of costlychanges during construction and the con-cern for safety in current LWRs. It would beexpensive, however, especially if a demon-stration p lant is nec essa ry.Improve the management of reactors underconstruction or in operation. Inadequatemana gem ent has bee n one of the majorcauses of construction cost overruns and er-ratic operation. Efforts are underway by theNRC and INPO to upgrade reactor manage-ment, and they should show results in im-proved training programs, better quality con-

trol and more reliable p erforma nc e. Thecongressional role in improving reactor man-agement would be oversight of the NRC, andsupport for improvements in analytical tech-niques and resolution of the remaining safety

issues.

OPTIONS

Photo credit: William J. Lanouette

This photo shows the campaign headquarters ofthe Project Survival Group which supported

Proposition 15, a California referendum prohibitingfurther construction of nuclear plants without adefinitive resolution of nuclear waste storage.The referendum was defeated in 1976, but a

similar California law was recently upheldby the Supreme Court

3. Revise the regulatory process. Many of thedifficulties experienced with licensing would

be avo ide d with future p lants throug h im-proved technology and management. Im-proving the predictabil ity and efficiency oflicensing is a prerequisite for any further

orde rs, how eve r. To a la rge extent this c ouldbe done administratively by the NRC without




4.

legislation. Som e of the eleme nts suc h a sc onsistent b ac kfit eva luations and preap -proval of reactor designs and sites discussedin the regulatory section above will probablyrequire at least strong congressional over-

sight and possibly legislation. Legislation thatmakes the process inflexible or restricts pub-

lic access could be counterproductive.Certify utilities and contractors. If effo rts toimprove reac tor manag ement a re o nly pa r-tially successful, stronger measures could bewarranted. A poorly performing utility canaffect the entire nuclear industry through theresponse of the public and the NRC to inci-dents with sa fety imp lic a tions. The NRCmight c onside r withdrawing the op eratinglicenses of utilities that do not demonstratecompetence or commitment in managingtheir nuclear plants. Evidence of capabilityof the utility and its contractors might be

made a prerequisite for a new constructionpe rmit to a lleviate c onc erns of the p ublic ,investors, and c ritic s abo ut the qua lity a ndc ost of the p lant.

5. Support R&Don new reactors. Some newreactor concepts have features that, ifproven out, could make them inherentlysafer than current operating plants, thusalleviating some of the concerns of the util-ities and the public. If advanced LWRs donot appear adequate to overcome theseconcerns, then the availability of an alter-native reactor, such as the HTGR, would beimportant. Research, development, and

demonstration of these technologies will benecessary to make them available.

6. Address the concerns of the critics. Im-proved public acceptance is a prerequisitefor any new orders. At present, the publicis confused by the controversy over safetyand is therefore opposed to accepting therisks of new reactors. The best way toreduce controversy would be to resolvesome of the disagreements between thenuclear industry and its critics. This could

be initiated by involving the critics moredirectly in the regulatory proc ess. Involv-ing knowledgeable critics in regulation orin the design and analysis of new reactors

7.

ma y be the o nly wa y to a ssure the pub licthat safety concerns are being addressedadequately.Control the rate of nuclear construction.Many of the concerns over nuclear poweroriginated from the early projections ofrapid growth, and expectations of a per-

vasive “ nuclear ec onom y. ” The p resentmodest projections appear less threatening,

but some people will oppose all nuclearpower as long as a major resurgence is

possible. Controlling the growth rate mightalleviate these concerns, thus reducing thecontroversy, rebuilding public acceptance,and making some new construct ionpossible.

None of the options described above will bevery effective by itself. Some c ould b e ve ry dif-fic ult to imp lem ent. It ap pea rs at least p ossible,

however, that combinations of these optionscould contribute to a much more favorable envi-ronment for nuclear power.

Whether any of these strategies would “work”is a func tion o f seve ral fa c tors including:

the extent to which Federal policy strategiesresolve the problems and make nuclear

power more attract ive,the electricity demand growth rate and theeventua l need for new p ow erplants, andthe improvement in designs and operationsin the ab senc e of p olic y

-initiatives.

The future o f the nuc lea r industry will beshaped by t he e volution of the se fa c tors. The

de gree to which the Fed eral Government shouldbecome involved (the first factor above) dependson an assessment of the uncertainties surround-ing the other two factors. Under some conditions(e.g., relatively rap id growth in d ema nd for elec -tric ity, reliable op eration of existing p lants andimproved technology available) a revival is quitepossible. Under other conditions, even a stronglysup portive po lic y strate gy co uld fa il. Suc c essfulimp leme ntation of a ny strate gy will dep end on

how well the concerns of all interested partieshave been addressed.



Chapter 3

The Uncertain Financial andEconomic Future

Photo credit: United Engineers & Constructors



Contents

Page

The Rec ent Past: Utilities Have Built Fa r LessTha n They Planned . . . . . . . . . . . . . . . . . . . . . 29

The Unc ertain O utlook fo r Elec tricity Dem a nd 31

The Top -Down Perspec tive o n Elec tricityDemand–Sources of Uncertainty . . . . . . . . 33

The Botto m-Up Perspec tive on Elec tric ityDem a nd —Source s of Unc erta inty . . . . . . . . 35

Co nc lusion . . . . . . . . . . . . . . , . . . . . . . . . . . . . 41

Reserve Margins, Retirements, and the Needfor New Plants . . . . . . . . . . . . . . . . . . . . . . . . . 42

Rate Regulation and Powerplant Finance . . . . . 46Utilities’ Current Financial Situation . . . . . . . . 46Implications of Utilities’ Financial Situation . . 50Possible Changes in Rate Regulation . . . . . . . 52

The C ost o f Build ing a nd Op eratingNuc lea r Powerpla nts . . . . . . . . . . . . . . . . . . . . 57The Rapid Increase in Nuclear Plant Cost . . . 58Reasons for increased Construction Cost . . . . 60The Increase in Nuclear Construction

Lead times . . . . . . . . . . . . . . . . . . . . . . . . . . . 62The Imp a c t o f Dela y on C ost . . . . . . . . . . . . . 63The Cost of Electricity From Coal and

Nuc lea r Plants. . . . . . . . . . . . . . . . . . . . . . . . 64Future Construction Costs of Nuclear

Powerpla nts. . . . . . . . . . . . . . . . . . . . . . . . . . 66The Financial Risk of Operating a

Nuclear Powerplant . . . . . . . . . . . . . . . . . . . 68Impact of Risk on the Cost of Capital . . . . . . 70

Nuclear Power in the Context ofUtility Strategies . . . . . . . . . . . . . . . . . . . . . . . . 71Imp lic a tion s fo r Fed eral Polic ies , , . . . . . . . . . 74

Ch a p te r 3 Refere nc es . . . . . . . . . . . . . . . . . . . . . 76

Tables

Table No. Page

3. Growth Rates in Electricity DemandGiven Different Price Responses . . . . . . . . . 34

4. Estimated Impact of Industrial Electrotech-nologies in the Year 2000 . . . . . . . . . . . . . . 38

5. Efficienc y Improve me nt Potentia l From Typic alto Best 1982 Model: Household Appliances 40

6. Possible Needs for New Electric-Generat-ing Capability to Replace Retired Powerplants,

Loss of Powerplant Availability, and Oil andGas Steam Powerplants . . . . . . . . . . . . . . . . 44

7.installed Capacity and Net Electricity Gen-eration b y Type of Ge nerating Ca pa c ity,1981-82 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45

8.

9.

10.

11.

12.

Numbers of 1,000 MW Powerplants Neededin the Year 2000 ....., . . . . . . . . . . . . . . . . 46Additions to Overnight Construction CostDue to Inflation, Escalation and InterestDuring Construction . . . . . . . . . . . . . . . . . . . 64Nuclear Plant Property and LiabilityInsura nc e . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68

Six Sets of Alternative UtilityCo nstruc tion Plans . . . . . . . . . . . . . . . . . . . . 73Cost and Volume of Various LoadMa na gem en t Devices . . . . . . . . . . . . . . . . . . 74

Figures

Figure No. Page

3. Planned Construction of Coal and

4.

5.

6.

7.

8.

9.

10

11.

12.

13.

14.

15.

16.

17

18

Nuclear-Generating Capacity, 1982-91 . . . . 30Comparison of Annual Ten-Year Forecastsof Sum me r Pea k Dem and . . . . . . . . . . . . . . 30A Comparison of the Growth Rates ofReal GNP and Electricity Sales, 1960-82 . . . 32Penetration of Air-Conditioning and

Electric Heating in Residential Sector . . . . . 39Projected Generating Capacity andAlternative Projections of Peak Demand . . . 43The Energy Source and Age of ExistingElectricity Generating Capacity . . . . . . . . . . 44The Electric Utility industry is Dependenton Externally Generated Funds to a GreaterExtent Than Other Large Capital Users . . . . 48Since 1967, the Fraction of Utility AssetsTied Up in “Construction Work in Progress”Has Ste a d ily Inc rea sed . . . . . . . . . . . . . . . . . 48Beginning in 1972, the Utilities’ Real Returnon Equity Has Been Eroded by Inflation . . . 48Since 1973, the Average Utility Stock Has

Sold Below its Book Va lue. . . . . . . . . . . . . . 49In 1970, a Majority of Utilities Were RatedAA or Better; in 1982, a Majority AreRated A o r Worse . . . . . . . . . . . . . . . . . . . . . 49Ov er Time , the Sha re o f AFUDC in Tota lEarnings Has Increased and the Share ofCash Has Fa llen . . . . . . . . . . . . . . . . . . . . . . 49Costs of Nuclear Units With ConstructionPermits Issue d , 1966-71 . . . . . . . . . . . . . . . . 59Total Capital Costs for Nuclear PlantsCompleted in 1971, 1978, and 1983-87in 1982 Dollars Without InterestDuring Construction . . . . . . . . . . . . . . . . . . . 59Trends in Material Requirements in

Estimates of PWR Construction Cost . . . . . . 61Construction Leadtimes for NuclearPowerplants . . . . . . . . . . . . . . . . . . . . . . . . . , 63



Chapter 3

The Uncertain Financial and Economic Future

If signif icant new electric-generating capacityis need ed in the 1990’s and be yond, and if nu-c lea r pow er is to p rovide a ma jor frac tion o f thatcapacity, utilities must order reactors some time

before the end of the decade. Util ity executiveswill co mp are the reliab ility, safety, a nd pub lic ac -c ep tanc e o f nuc lea r pow er (issues that are ex-plored in other chapters in this report) relative

to other types of generating capacity, especiallyto coal. But above all, they will treat the ques-

tion of whether to order a nuclear plant as a stra-teg ic d ec ision to b e ta ken in the c ontext of ex-pec ted d ema nd for elec tricity and the c urrent andfuture financial status of the utility. Utility ex-

ec utives will c om pare the risks and returns from

construction of more nuclear powerplants withthe risks and returns of severa l othe r op tions formee ting their pub lic o bliga tion to p rovide a de -

quate electric power.This c ha pter will explore the e leme nts of stra-

tegic choice for utilities that arise from the uncer-ta inty of future ra tes of g row th in elec tric ity de-mand, the uncertainty of economic return on in-vestment, and the uncertainty of construction andop erating c osts of nuc lear pow erplants. The c hap -te r will also a ssess the reg iona l and na tiona l im-plications of individual utility choices.

THE RECENT PAST: UTILITIES HAVE BUILT FAR LESSTHAN THEY PLANNED

Utilities have built far less new electric-generat-ing capacity in the 1970’s and early 1980’s thanthey expected to a decade ago. In 1972, the peak-year of generating capacity forecasts, utilities

overestimated construction in the last half of the1970’s by 25 percent and overestimated construc-

tion in the first half of the 1980’s by 60 percent(46). Utility pred ic tions of future g ene ra ting c a -

pacity declined over the decade but sti l l greatlyexcee de d a c tual construction. The ac tual gener-

ating ca pa city of 580 giga watts(GW) * in the sum-er of 1982 was about 75 GW lower than whathad bee n forec ast a s rec ently as 1977 (70).

Both nuclear and coal plants were canceled inthe late 1970’s and early 1980’s, but nuclearplants were canceled in far greater numbers andwith far greater cost in sunk investments. Over100 nuclear units were canceled from 1972 to1983, more than twice the number of coal units(29,35). Almost $10 billion (1982 dollars) in invest-ment costs was tied up in the 26 sites (some with2 units) where at least $50 million per site had

already been spent (35). Another 11 nuclear unitstotaling $2.5 billion to $3 billion in construction —

*One gigawatt equals 1,000 MW (1,000,000 kW) or slightly lessthan the capacity of the typical large nuclear powerplant of 1,100to 1,300 MW. GW as used in this chapter always refers to GWeor gigawatts of electricity.

costs are expected to be canceled and another5 units with $3 billion to $4 billion costs may becanceled (35,53).

Of 246 GW of orders for nuclear plants everplaced , about 110 GW have bee n c anc eled. Allof the 13 orders plac ed sinc e 1975 have be encanceled or deferred indefinitely and no neworders have be en p lac ed sinc e 1978.

Utilities expec t to c omp lete ma ny of the nu-clear plants under construction but have notplanned to build any more. Utilities still, however,are planning to build new coal plants. A total ofabout 43 GW of coal construction is planned forc om pletion from 1988 to 1991 (see fig. 3) b ut o nly11 GW of nuclear capacity is planned (much ofwhich is likely to be canceled) (68).

One obvious reason for so many canceled anddeferred plants is that from 1973 to 1982, elec-tricity load grew at less than half the pace (2.6pe rc ent p er yea r) that it had grown from 1960to 1972 (7.1 percent per year). Most utilities in

the early 1970’s used simple trend-line forecaststhat took neither gross national product (GNP)

or respo nse to elec tricity pric e into a c c ount. Theywere unprepared for the change in electricity

growth rates. Figure 4 shows a remarkable down-

29




Figure 3.– Planned Construction of Coal and Nuclear-Generating Capacity, 1982=91

1982- 1984- 1986- 1988- 1990-1983 1985 1987 1989 1991

SOURCE: North American Reliability Council, Electric Power Supply an dDemand 1982-1991, August 1982.

Figure 4.—Comparison of Annual Ten-Year Forecastsof Summer Peak Demand (contiguous U.S.)

800

700

650

400

350

30074 75 767778798081828384858687 8889909192

SOURCE: North American Electric Reliability Council, Electric Power Supply and Demand, 1981-1991, August 1982 and advanced release of theprojections for 1983-1992 in April 1983.

ward trend in utility forecasts of future loadgrowth. it is no surprise that utilities, making plansin the late 1960’ s and ea rly 1970’s, and unab leto foresee the economic changes of the 1970’s,planned to build more generat ing c ap ac ity thanwa s eventua lly need ed . Even w ith all the c anc el-lations and deferrals, there was almost a

50-percent increase in electric-generating capac-ity from 1973 to 1982 while the a verag e reservemargin* increased from about 21 percent in 1973to about 33 percent (68,70). In between, thereserve margin peaked at about 37 percent as util-ities c omp leted and brought o nline p lants thathad begun construction before the slower loadgrowth was recognized as the norm rather thanas an anomaly (58).

In a dd ition to the slowd own in elec tric loa dgrowth, powerplants also have been canceledand deferred due to the widely acknowledgeddeterioration in the financial condition of utilities.

At the be ginning o f the 1970’s, utilities enjoyedgo od financ ial hea lth. Almost 80 percent hadbo nd ra tings (Sta nda rd an d Poo rs) of AA o r AAA.Utility stoc k sold we ll ab ove boo k value so t ha tthere was little difficulty financing new generatingcapacity from new issues of either debt or equity.

By 1981, however, utilities were in a greatlyweakened financial condition. Currently, therea re no elec tric utilities with AAA bo nd ra ting s a ndless than a fourth with AA ratings. At its low pointin 1981, utility stock sold on average at only 70pe rc ent o f b oo k value. This me ant tha t a ny issue

of new stock to pay for new generating plantwould dilute the value of the existing stock (seeb ox A la ter in this c ha p ter). The fina nc ial de terior-ation was influenced in part by the general finan-cial conditions of the decade, especially the rapidrates of inflation and the poor performance of thestoc k ma rket. Beyond the influenc e o f ge neraleconomic conditions, however, the financialstatus of utilities deteriorated because of the enor-

*“Reserve margin” is defined as the percent excess of “plannedresources” over “peak demand” where “planned resources” in-cludes instaIled generating capacity plus scheduled capacity pur-chases less sales.



Ch. 3—The Uncertain Financial and Economic Future q 31

mous strain of financing requirements for newplants. Despite many cancellations and deferralsof p owe rplants, new plant financ ing increasedfrom a bout $10 billion in 1970 to over $28 billionin 1982, of which more than two-thirds had tobe fina nc ed externally (45).

The bigg est ince ntive for utilities to c an c el an dpostpone more nuclear than coal plants, was themore than fivefold increase in the constant dollarcost of nuclear plants from plants completed in1971 to p lants sc hed uled fo r c om p letion in the1980’s com pa red to the a pp roxima tely threefoldincrease in the constant dollar cost of coal plants

(55,56). Another incentive to favor coal plantswa s their shorter Iea dt ime, an a verag e o f 40 to

50 percent fewer months than for a nuclear plant(l).

Loo king a hea d , the p rospec ts for substantial

numbers of new central station powerplants ap-

pea r fa irly unc erta in. The p rospec ts for morenuclear p lants app ea r even m ore unc ertain. Thereasons why this is so are laid out in the rest of

the c hap ter. The next sec tion desc ribes the unc er-ta inty ab out the futu re g row th rates in elec trici-ty demand. With some assumptions about thefuture it is reasonable to expect that the fairly slowgrowth rates (1 or 2 percent per year) of the pastfew years will continue. With equally plausibleassumptions, however, electricity load growthc ouId resume at rate s of 3 to 4 pe rc ent p er year.The source s of unc ertainty a re desc ribed in the

next section.

The third sec tion explains why utilities c an a f-

ford to wait awhile before ordering powerplantsin large numbers, since reserve margins are nowso high. Sooner or later, however, as this sectionpoints out, some number of new powerplants willneed to b e b uilt to replace ag ing p owerplants andmee t e ven m od est increases in elec tric loa d.

The fo urth sec tion of the c ha pter presents anargument that there may be systematic biases in

rate reg ulation tha t d isc ourage those types ofgenerating capacity that are of high capital costand high risk relative to other types. Over the longrun such rate regulation would discourage fur-ther construction of large coal and nuclear plants

even when increased load and replacement ofexisting plants would make it sensible to construct

central station plants, for which capital costs arehigh rela tive to fuel co st.

The fifth sec tion o f the c ha pte r lays out the un-certainties involved in constructing and operatinga nuclear plant which discourage utilities fromordering more nuc lea r plants even when theydecide to order more central station powerplants.

Construction c osts ha ve risen muc h fa ster tha ngeneral price increases and vary severalfold fromplant to plant even when built the same year. Inaddition, there is a financial risk of at least severalbillion do llars from an ac c ident tha t d isables a

powerplant and more from one that causes dam-ag e to pub lic hea lth a nd prope rty. To da te insur-ance is available to cover only a fraction of thisrisk.

Given the uncertainties of demand and nuclearc onstruction c ost, utility dec isions to c anc el nu-c lea r po werplants and some c oa l plants havebee n sensib le, and in the short-te rm interests ofthe ratepayers. Over the long run, however, ifratemaking discourages electric-generating tech-

nologies of greater capital cost and greater risk,further investment in nuclear powerplants could

be disc ourage d even if it we re in the longer terminterests of ratepayers.

The fina l sec tion of t he c hap ter de sc ribes the

choices utilities have and the choices they seemto be making. Under a few specific assumptionsabout changes in outside circumstances and rateregulation incentives, utilities could order nuclearplants ag ain. It a pp ea rs now, how ever, that the ywill avoid central station construction as long aspossible and then build coal plants.

THE UNCERTAIN OUTLOOK FOR ELECTRICITY DEMAND

From 1973 to 1982, annua l inc reases in e lec -tric ity de ma nd ave rag ed 2.6 perce nt. If these

growth rates were to c ontinue for the next 20years, they would provide no more than a weak




stimulus to further building of central stationpo we rplants, inc luding nuc lear pow erplants. (See

the detailed discussion of capacity requirementsin the next section.) It would be possible for mostutilities, with som e effort, to a void build ing c en-tral station powerplants altogether until the late1990’s by encouraging conservation, load man-

agement, cogeneration, and small sources ofpower from hydro and wind; or by purchasingfrom U.S. utilities with exce ss c apac ity or from

Ca nad ian utilities; or by keep ing existing p lantsonline pa st no rma l retirem ent ag e. These stra te-gies a re d isc ussed later in the c hap ter.

What are the chances that the average elec-tricity demand growth rate will be significantlyhigher or lower than 2.6 percent per year? A sig-nific antly lower growth rate wo uld ma ke it d iffi-cult to justify any major construction of centralstation powerplants. A significantly higher growthrate wo uld ma ke a strate gy o f lit t le o r no po we r-plant c onstruc tion d iffic ult to susta in.

Published projections of electricity demandreflect considerable uncertainty about futuregrowth rates. As is clear from figure 4 above, theutilities’ o wn e stimate s of future pea k dema ndhave d rop ped eac h year since 1974 and now

average an annual increase of 2.9 percent from1983 to 1992 (70). Some studies (e.g., the EnergyInformation Agency and Starr and Searl of EPRI)project higher rates of electricity growth than theelectric utilities do, although none project morethan 4 percent annual growth through 1990 (27,

51 ,83). Only one (Edison Electric Institute) projects more than 5 percent from 1990 to 2000. Sev-eral studies (e.g., Audubon and the Solar EnergyResearch Institute) on the other hand, project

very low ra tes of a nnua l elec tric ity grow th of Oto 1.5 percent (77,84).

One of the reasons for this range is differentassumptions about the future growth rates in

GNP. The p rojec tions of fa ste r elec tric ity g rowthassume a range of 2.5 to 3.0 percent annual GNPgrowth p er yea r (51). The p rojec tions of slowe relectricity growth assume somewhat slower

growth rate in real GNP, a range of 2.0 to 2.8percent per year (77). In general, however, allthese projections assume that the United Stateshas a “ ma ture” ec onom y and inc rea ses in rea l

GNP faster than a n a verage of 3.0 percent pe ryear are not likely.

The p rojec tions d isagree more signific antlyabout the likely future relationship betweengrowth rates in GNP and growth rates in elec-tricity demand. Projections of faster electricitygrowth assume that electricity will increase fasterthan GNP. Projections of slower electricity

growth asume that electricity demand will in-c rease signific antly less fast tha n GNP.

The ratio be twe en e lectric ity grow th ra tes an dGNP growth rates has indeed dropped since the1960’s. As is shown in figure 5, electricity growthrates were about double GNP growth rates in the1960’s and approximately equal to GNP growthrate s (exc ep t fo r rec ession ye a rs) in the 1970’s.Those expe c ting fa st g row th rate s in e lec tricityregard the late 1970’s as an anomaly and expect

a resurgence of a ratio of electricity to GNP

grow th o f mo re t han 1.0. Those expe c ting slowgrow th rates in elec tricity assume tha t the ra tioof electricity growth to GNP growth will fall stillfurther, well below 1.0. They expe c t tha t e lec tric i-ty will continue to behave like other forms of en-ergy for which ratios to GNP have dropped stead-ily sinc e 1973.

The source s of unc ertainty in elec tric ity de -ma nd forec asts are very evide nt from a loo k at

Figure 5.—A Comparison of the Growth Rates of RealGNP and

-Electricity Sales, 1980=82

1 1 1 1

1960 62 64 66 68 70 72 74 76 78 80 82SOURCE: Craig R. Johnson, “Why Electric Power Growth Will Not Resume, ”

Public Utilities Fortnigfhtly, Apr. 14, 1983 from data in the EIA, AnnualReport to Congress 1982 and the Edison Electric Institute,



Ch. 3—The Uncertain Financial and Economic Future q33

the uncertainty surrounding the factors under-lying the forec asts. The unc erta inty exists bothin convent ional macroeconomic approaches toforecasting which relate electricity growth to ex-pected changes in GNP, electricity prices and thepric es of comp et ing fuels. (This is some time s re-ferred to a s the to p-do wn a pp roa c h.) There is

comparable uncertainty about the factors under-lying engineering or end-use projections of elec-tric ity use. This ap proa c h (som etime s c alled bo t-tom-up analysis) identifies possible technicalchanges in the use of electricity that are econom-ically feasible—such as improvements in appli-ance and electric motor efficiency, opportunitiesfo r fuel switching, and new electricity-using in-

dustr ia l technologies–and then estimates the

likely market penetration of these changes.

The Top-Down Perspective onElectricity Demand—Sources

of Uncertainty

One of the advantages of top-down analysis ofelectricity demand is that uncertainty is confinedto only a few powerful variables—future growthin GNP, changes in electricity prices, and changes

in the prices of competing fuels and the respon-siveness of electricity demand to each.

The Influence of Economic Growth. -Futuregrow th o f GNP is a ma jor source of unc ertainty,both because income and industrial productionare assumed by economists to have major im-

pacts on electricity demand, and because of somedeep uncertainties about the future direction ofthe economy. Even the fairly narrow range of

GNP growth rates of 2 to 3 percent that has beenassumed by the major electricity demand projec-tions implies a range of electricity demand growthrates (assuming no price influence) of about 2 to3 perce nt over the long run if electric ity de ma ndfollow s the income response pa tterns identifiedin the past (79). Many observers concede therang e is even wider, Those w ith private m isg iv-ings about the future health of the economy ac-cept the possibility of an annual rate of GNP

growth lower than 2 percent. Optimists abouteconomic renewal and increased productivity

suggest the potential for a higher rate of growth.

Regional uncertainties about economic growthare mo re e xtreme than national ones. Inco meand industrial output have fallen in some regionsas a result of the recent recession and the extentof long-term recovery from the recession in thesereg ions is unc lea r. Rap id p op ulation g row th is ex-pe c ted to o c c ur in the South a nd Southwe st.

Electricity Prices.– Future electricity prices and

their imp ac ts are a sec ond source of unc ertaintya bo ut elec tric ity de ma nd g row th. This is b oth b e-cause there is disagreement about future changein electricity prices and because there is uncer-tainty about how electricity demand responds toelec tric ity p rices.

From 1970 to 1982, averag e elec tric ity pric es

increased in constant dollars at about 4 percentper yea r, reve rsing a 20-year trend o f dec rea s-ing rea l p ric es (26). The re is c onside rab le d is-

agreement about the future course of electricityprices even though they should be easier to pro-

ject than oil or gas prices because they are largelydetermined by regulatory rules that are predict-ab le. The Energy Information Administration in

the Department of Energy (DOE) has consistent-ly projected very slow increases in real electrici-ty p rices of less tha n 0.5 perce nt p er yea r until

1985 and 1.4 pe rcent p er yea r after tha t (27). TheOffice of Policy Planning and Analysis, also ofDOE, projected somewhat more rapidly increas-ing electricity prices, at 2.4 percent per year un-til 1995 with level prices after that (20). Finally,

Data Resources Inc. (DRI) has projected sharplyincreasing electricity prices for both industrial andresidential users of 3.7 percent per year until themid-1 980’s, slower increases until 1990 and less

than 0.5 percent per year from 1990 to 2000 (18).

Forecasts of electricity prices disagree principal-ly about the future cost of coal for electricity, thefuture construction cost of nuclear and coal pow-erplants and the future rate regulation policiesof Pub lic Utility Comm issions. Sta b ilizing of elec -tricity prices in the 1990’s is expected to occurbecause of a growing share of partially depreci-

ate d plants in the rate ba se a nd little ne w construc tion. (See the d isc ussion o f the se fac to rs inlater sections of this chapter.)




Response of Electricity Demand to ElectricityPrices.—There is generally less agreement about

the impact of electricity prices on electricity de-mand than there is about the impact of changesin GNP. Most analysts agree that the short-runresponse of electricity demand to an increase inelectricity prices is very limited—l O to 20 percent

of the price increase. Based on comparisons fromState to State, however, analysts estimate thelong-run response to be much greater–50 to 100percent of the price increase. (The response toa price increase is always a decrease in demand.)For example, an increase of 2 percent in elec-tricity prices would be expected to result in ashort-run d ec rea se in elec tric ity dem and of 0.2to 0.4 percent (from what it would have beenotherwise), but a long-run decrease in electrici-ty dema nd of 1 to 2 percent.

Prices of Competing Fuels.–Very few analysts

have attempted to estimate the long-run responseof electricity demand to changes in the prices ofother (and competing) forms of energy of whichthe principal competitor is natural gas. Of theseattempts, the consensus is that electricity demandwould be expected to increase (over the long run)ab out 0.2 pe rc ent for every 1 percent inc rea sein natural gas prices.

The Energy Informa tion Ad ministration p rojec tsthat natural gas prices will increase at more than10 percent a year (in constant dollars) until 1985and mo re slow ly, at 5 percent p er year after that

until 1990 (27). DRI, on the other hand , projec ts

som ew ha t slower inc rea ses of about 3 pe rc entper year through the 1980’s and 1990’s (18).

In some areas, especially New England, oil is

the chief competitor to electricity and the chiefsource of unc ertainty. Oil prices a re no w highe rthan natural gas prices and are projected to in-c rea se b ut mo re slow ly tha n na tural ga s pric es.

The Impact of Prices on Demand .–The com-bined effect of uncertainty about future electric-ity and natural gas (and oil) prices and uncertain-

ty about how electricity demand responds tochanges in electricity prices is enough to explaina range of uncertainty in electricity demand fromvery slow g rowth to quite rap id g rowth. This isillustrate d in tab le 3. If GNP is p rojec ted to growat 2.5 percent (the midpoint of the range assumedin current forec asts) a nd the long-run respo nseto increases in income is assumed to be 100 per-cent, the effect of price and price response as-

sump tions is to p rod uc e a projec tion of 1.1 per-cent annual increase in electricity demand, at thelow end, and of 4 percent at the high end. Itwould be possible, for example, for electricity de-ma nd to grow a t 4 percent p er yea r, if elec tric itypric es increase a t no m ore than 1 pe rc ent p eryear (in constant dollars) while natural gas pricesincrease a t 10 perce nt per yea r, and there is a

relatively small long-run decrease in demand inresponse to an electricity price increase (definedas a long-run elasticity of – 0.5).

Timing of Respo nse to Pric es.–Unfo rtuna te ly

no analyses have been published of the long-run

Table 3.—Growth Rates in Electricity Demand Given Different Price Responsesa

Annual increase in electricitydemand given:

Rates of electricity and gas Low long-run High long-runprice increase price response price response

1. High electricity prices (2%/yr)Moderate gas prices (3%/yr) . . . . . . . . 2.1%/yr 1.1%/yr

2. Low electricity prices (1‘\O/yr)Moderate gas prices (3%/yr) . . . . . . . . 2.6%/yr 2.1%/yr

3. Low electricity prices (1%/yr)High gas prices (5%/yr). . . . . . . . . . . . . 3.0%/yr 2.5%/yr

4. Very high gas prices (10%/yr)Low electricity prices (1%/yr). . . . . . . . 4.0%/yr 3.5%/yr

aAl~O ~alled price elaStiClt&. See Assumptions.

Assurrrptlons: GNP increases at 2.5 percent par year; Income elasticity of electricity demand - 1.0; response of electricitydemand to gas price (cross-elasticity) - 0.2; response of electricity demand to electricity price (own-elasticity): a) low= 0.5, b) high = –1.0.

SOURCE: Off Ice of Technology Assessment based on a presentation by James Sweeney to an OTA workshop.




response of electricity demand to electricityprices in the crucial decade since the 1973 oilemb argo. The e stima tes mentioned ab ove we rea ll based on d a ta up to 1972. The c hief rea sonis because this analysis cannot be done withoutconsideration of the timing of the long-run re-sponse. Not only have electricity prices (in con-

stant dollars) changed from decreasing to increas-ing, but c om pe ting fuel pric es (whic h a lso a ffectelectricity demand) changed from decreasing toincreasing even more dramatically.

The length of time it takes for the long-run pric eresponse to be felt is crucial to making any esti-mate of this response. If the “long run” is 3 to4 years, we ha ve a lrea dy seen m uch of the re-

sponse to the price increases of the 1970’s. If the“ long run” is 10 yea rs or longer, we a re just nowbeginning to witness the effects of actions takenin response to those price increases.

This lac k of und ersta nd ing of ho w long it ta kesfor the full long-term response to increases inelectricity prices makes it difficult to predict whatstill remains of consumer and industrial responsesto the increasing electricity prices of the lastdecade. If the response takes 10 years or longer,the effects will last until the early 1990’s.

Elec tric ity Ra te Struc ture.-The unc erta inty o fprice impacts is further complicated because fore-casts of average electricity price do not fully cap-ture the potential price changes that will influ-ence electricity demand. Decisions of industry

and consumers are also influenced by the priceof an additional unit of electricity, that is the marginal price of electricity. Utilities have recentlybegun to shift from “declining block rate” struc-tures (in which e ac h ad d itiona l bloc k of units ofelec tric ity costs less tha n p revious b loc ks) to in-creasing block rate structures (in which each ad-ditional block costs more than previous blocks).There is no c urrent survey o f da ta on utility ra testructures, but a crude estimate can be made thatas many as one-fifth of all utilities may have in-c reasing bloc k rates for househo lds. The numb erof such utilities is likely to continue to grow.

Regional Differences in Demand Response toPrice Increases for Individual Utilities. -Anothersource of uncertainty is that individual utilities will

have very different experience in electricity prices

from the national average. A recent regional anal-ysis of projected changes in electricity pricesshows a mixture of declining electricity prices insome regions and increasing electricity prices inothers (48). Rea l elec tric ity p rices in the Moun-tain region are forecast to drop by an average ofmore than 3 percent per year (in constant dollars)

until 1987 and then stay nearly stable until 2000.Mea nwhile, in the West South Cen tra l reg ionelectric ity pric es are p rojecte d to increase b y anaverag e o f 4.6 perce nt per year until 1991 andthen taper off slowly until the year 2000. Pricechanges as different as these will inevitably in-duce a wide regional variation in demand growthrates. Declining rates in the Mountain region

should eventually stimulate increases in electricityde ma nd while the op po site o c c urs in the WestSouth C ent ral region . This regiona l va riation inbo th p resent a nd projected electric ity pric es willcomplicate and perhaps delay industry’s invest-ments to improve e ffic ienc y.

The Bottom-Up Perspective onElectricity Demand—Sources

of Uncertainty

Within an overall framework of economicgrowth rates and changes in relative energyprices, bottom-up or end-use analysis offers acloser look at how electricity customers might ac-tua lly cha nge their pa tterns of e lect ricity use inresponse to prices and income changes. Industrial

c ustom ers purc ha se the m ost elec tric ity, ab out38 perce nt of the app roxima tely 2.1 billion kwhsold in 1981. * Residential customers are close

be hind w ith 34 percent of all sales in 1981. Com-mercial customers and other customers pur-chased 24 and 4 pe rc ent, respe c tively.

Given a range of plausible assumptions abouthow customers are likely to change their patterns

of electricity use over the next two decades,growth rates in electricity demand that range from1 percent p er yea r to a s high a s 4 percent p eryea r a re po ssible.

From a close look at each sector it is clearthat a few variables are far more important than

* 1981 data are used because industrial purchases had fallen to35 percent of the total in 1982 as a result of the economic recession.




others. Industrial electricity sales will be strong-ly influenced by the output experience of severalkey industries such as steel and aluminum, by themarket penetration of greater efficiency in the useof electric motors, and by the success of severalimportant new electrotechnologies for replacingoil and natural gas sources of industrial process

heat.The fut ure rate o f household forma tion w ill a lso

strong ly influenc e residential sa les. How fast theuse of electric heat and central air-conditioningsp read s into new and existing housing units is a lsoimportant, as is the success of high-efficiency ap-p lianc es, air-co nd itione rs, and hea t pump s.

For future c om me rc ial sa les of elec tric ity theimportant future influences will be: the rate ofconstruction of new commercial buildings; theprospects for significantly more efficient air-con-ditioning and lighting; and the potential for signifi-

cant increase in electricity used for computersand other automation.

Utilities, in fact, are beginning to monitor suchvariables more closely in the effort to reducesome of the uncertainty about future growth ratesin elec tric ity. Utilities are inc reasingly turning toend -use m od eling o f future dem and . This is in

part because such models can be used to assessthe impact of uti l i ty conservation and load man-ag em ent p rog ra ms, but it is also po ssible to m oni-tor important indicators of future customer be-havior. Some utilities and State governments are

indeed undertaking load management and con-servation programs in order to directly influencecustomer behavior and reduce future uncertainty.

There is gene ral ac c ep tanc e that the high a nd

low end of the bottom-up range is influenced bythe key variables of price and income (GNP) usedin top -do wn ana lysis. Slow inc rease in elec tric i-

ty prices is a weak stimulus to improvements inthe e ffic iency of elec tric ity use even if they are

technically feasible, while rapid electricity priceincrea ses a re a strong stimulus to effic ienc y im-provem ents. Simila rly, rap id inc reases in na turalgas prices relative to electricity prices will stimu-

late the adoption of new electrotechnologies thatsubstitute for natural gas. Less rapid price in-creases will provide less stimulus.

Industrial Demand for Electricity .-The elec-tric-intensity of U.S. industry (electricity used per

unit of output) has held steady since 1974. Thisis despite a steady decrease for more than twode c ad es in the overall use o f energy p er unit ofindustrial outp ut. The stead y rela tionship of e lec -tricity to industrial output is the product of two

opposing trends: a steady increase in electricity-use in industries which are heavy users of elec-tric ity, and a stea dy decrease in the proportionof output from those same industries relative tototal U.S. industrial output (58). Looking ahead,there is considerable uncertainty about boththese trends.

Electricity use in industry is concentrated. Just13 specific industrial sectors* consume half of allindustrial electricity. These a re: primary a lumi-num, blast furnaces, industrial inorganic and or-ganic chemicals not elsewhere classified, petro-leum refining, p apermills, misc ellaneous p lastic

products, industrial gases, plastics materials andresins, paperboard mills, motor vehicle parts, al-kalis and chlorine, and hydraulic cement.

Future rates of output growth in several of theseindustries are highly uncertain. For example, theindustrial output of primary metals, which in-cludes both aluminum and steel, decreased byabout 1.0 percent per year between 1972-74 and

1978-80 (57). Capacity expansion plans for bothsteel and aluminum a re d own a bo ut two-thirdsfrom the ir level a d ec ad e a go (50).

Within the chemicals industry, electricity isused hea vily in the p rod uc tion of seve ral basicc hem ica ls suc h a s oxygen (where it is used forrefrigeration), and chlorine (where it is used forelectrochemical separation). Although the chemi-c als industry as a whole g rew by a n a verag e o f4.5 percent per year from 1974 to 1980, the

production of these basic electrically producedc hemica ls grew o nly 1 to 2 percent p er year (ox-ygen and chlorine) or actually decreased (acet-ylene and phosphorus) (1 1). Observers of the

chemical industry expect this trend to continueas demand for basic chemicals becomes saturated

and som e p roduc tion m oves oversea s. The U.S.

*As identified by four-digit Standard Industrial Classification (SIC)

codes.




chemical industry is expected to concentrate in-creasingly on small volume specialty chemicals,which use relatively little elec tric ity for p rod uc -tion.

Sinc e c hem ica ls, iron and stee l and p rimaryaluminum alone a c c ount for more than a thirdof e lect ricity use in industry, a 2 pe rc ent lag ingrowth of these electricity-using industries behindgeneral industrial output would alone cause a lag

of about 0.5 percent in electricity demand behindoverall growth in industrial output.

In addit ion to being concentrated by type ofindustry, elec tricity use in ind ustry is a lso c on-centrated by function. Uncertainty about the di-rection of trends in electricity use in electric-intensive industries comes about because elec-tricity use will probably become more efficient

in two of the functions (elec tric moto rs and elec-trolysis) and is likely to expand into significant

new uses in the third func tion (elec tricity to sup p -ly or substitute for process heat).

About half of all electricity use in industry is

used for electric motors, including compressors,fans and blowers, and pumps (3). Improvementsin the efficiency of electric motors are likely to be

c ontinuous for 10 to 15 yea rs through improve-ments in the motors themselves and through im-proved efficiency of use which takes advantageof new semiconductor and control technology.Thus, elec tric ity use p er unit of o utp ut fo r thesepurposes could decrease by 5 percent (if there

is little price stimulus) or up to 20 percent (if thereis significant p ric e stimu lus). Some of th is im-

proved effic ienc y should c ome ab out a s a resultof past price increases, as capital stock turns over.imp etus for the rest will dep end on future elec -tricity price increases, and the cost of installingthe c ontrol technolog ies.

Another 15 to 20 percent of all industrial elec-tricity is used for electrolysis of aluminum andchlorine (3,81 ). Aluminum electrolysis is morelikely to decrease than increase as a fraction ofindustrial use, because efficiency improvementsof 20 to 30 percent are technically possible fromseveral technologies and are probably necessary(given sharply increasing prices for electricity inthe No rthwe st, Texa s, and Lou isiana where p lantshave been located) to keep aluminum produc-

tion in the United States competitive with alumi-num p rod uc tion ove rsea s (81).

Electric process heating in industry accounts foronly about 10 percent of current uses of electrici-ty but has great potent ial to bec ome much m oreimportant as new electric process heating tech-niques are developed that ma ke b etter use of

electricity’s precision and ability to produce veryhigh temperatures. In some important high tem-perature industries such as cement, iron andstee l, and glassma king, e lec tric ity ma kes up 20to 35 percent of all energy use and could as muchas double its share.

Some techniques being d eveloped co uld havevery la rge impa c ts on e lec tricity dem a nd . Theseinclude plasma reduction and melting processesfor primary metals production, and inductionheating for shaping and forging. Other techniquessuch as lasers and robotics, however, are likelyto have small impacts on electricity demand be-cause only small amounts of electricity are usedfor ea c h a pp lica tion. The b igge st lasers, for ex-ample use only about 1 to 2 kW. Most use under100 W. Table 4 shows an assessment of the rela-tive impa c t likely from e ac h of the new er tech-niques (81).

Great uncertainty surrounds the contributionto industrial electricity demand from the most im-po rta nt new e lec trote c hnolog ies. The iron a ndstee l ind ustry could expe rienc e the g rea test in-crease in electricity demand. Production and

profit levels, however, in that industry are uncer-tain. Overall growth in the steel industry is pro-

jected to be only about 1.5 percent per year to2000 (81 ). The c on tribution o f elec tric a rc stee l-ma king from sc rap iron is expe c ted to inc rea sefrom ap proxima tely one-fourth to a t least o ne-

third of the total, but the potential for plasmareduction and melting is more uncertain partlybec ause there ma y be too litt le c ap ital to take ad -vantage of technological advances (81). If newtechnologies penetrate slowly, the impact on

electricity demand could be minimal until the late1990’s. It is conceivable, however, that rapid pen-

etration c ould oc c ur with a few ve ry suc c essfulnew techniques which in turn may increase theU.S. steel industry’s co mp etitive position andprospects for growth. In such circumstances,

25-450 0 - 84 - 4 : QL 3



38 . Nuclear Power in an Age of Uncertainty

Table 4.—Estimated Impact of Industrial Electrotechnologies in the Year 2000

Rough estimates ofGW of capacity Change Load Level of

1983 2000 to 2010’ factor uncertainty

3.4.

5.

6.

7.

8.9.

10.11.

12.“ 13.

14.

15.

16.17.

18.

19.20.

Arc furnace steelmaking. . . . . . . .Plasma metals reduction . . . . . . .Plasma chemicals production . . .High-temperature electrolysis

(aluminum and magnesium) . . . .Induction melting(casting) . . . . . . . . . . . . . . . . . . . . .Plasma melting . . . . . . . . . . . . . . .

Induction heating (forging) . . . . .Electro slag remelting. . . . . . . . . .Laser materials processing . . . . .Electron-beam heating . . . . . . . . .Resistance heatingand melting . . . . . . . . . . . . . . . . . .Heat pumps . . . . . . . . . . . . . . . . . .Electrochemical synthesis,electrolytic separation,chlorine/caustic soda . . . . . . . . . .Microwave andradiofrequency heating . . . . . . . .

Ultraviolet/electron-beamcoating . . . . . . . . . . . . . . . . . . . . . .Infrared drying and curing . . . . . .Electrically assistedmachining and forming. . . . . . . . .Low-temperature plasmaprocessing . . . . . . . . . . . . . . . . . . .Laser chemical processing . . . . .Robotics . . . . . . . . . . . . . . . . . . . . .

3.5-4.500

8-93-4

0.05

5-70.075

0.0005-0.0010.006-0.008

0.3-0.50.1

4-5

0.5

0.01-0.020.4-0.5

<0.001

<0.0010

<0.001

5-72-43-5

9-124-5

? b

8-100.15

0.001-0.0020.015-0.02

0.4-0.61-3

4-5

7

0.5-10.5-1

<0.001

<0.001<.5GW C

<0.001

+++

O or –o

+

o+++

O or –+

o

+

+o

—

— — —

HighHighHigh

HighLow andoff peakHigh

ModerateModerateModerateModerate

HighHigh

LowHighHigh

LowModerate

VeryhighModerateModerateModerateModerate

LowModerate

High Low

Moderate High

Moderate ModerateModerate Low

—

— — — — —

aExplanation. “ + “ = Increase to 2010; 4’0” = stable to 2010; “ – “ = decrease to 2010,bAny capacity in plasma melting directly competes with electric arc furnace steelmaking.cThis will probably only be used in U.S. Government-owned f’CilltleS for Uranium isotope Separation.

SOUR C E: Table prepared by Philip Schmidt, based on research for a repori publicized by EPRI in early 19S4, Electricity and Industrkl Productivity: A Technica/ and Economic Perspective (81).

there could be a substantial impact on electrici-

ty demand by the early 1990’s.

The role o f cog enera tion* in sa tisfying futureindustrial elec tric ity de ma nd is a fina l source ofuncertainty. (Cogeneration is discussed further inthe sec tion on utility strateg ies and wa s a sub - jec t of a rec ent O TA rep ort Industrial and Com- mercial Cogeneration (72).) A recent study by

DOE estimated that cost-effective opportunitiesexisted in industry for about 20 GW of self-con-tained cogeneration without sales to the outsideelectrical grid, a potential increase of slightly morethan the c urrent installed c ap ac ity of ab out 14

GW with about 3 GW additional capacity in theplanning stage (72). if fully realized, an increase

*The combined production of electricity and useful steam (or hotwater) in one process.

in cogeneration of this magnitude would reduce

the growth rate of purchased electricity by about0.5 percent per year. Cogeneration is beingadapted more slowly than the market potentialindicated earlier partly because the success of in-dustrial energy conservation programs has re-duc ed industrial req uirem ents for stea m.

For the 1980’s, the highest growth rate in indus-trial electricity demand is likely to be no more

than equal to industrial output, and growth ratesc ould fa ll as low as 2 pe rc enta ge po ints be lowindustrial output growth. For the 1990’s, it is quitepossible that electricity demand could grow faster

than industrial output (20). The reasons for possi-ble faster growth in industrial electricity demandin the 1990’ s tha n in the 1980’s should be c lea rfrom the abo ve d isc ussion. The outp ut o f suc helectricity-using industries as steel and aluminum




is likely to continue to grow very slowly during

the 1980’s, but may grow more rapidly in the1990’s. Tec hno log ies for imp roving the e fficien-

c y of use o f elec tric mo tors a re a lso likely to ha vethe greatest impact in the 1980’s, while new elec-trotechnologies for substituting electricity forprocess heat are not l ikely to have a big impact

in the 1980’s because they are now a small shareof total electricity use. Even if they grow rapidly,they cannot have a major impact on overall in-dustrial electricity demand until they are a largerfraction of the whole, sometime in the 1990’s.

Residential Demand for Electricity .-From abottom-up perspective, what happens to futureelectricity demand from households dependsboth on how fast the total number of householdsincreases and on what happens to electricity useper household.

There is uncertainty, first of all, about the rateof household formation. Over the decade from1970 to 1980, the U.S. population formed house-holds at a rate much faster than populationgrowth. In c urrent c ensus p rojec tions, this trendis expe c ted to c ontinue throug h the 1980’ s, re-sulting in a fairly rapid rate of household forma-tion of 2.2 percent per year and a further dropin household size from about 3.2 people perhousehold in 1970 to 2.8 people in 1980 to 2.5

people per household in 1990. On the otherhand, were the U.S. taste for living in smaller andsmaller households to become less important, thegrowth rate in household formation could fall to1 percent per year or less.

The p ote ntial for increased use of e lec tric ity perhousehold largely comes about because the num-ber of households with air-conditioning and elec-tric he a ting is still inc rea sing. This is sho wn infigure 6. As of 1979 only 16 percent of all house-holds had electric he at, but a bo ut half of new

dwe lling units we re he a ted elec tric a lly. As newdwelling units replace existing ones, the percentof total dwelling units with electric heat coulddouble or triple. A doubling of the share of allhouseholds heated electrically (assuming no in-

crease in the efficiency o f spa ce heat) wo uld ad dabout 0.7 percent per year to household growthin elec tric ity de ma nd . The pote ntial for increaseduse of e lec tric spac e hea ting, howe ver, will beinfluenc ed by the relative c ost of e lec tric hea t

Figure 6.—Penetration of Air-Conditioning and ElectricHeating in Residential Sector

1940 1950 1960 1970 1980

r Percentage of new Percentage of newhouseholds with air households withconditioning central air

conditioning

Percentage of new

IPercentage of new

households with households withelectricity for heat pumpsprimary heating

P e r c e n t a g e o f t o t a l - - - Percentage of totalhouseholds with households with

air conditioning electricity forprimary heating

SOURCE: Department of Energy, The future of Electric Power in America.Economic Supply for Economic Growth, Report of the ElectricityPolicy Project, June 1983, This graph was prepared from data in theEIA residential energy consumption surveys

which will in turn be reduced by increases inelectric heating efficiency. About 70 percent ofnew households have air-conditioning comparedto about 55 percent of e xisting ho useho lds, so

modest increases in electricity use from air-conditioning are also likely.

The use o f electric ity to heat wa ter may e xpand

beyond the 30 percent of households that nowuse it and c ould a s muc h as doub le if there is abig decrease in the relative cost of electric andga s-heated hot wa ter. The de ma nd for other elec-tric appliances is considered largely saturated and



Ch. 3—The Uncertain Financial and Economic Future • 41

t inue, and commercial square footage continuesto g row mo re slow ly than G NP, com me rc ial elec-tricity use will only increase as fast or faster thanGNP as long as electricity use per square footc ontinues to inc rea se.

Elect ric ity use p er squa re foot in c om merc ial

buildings may continue to increase for severalreasons. Only 24 percent of existing commercialbuilding square footage but almost half (48 per-c ent) of the ne w building squa re foo tag e is elec-trically heated (33). If these trends continue theshare of buildings that are electrically heated

could double.

Air-conditioning in commercial buildings isprobably saturated. About 80 percent of all build-ings have some air-conditioning. Small increasesin electricity use per square foot can come aboutby air-conditioning more of the building and byreplacing window air-conditioners with central

or package air-conditioners. Window air-condi-tioners c oo l 20 pe rc ent of the e xisting b uild ingstock, but only 9 percent of the newest buildings(33).

Greater use of office machines and automationmight increase electricity use both to power thema c hines and to c ool them in offic e b uildings,stores, hospitals, and schools. Machines, how-ever, are less likely in churches, hotels, and othercategories of commercial buildings.

The p ote ntial for inc reased effic ienc y of elec-tricity use in commercial buildings is less wellknown than for residential buildings becausecommercial buildings are very diverse and thepotential for increased efficiency depends part-ly on success in balancing and integrating thevarious energy loads: lighting, cooling, heating,refrige rat ion, a nd ma c hines. OTA a na lyzed the

theoretical potential for reductions in electricityand fuel use in commercial buildings in a recentrepo rt, Energy Efficienc y of Buildings in Cities (71),

and found that electricity use for lighting and air-conditioning in commercial buildings can be re-duced by a third to a half. Heating requirements

also c an be red uc ed substantially b y recyc lingheat generated by lighting, people, and officemachines from the building core to the periphery.There is still very little verified doc ume nta tion o f

energy savings in commercial buildings, however,

and therefore, considerable uncertainty remainsabout the potent ial .

The results of the bot tom-up ana lysis for theco mmercial sec tor indicate a dem and for elec tric-ity that will increase for a few years at about therate of increase of GNP but with wide ranges of

uncertainty. It could be higher as a result of fasterpenetration of electric space heating and moreair-cond itioning a nd big load s from offic e m a-c hines, or low er as a result of b ig improvementsin the efficiency of commercial building electric-ity use. By the 1990’s, however, when the de-ma nd fo r electric hea t in c omm ercial buildings

will be largely saturated, the trend rate of growthin electricity demand is likely to settle to thegrowth in commercial square footage, somewhatless tha n the grow th rate of G NP.

Conclusion

Utility executives contemplating the construc-tion of long Ieadtime powerplants must contendwith considerable uncertainty about the probablefuture g row th rates in elec tric ity dem and , The

range of possible growth rates encompasses lowave rag e g row th rate s of 1 or 2 pe rc ent p er year,which would justify very few new large power-p lants, up to fa irly high g row th rates of 3.5 to 4percent per year, for which the pressure to buildseveral hundred gigawatts of new large power-plants is great, as will be clear from the discus-sion in the next section.

The sources of unc erta inty are m any. Future

trends in electricity prices are viewed differentlyby different forecasters, because there is disagree-ment about future capital costs of generating ca-pacity, future rates of return to capital and futureprices of c oa l and na tural ga s for elec tric ity. There

also is uncertainty about how consumers and in-dustry will respond to higher prices, given manytec hnic al op po rtunities for imp roved effic ienc yin appliance use and industrial electricity use and

increasing numbers of promising new electro-technologies that could substitute for the use of

oil and natural gas for industrial process heat. Sev-eral of the industries, such as iron and steel, how-ever, where the new electrotechnologies couldhave the greatest impact, face an uncertain fu-ture.




At present, utility strategy for supplying ade- ficulties and regulatory disincentives for large cap-quate electricity is influenced heavily both by the ital projec ts. The influenc e o f utility ra te reg ula-rec og nition o f unce rtainty abo ut future g rowth tion and current utility strategies are discussedin electricity demand and by recent financial dif- Iater in the chapter.

RESERVE MARGINS, RETIREMENTS, AND THENEED FOR NEW PLANTS

Powe rplants are planned and ordered ma ny

years in ad vanc e of whe n they are need ed . inorder to p rod uce po we r for sale b y 2000, nuclearpowerplants on a typical schedule would haveto be ordered by 1988, or 1990 at the latest, andcoal plants or nuclear plants on an acceleratedschedule, must be ordered by 1992 or 1993.Sources of generating capacity with shorter lead-times, such as gas turbines or coal conversion,need not b e o rde red be fore 1996 or 1997.

There is c onsid erab le disa greem ent a bo ut howmany new powerplants will be needed by 2000.Those w ho b elieve tha t large num bers of newpowerplants will be needed (several hundredGWs) anticipa te rap id growth o f electric ity de -mand (3 or 4 percent per year), expect that largenumbers of existing plants will be replacedbecause of deterioration in performance or retire-ment due to age and economic obsolescence,and expect only modest contributions from smallpower production (20,83).

On the other hand, some believe that no new

pow erplants or very few (a few d ozen G W) willbe needed before 2000 because they anticipateonly slowly growing electricity demand (1 or 2percent per year), expect little or no need toreplace existing generating capacity, and expect

substantial contributions to generating capacityfrom small sources of power such as cogenera-tion, geothermal, and small-scale hydropower(83).

This sec tion lays out the rang e of possibilitiesfrom no new powerplants to several hundred

gigawatts of new powerplants arising out of dif-

ferent combinations of growth in electric demandand varying utility decisions about powerplantretirements and use of small power sources.

The e lec tric utility ind ustry current ly projec tsaverage growth in peak summer demand of 3.0percent per year between 1982 and 1991 (68). *The industry has a lso p lanne d for an inc rease inelectric-generating resources of 158 GW by 1991bringing the total generating capability up to 740GW (68). Only 13 GW of sc hed uled retirem entshave been included in the 1991 estimate (69). Atthe same time the current reserve margin** of33 percent is forecast to fall to 20 percent. As a

rule, utilities like to maintain a reserve marginof 20 percent to allow for scheduled maintenanceand repair and unscheduled outages. Individualregions may require higher reserve margins if theyare poorly connected to other regions, if they aredependent on a small number of very large plantsor if they are dependent for a large share of gen-erating c ap ac ity on o lde r plants or plants thatburn expe nsive oil and na tural ga s.

As shown in figure 7, the planned resources of740 GW scheduled for 1991 would allow a re-serve ma rgin of 20 pe rc ent to be ma intained un-

til 1996 if electricity peak demand grows only at2 percent per year, or until 2000 if electricity de-mand grows only at 1 percent per year. On theother hand, the average reserve margin will fallbelow 20 percent by 1987 if electricity demandincreases at 4 percent per year.

However, the number of new powerplants thatmust be built to maintain a given reserve margindoes not depend only on the rate of increase in

*This had dropped to 2.9 percent per year for 1983 to 1992 inthe 1983 North American Electric Reliability Council Forecast ofElectric Power Supply and Demand (70).

**’’ Reserve margin” is defined as the percent excess of “plannedresources” over “peak demand” where “planned resources” in-cludes: 1 ) installed generating capacity, plus 2) scheduled powerpurchases less sales.




Figure 7.—Projected Generating Capacity and Alternative Projections of Peak Demand

Load growth Load growth Load growth Load growth Load growth

of 1%/yr of 2%/yr of 3%/yr of 3.5%/yr of 4%/yrq . . . . . . . . . - - - - - - — - - — — .

1,400

1,200

1,000

800

600

400

200

0

1980 1985 1990 1995 2000 2005 2010

Year

NOTE. “Planned resources” IS defined by NERC to include: 1) Installed generating capacity, existing, under construction or in various stages of planning; 2) plusscheduled capacity purchases less capacity sales; 3) less total generating capacity out of service in deactivated shutdown status.“Reserve margin” given here IS the percent excess of “planned resources” over “projected peak demand. ”

SOURCE” North American Electric Reliability Council, Electric Power Supply and Demand 1982-1997, August 1982 and Off Ice of Technology Assessment

electric pe ak de ma nd. It also d ep end s on howmuc h existing g enerating c ap ac ity must be re-placed because powerplants are retired, due toag e o r to ec onomic obsolesce nce, or bec ause ex-isting powerplants must be derated to lower elec-tricity outputs.

Retirements Due to Age.–The “book lifetime”

of a powerplant, used for accounting purposes,is usually 30 to 40 years. Over this period theplant is gradually depreciated and reduced as arecorded asset on the utility’s books until, at theend of the period, it has no more book value and

produc es no return on c ap ital. How ever, in p rac -tice powerplants may continue to operate for 50yea rs or more. As of 1982 there w ere a bout 10GW of generat ing c ap ac ity that we re mo re than

40 years old, more than a qua rter of the to talgenerating capacity that was in service 40 yearsa g o .

In fact, the bulk of the current generating ca-pacity of the United States is comparatively new.Over half has been built since 1970, as shownin figure 8. The numb er of p lants tha t w ould beretired by 2000 varies greatly with the assumedplant life. In the unlikely event that a 30-year lifewould be used, over 200 GW would be retired

by 2000 (see table 6). A 50-year schedule wouldretire only 20 GW.

Ec ono mic Ob solesc enc e. -From 1965 to 1979a large number of steam-generating plants usingoil or natural gas were built (see fig. 8). They were




Figure 8.–The Energy Source and Age of ExistingElectricity Generating Capacity

Energy source

Other

Nuclear

Water

Gas

Coal

Oil

Before 1950- 1955- 1960- 1965- 1970- 1975- 1980-1950 1954 1959 1964 1969 1974 1979 1982

SOURCE: Energy Information Administration. Inventory of Power Plants in theUfnited States, 1981 Data from the ‘Generating Units Reference File:

Table 6.—Possible Needs for New Electric.Generating Capability to Replace Retired

Powerplants, Loss of Powerplant Availability, andOil and Gas Steam Powerplants

Cumulative replacementcapacity

(GW) needed by:

1995 2000 2005 2010

If existing powerplantsare retired after:30 years. . . . . . . . . . . . . . . . 155 230 395 51040 years. . . . . . . . . . . . . . . . 55 105 155 23050 years. . . . . . . . . . . . . . . . — 20 55 105

If all 011 and gas steamcapability is retired asfollows:All . . . . . . . . . . . . . . . . . . . . 152 152 152 152Half . . . . . . . . . . . . . . . . . . . 76 76 76 76All oil and gas capacityabove 20 percent of region(3 regions) . . . . . . . . . . . . . . 55 55 55 55

If average coal andnuclear availabilityslips from 700/0 to:About 65% . . . . . . . . . . . . . 21 21 21 21About 600/0 . . . . . . . . . . . . . 42 42 42 42

SOURCE: Office of Technology Assessment.

ordered before the 1973 oil embargo. Under cur-rent price forecasts, oil prices are expected to re-main fairly stable until the late 1980’s or early

1990’s and then increase substantially (by about60 perc ent) from 1990 to 2000 (27). The p ric e o f

natural gas to utilities is expected to increase

steadily through the 1980’s as the long-term con-tracts for natural gas sold at relatively low pricesexpire and are replaced by contracts for more ex-pensive gas.

As of 1981, there were 152 GW of oil and natu-ral g as stea m-gene rating c ap ac ity. Tog ether they

totaled 27 percent of all generating capacity butproduc ed only 22 perce nt of a ll elec tric ity. Asshown in table 7, oil-fired steam plants producedonly ha lf a s muc h e lectricity relative to their sha reof generating capacity. Natural gas-fired steampIants, on the other hand , produc ed a greate rshare.

Even though oil and natural gas will be expen-sive, plants burning these fue ls c an be used aspart of the reserve margin. Oil and gas are, in fact,

just about 20 percent of two regions, the South-east (SERC)* and the West (WSCC). The fraction

of oil- and gas-generating capacity, however, ismuch larger than 20 percent in three regions:Texas (ERCO T) about 72 p erc en t, Southwest

Pow er Poo l (SPP) ab out 56 percent, and theNorthea st Power Co ordinating Counc il (NPCC)

about 51 percent (68). If oil and gas steam plantswere retired continuously in these regions untilthey formed no more than 20 percent of total

generating capacity, the total retired would beabout 55 GW.

Loss of Availab ility of Generating C apac ity.– The p erce nt of time tha t nuc lear and fossil b ase-

Ioad plants were available to generate electrici-ty averaged around 70 percent** over the decade

of the 1970’s (67). If there were a reduction from70 to 65 perce nt in the a verage ava ilab ility ofnuclear and coal powerplants this would be theeq uivalent of a loss of 21 GW out o f a to ta l c ur-rent coal and nuclear-generating capacity of 294

GW (see table 6).

A recent study for DOE assesses the prospectsfor changes in average availability (82). Statistical

*These are the regions of the Northeast Electric Reliability Council

(NERC).

**The availability figure used here is equivalent availability andincludes service hours plus reserve hours less equivalent hours ofpartial outages. (66) From 1971-80, nuclear pIants and coal plantsover 575 MW averaged 67.8 percent in equivalent availability andcoal plants from 200-574 MW averaged 74.3 percent in equivalentavailability.



Ch. 3—The Uncertain Financial and Economic Future 45

Table 7.—installed Capacity and Net Electricity Generation by Type of GeneratingCapacity, 1981-82

NetInstalled electrical

generating energycapabiIity, Percent generation, Percent

summer of 1981-82 of1981 (GW) total (billion kWh) total

Steam — coal . . . . . . . . . . . . . . . . . 243 42 1,177 52

Steam — oil . . . . . . . . . . . . . . . . . . 89 16 190 8Steam — gas . . . . . . . . . . . . . . . . . 63 11 320 14Nuclear power . . . . . . . . . . . . . . . 51 9 274 12Hydro electric . . . . . . . . . . . . . . . . . 66 12 261 12Combustion turbine — oil . . . . . . . 34 6 3 —

Combustion turbine — gas. . . . . . 6 1 7 —

Combined cycle oil . . . . . . . . . . . . 3 — 2 .Other . . . . . . . . . . . . . . . . . . . . . . . . 17 3 26 1

Total . . . . . . . . . . . . . . . . . . . . . . 572 100 2,260 100

SOURCE North American Electric Reliability Council, Electric Power Supply and Demand 1982-1991, August 1982.

evidence from the past two decades would sup-po rt a n estima te o f a loss of 3 to 5 pe rc enta gepoints of average availability for every 5-year in-

crease in average age of coal plants. Lookingahead, there could also be losses in availability

of several percenta ge p ointsdue to em ission con-trols and requirements for low sulfur coal that is

at the sam e time o f lowe r com bustion q uality.

Offsetting these tende nc ies to red uce d ava ila-b ility, howe ver, there a re a lso fo rc es tha t mightincrease average availability. The utility industryhas completed a period of construction of coalplants with poor availability, and the newestplants (from the late 1970’s and early 1980’s)should have substantially higher average availa-bility. If this were to continue, overall availabili-ty could increase. If utilities invest in higher avail-abilities (e. g., by converting forced draft boilersto balanced draft), this will also increase availabili-

ty (82). It is c lear that atte ntion to fue l qua lity a ndgood management also can raise availabil it ies.Som e Public Servic e C omm issions (e.g., Mic hi-gan) are including incentives to improve availa-bilities in utilities’ rate of return formulas.

On balance, it is unlikely that availability willincrease or decrease dramatically. If a change inavailability should occur, however, it would havea noticeable impact on the need for new capaci-

ty. A 10-percentage-point change could imply anincreased (or reduced) need for powerplants of

more than 40 GW by 2000.

Summary-The Need for New Powerplants.–the need for new powerplants depends on boththe growth rates in electricity demand and on the

need for replacement of existing generating ca-pa c ity. Tab le 8 summ a rizes mo st o f the ran ge ofdisagreement and its implications for new power-plants. Estimates of growth in electricity demandrange from 1 to 4 perce nt per year. (The ta b le

shows the implications of electricity demandgrowth rates of 1.5 to 3.5 percent.)

judgments about replacement of existing plantscan, somewhat arbitrarily, be divided into high,medium, and low replacement. A high-level re-placement of about 200 GW by 2000 would benecessary to: offset a slippage of about 5 percent-age points in availability, meet a schedule of

40-year life expectancy for all powerplants, andretire about half the oil and gas capacity in thisc ountry (see tab le 6). A low -level rep lac eme nt of50 GW would meet a 50-year schedule, retire alitt le oil and gas capacity and would assume noslippage or an actual increase in average availa-bility.

If these a lterna tive rep lac em ent a ssump tions

are c omb ined with alternative g rowth rate as-sump tions (tab le 8), they lead to a wide rang e o fneed s for new plants. About 454 GW of new c a-pa c ity would b e need ed , for examp le, by 2000

(beyond NERC’s planned resources for 1991) tomeet a 3.5 percent per year increase in peak de-mand for electricity and the high replacement re-




Table 8.—Numbers of 1,000 MW PowerplantsNeeded in the Year 2000

(beyond current utility plans for 1991)

Levels of replacement Electricity demand growth

of existing plants 1.5%/yr 2.5%/yr 3.5%/yr

Low: 50 GW; Replace allplants over 50 years old . . 9 144 303

Moderate: 125 GW; Replace

all plants over 40 yearsold; plus 20 GW of oil andgas capacity . . . . . . . . . . 84 219 379

High: 200 GW; Replaceplants over 40 years plus95 GW of oil and gascapacity . . . . . . . . . . . . . . 159 294 454

NOTES: 1. Planned generating capacity for 1991 is 740 GW, 158 GW more than1982 generating sources of 582 GW. Starting point for demand calcula-tions is 1982 summer peak demand of 428 GW.

2. The calculations assume a 20- percent reserve margin, excess of plann-ed generating resources over peak demand.

qu irem ents, as is pointed out in rec ent wo rk at

the Electric Power Research Institute (83). On theother hand, a low replacement requirement com-bined w ith only 1.S pe rc ent de ma nd g row th peryear would req uire a lmost no new c ap ac ity.

This the n is the d ilem ma for utility stra teg ists.A shift of o nly 2 pe rc enta ge po ints in d ema nd

growth combined with a more stretched outretirement schedule can reduce the requirementfor new powerplants from hundreds of gigawatts

(a number requiring a capital outlay of $0.5trillion to $1 trillion 1982 dollars) to almostnothing. Som e o f the fa c tors a ffec ting utilitychoice of strategy, given this situation, are dis-cussed in the next sections.

SOURCE. Office of Technology Assessment.

RATE REGULATION AND POWERPLANT FINANCEAlthough the national average reserve margin

will not dip below safe limits until well into thenext decade, individual utilities may considerordering powerplants before 1990 for any of threerea sons: to replac e e xpensive oil or natural ga sgenerating capacity, to anticipate growth in their

reg ion, or to start a long Iea dtime p lant w ell be -fore it ma y be ne ed ed in the 1990’s. This sec tiondescribes the framework of rate regulation with-in w hich suc h a dec ision is ma de . The next sec -tion explores the broader strategic options for

utilities.

Utilities’ Current Financial Situation

Although the financial situation of utilities is im-proving slowly, they are still in a greatly weak-ened financial condition compared to their situa-tion in 1970. The series of figures 9 throug h 14prepared for the Department of Energy shows theorigins of the financial difficulties of the 1970’s

until the relative improvement of 1982 (describedbelow).

Utilities raised enormous amounts of capital inexternal financing in the 8 years from 1973-81,more than double the requirements of the tele-phone industry-the ne xt most c ap ita l-intensiveindustry (fig. 9). In the process, more and more

of utility assets bec am e tied up in co nstructionof new generating capacity (f ig. 10), which equaleda quarter of all utility assets as of 1981. At thesame time, eve n with high rate s of infla tion thenom inal return on e q uity w as kep t c onsta nt. Thusthe real value of utilities’ return on equity de-

clined sharply (fig : 11).

As a consequence of high inflation, enormousamounts of investment and large fractions ofassets under construction, there was weakeningof many indicators of financial health that are

wa tc hed c losely by investors. The a mo unt o fearnings paid out as dividends increased, leav-ing less for reta ined ea rnings to financ e futureprojec ts. The ratio o f op erating inc om e to interest

on debt—pretax interest coverage—fell to disturb-ingly low leve ls. The c ost o f c ap ital from issuesof new stoc k and bo nd sales rose a c c ordingly.After 1973, far more utility bonds were down-graded than upgraded; the number of utilitybonds rated only “medium grade” BBB, in-creased from 10 to 43 (fig. 12). A lower ratingusually means that investors require a higher yieldin order to purchase the bond, and institutional

investors may not be willing to purchase the bondat all (45). Similarly, the average market value ofutility stock fell steadily from its high of about 2½times book value in 1965, to a level equal to book




value in 1970 to less than book value in 1974(fig. 13). When stock sells below book value,

there a re two c onseq uenc es. The utility mustissue more stock, (and pay out more dividends)to raise each new unit of capital than the valueof each existing unit of capital, and the value ofstock for the existing stockholders is diluted. (See

box A for a discussion of market-to-book ratio andstock dilution. )

One of the most serious problems for utilitieshas been a steady squeeze on cash flow. Asshown in fig. 14, almost 50 percent of utilities’nom ina l return on e quity wa s pa per ea rnings inthe form o f allow anc e for funds used during c on-

struction (AFUDC). (See box B.) One result is thatutilities have retained less and less of their earn-ings. The sha re of ea rnings pa id to stoc kholdersas dividends—the dividend payout ratio—increased from 68 percent in 1970 to 77 percentin 1981, For some companies it was above 100

percent, which means they paid out more than

they earned (45).

In 1982 and 1983, there was some improve-ment in utility financ ia l hea lth. As of Dec em ber1983, market-to-book ratios were up to an aver-age 0.98, and 51 out of 103 utilities had market-to-book ratios of more than 1.0 (59). Althoughmore bonds are being downgraded than up-




The History of the Deterioration in the Financial Health of Electric Utilities, 1960=82

Figure 9.—The Electric Utility Industry is Dependent on Externally Generated Fundsto a Greater Extent Than Other Large Capital Users

173Total External Financlng,a 1973-81

79

3635

15 141

90

FExternal Financing’ as a Proportion of Total Capital

80 Expenditures, 1973-81

aNet of debt retirements.blncludes e~Ploratlon and prduction by companies Prlfnadb lnvolv~ ‘n ‘efining.

Figure 10.–Since 1967, the Fraction of Utility Assets Tied Up in"Construction Work in Progress" Has Steadily increased

Other assets

225

150

75

(6%)

.

Other assets

Other assets(8%)

1967 1973 1981

fconstruction work In progress (cwlP) refers to plant IXIlntl built, but not Yet In sewice.

Figure 11.—Beginning in 1972, the Utilities' Real Return on Equity Has Been Eroded by Inflation

14 %

12%

10%

8%

6%

40%

2 %

o

–2%

(utility Industry annual average return on common equity (end of year))

Nominal Roe

1 1 1 i i i i 1 i 1 1 1 i1967 1969 1971 1973 1975 1977 1979 1961

eNomlna\ Roe deflated by GNP implicit price deflator (measured at fourth quarter SnnuallY).




Figure 12.—Since 1973, the Average Utility Stock has Sold Below its Book Value a

2.21 I

62 64 66 68

1.39 1.41 1.29

I I70 72

1.090.91

74 76 78 60 62

cAverage of high and low value for the Year..‘August 1962.

Figure 13.—ln 1970, A Majority of Utilities Were Rated AA or Better in 1982, a Majority Are Rated A or Worse

20

AAA

69Key

I I I1970 1962

50

4341

!

26

10

I3

0 1 0 1

AA A BBB BB B

%e box A for an explanation of the implications of stock selling below book value.

Figure 14.-Over Time, the Share of AFUDC in Total Earnings Has Increased and the Share of Cash Has FallenUtility Industry Annual Average Return on Equity (end of year )

14 r Nominal Roe.

1967 1969 1971 1973 1975 1977 1979 1961

SOURCE: Booz-Allen & Hamilton Inc., The Financial Health of the Electrlc Utility Industry, prepared for the Department of Energy October 1962, using data from UtilityCompustat; U.S. Department of Commerce, Survey of Current Business; Edison Electric Institute, Statistical Yearbook of the Utility Industry; Standard andPoor’s Bond Guide; Energy Information Administration, Statistics of Privately Owned Utilities.




graded , the number of net d owngradings has

been reduced sharply. By late 1983, the averageearned return on common equity for the 100largest electric utilities was up to 14.1 percent,more than 200 basis points higher than theearned return in 1980 (59).

Implications of Utilities’Financial Situation

There a re tw o w ays of looking a t the c urrentfinanc ia l situation o f the u tilities and the incen-tives to build more plants. From one perspectivethe electric utilities have shared in general eco-

nomic problems and have not fared as badly assom e ind ustries. The ma rket-to -bo ok rat ios of a llindustrial stocks fell over the 1970’s, although onaverage the industrial stocks stayed well above

a market-to-book ratio of 1.0. For the first part

of 1982, the return on investment of the electricutilities ranked 14th out of 39 industries, wellabove the average for such industries as chemi-cals, appliances and paper (41).

From this point of view, there is no need forfurther concern about near-term utility solven-

c y. The w orst of the utilities’ p rob lems a re com -ing to an end, and their financial situation shouldimprove gradually. Public utility commissions(PUCs) have responded by increasing the allowedrate of return to utilities, and the Fed eral Go vern-ment has provided additional relief from cashflow shortag es in the 1981 Ec ono mic Rec ove ry

Tax Ac t through liberalized dep rec iation a llow-anc es. The ta x law further manda tes tha t thesemust be “normalized” (retained by the utility)rather than “flowed through” to the consumer



Ch. 3—The Uncertain Financial and Economic Future Ž 51

in lowe r elec tric ity rate s. Further relief for tho seutilities, with problems selling their stock tofinance large construction programs, has also

be en a va ilab le from the 1981 Ta x Act b ene fits forreinvestme nt o f d ividend s in p urcha ses of ne wutility stock. Over the next few years, externalfinancing needs should diminish gradually, as

total construction expenditures decrease slight-ly and internal funding–from reta ined ea rnings,depreciation and deferred Federal taxes–in-creases.

From another perspective, however, there is

still cause for concern. Inflation has broughtabout several distortions in the ratemaking proc-ess that ma y need to b e c orrec ted before the nextround of orders for new generating capacity.From this point o f view , the issue is no t w hethe r

utilities need rate relief to keep from going bank-rupt, but w hether the trea tment o f ca pital in ratereg ulation need s to b e a djusted for the impa c tof inflation in order to prevent allocation of utilityresources away from capital-intensive electrici-ty-generating processes—beyond the point whereit would be beneficial to the ratepa yer.

It is not enough, from this perspective, for utili-ties to recover their financial strength during thecoming decade of slowed construction programs.

In the 1990’s, when the utilities need capital againfor construction, investment advisors will havemod els that predict a de terioration in financ ialhealth associated with a large construction pro-gram unless rate regulation can be expected to

give an adequate return on capital right throughthe c onstruction p eriod .

The c onc ern that elec tric ity ra te regulationneeds to be adjusted for the impact of inflationc an b e summ ed up in seve ral po ints. The firstproblem is erosion in the definition of cost ofc ap ital. Bac k in 1962, elec tric ity d ema nd wa sgrowing rap idly. There wa s rela tively little infla-tion, and unit costs of generating electricity weredecreasing. Utilities were allowed an average11.1 percent return on equity and actually earnedslightly more than that—about 11.3 percent—

right in line with the average of Standard & Poor’s400 industrial stoc ks (45).

By 1973, this situation had changed. Demandfor electricity was growing more slowly, inflation

had increased greatly, and the unit co sts of p ro-duc ing electric ity ha d be gun to inc rea se. From1973 to 1981, the return on equity earned by

utilities (10.9 percent) fell substantially below thereturn on equity allowed by State PUCs (1 3.3 per-c ent). Althoug h the a llowe d return c am e c loseto the return on equity earned by the 400 indus-

trials (14.3 percent), the earned return fell wellbelow. If the return on eq uity is designed to e x-clude AFUDC-deferred paper earnings, it fell farbelow the return on industrials (see fig. 14).

For utilities to earn a return on equity signifi-

cantly below that earned by industrial stocks rep-resents a change from past regulatory practice.The basis for determining ra te of return has itslegal foundation in two cases–the 1923 Bluefield Water Works* c ase a nd the 1944 Hop e Natural Gas Co. case.** These c ases estab lished threeprinciples:

1. the utility can charge rates sufficiently highto maintain its financial integrity;

2. the utility’s ra tes may c ove r all leg itimate ex-pe nses inc luding the c ost o f c ap ital; and

3. the utility should be able to earn return ata rate that is comparable to companies ofc om parab le risk.

Although in principle PUCs allowed rates ofreturn on equity comparable to companies of

comparable risk, in practice they failed to adaptrate regulation practices to accommodate infla-tion. The prac tice of using a n historic a l test yea rrathe r than a future yea r to d etermine incom eand expenses is one example. Politically, it oftenwa s d iffic ult for PUCs to grant fu ll rate increa ses

requested by utilities when inflation caused themto return yea r after yea r. As prec ed ents ac c umu-lated in ea c h State it bec am e ha rde r for individua l

Commissioners to argue for a restoration to thefu l l Hope/B luef ie ld de f in i t ion o f cos t o fc ap ital. * * *

q Blue field Water Works and Improvement Co. v. West Virginia

Public Service Commission, 262 ‘US 679, 692 PUR 1923D il .**Federal Power Commission v. Hope Natural Gas Co. (1944),

320 US 591, 60351 PUR NS 193.***It is in this context that one Wall Street partici ant in an OTA

rworkshop recommended a high-level commission o rate regulators,utility executives, and investment advisors to reexamine theHope/Bluefield principles, determine the problems of implement-ing the principles in times of inflation, and make recommendationsthat take into account the political realities of a Federal system.




A second problem, exacerbating the first, is thatof “ ra te shoc k” w hich a rises from the front-endloading of rate requirements for large capital in-vestme nts. This pheno me non (explained in bo xC) is noticeable at low rates of inflation and isstriking at high rates of inflation. Assuming 7 per-cent inflation, for example, the cost of electrici-

ty in constant dollars from Nuclear Plant X shownin box C would be 9.5¢/kWh the first year andonly 1.5¢/kWh the 20th year. For large plantsentering the rate base of small utilities, the in-

c rease c an b e 20 to 30 pe rc ent the first yea r. Thisprob lem is exac erba ted the m ore AFUDC (seebox B) is included in the cost of the plant as it

enters the rate base. (AFUDC is described furtherin the next section on the risks of constructing

nuclear plants because the impact of AFUDC islargest for plants with the high capital cost andlong duration of nuclear plants.)

The c om bina tion o f the very high rate req uire-me nts in ea rly yea rs and low ra te req uirem entsin subsequent years discourages multidecadeplanning of generating capacity. While short-term

rate increases must be tolerated to realize long-term reductions in real electricity rates, there areintense political pressures on public utility com-missioners to hold down these short-term rates

(88).

A final cause for distortion in utility rate regula-tion is the generally practiced fuel pass-throughwhich allows utilities to pass changes in fuelpric es onto c onsumers without go ing ba c k to the

PUC for a n inc rease in rates. This ha s bee n a use-ful device for avoiding damage to utility cash flowfrom the volatile changes in fuel (oil and naturalgas) prices in the 1970’s but it has had the in-advertent effect of shielding utilities from the ef-fects of inflation in fuel costs while they have notbeen shielded from increases in the cost of capi-ta l. For a utility fac ed with cap ital expend ituresto avoid fuel costs—through rehabilitating a plant,building a new one, or investing in load manage-ment—there is a theoretical incentive to stick with

the fuel-burning p lant a s long as fuel c osts a rerecovered immediately and capital costs are re-

c overed late a nd not fully.

Many utilities continue to base their generatingcapacity decisions on what will minimize lifetimec osts to rate payers. How ever, som e u tilities are

beginning to say openly (seethe later discussionof utility strateg ies) tha t they a re a ttem p ting tominimize c apita l req uirem ents ra ther than tota lrevenue requirements, to protect the interests oftheir stockholders to the possible long-term detri-ment of the ratepayers.

Possible Changes in Rate RegulationThere ha ve b ee n ma ny spe c ific prop osa ls for

utility ra te reg ulation. Som e a re d esigned to e n-courage conservation, load management, or therehabilitation of existing powerplants. Others aredesigned to encourage the construction of newpowerplants, especially when they are intendedto d isp lac e pow erplants now b urning oil or nat-ural gas.

This assessme nt d oe s not d ea l with the c om -plex subject of rate regulation reform in any de-tail. However, it is useful to describe briefly some

of those reform proposals that are specifically in-tended to offset those aspects of rate regulationthat discourage capital-intensive or risky projects.These refo rm proposa ls wo uld b e m ost likely toimprove the prospects for more orders of nucle-ar powerplants.

Construction Work in Progress (CWIP).–Thesimplest of the proposed changes in rate regula-tion is to allow a large fraction or all of a utility’sCWIP in the rate base. CWIP is advocated be-c ause it red uc es or elimina tes AFUDC. This inturn increases utility cash flow and the quality ofearnings and reduces the likelihood of rate shockbecause the rate base of the new plant includeslittle or no AFUDC.

One a rgum ent fo r inc luding CWIP in the ra tebase is that electricity rates come closer to reflect-ing the true c ost of incrementa l elec tric ity de -

mand, providing a more accurate incentive forconservation. Opponents of CWIP in the ratebase, howe ver, fea r that utilities will lose the in-c entive to keep plant c osts do wn a nd to finishthem on time. Furthermore, opponents claim,utilities may return to overbuilding. Many PUCshave responded to these concerns by including

only a portion of CWIP in the rate base (62).

Phased-in Rate Requirements. -At least sixStates have developed methods of phasing in therate requirements for large new nuclear power-




Box C.—Utility Accounting and the Origins of “The Money-Saving Rate Increase”1

The conventions of utility accounting have created a dilemma that affects all investmentchoicesbetween a capital-intensive plant (e.g., a nuclear or baseload coal plant) and a fuel-intensive plant (e.g.,a combustion turbine or an oil or gas steam plant}. if two plants (one of each type) have equal Ievelizedannual cost, and equal Iifecycle cost, the capital-intensive pIant will cost consumers far more in earlyyears and the fuel-intensive plant will cost far more in later years. This situation causes a dilemma for

oil- and gas-using utilities who wish to substitute coal or nuclear plants for their oil or gas plants.Whentheir analysis convinces them that the Iifecycle cost of the coal or nuclear plant will be less, they stillmust face a “money-saving rate increase” to cover the early years extra cost of the capital-intensive plant.

An Example. The dilemma is illustrated by a specific example in table Cl. A nuclear plant calledNuclear Plant X has been constructed for $2 billion (including accumulated AFUDC) and is about tobe placed in service to replace an oil plant that produces the same amount of electricity. The oil plantis old and fully depreciated and earns no return to capital. When the nuclear plant goes into servicethere will be a net fuel saving the first year of $263 million, which equals the value of oil saved lessthe cost of nuclear fuel for the nuclear plant. At the same time, accounting and ratemaking conventionsdictate that the first-year capital charge for the nuclear plant will be $471 million, or $208 million morethan the fuel savings. In the fifth year the capital charge has dropped to $364 million, less than thefuel savings for the first time (if the cost of fuel escalates at 9 percent per year), and by the eighth year

the cumulative capital charge of the nuclear plant will be less than the cumulative savings resultingfrom lower fuel costs. In the 15th year, the nuclear plant costs only $268 million in rate requirements

and saves $878 million in fuel costs.If fuel costs escalate more slowly (at only 5 percent per year), the results are shown in the right-

hand columns in table Cl. Annual capital charges for Nuclear Plant X drop below annual fuel savingsduring the sixth year and Nuclear Plant X breaks even on a cumulative basis by the 12th year. In bothcases, Nuclear Plant X will cost less over the 30-year life of the plant (if a discount rate of 12 percentis used). In the first case (with 9 percent fuel escalation), Nuclear Plant X will cost about $3.1 billionin lifetime discounted rate requirements and will save $5 billion. In the second case (with 5 percentannual fuel cost escalation), Nuclear Plant X will save $3.5 billion. In both cases the electricity ratepayerswould be better off with Nuclear Plant X over the long run. However, consumers would be worse offin the short run, because of the high capital charges at the beginning of plant operation, which translateinto high electricity rates.

Two Other Examples. To take another example of this phenomenon, which is sometimes referredto as “front-end loading” of capital costs, suppose another Nuclear Plant Y, with identical constructioncost (in 1982 dollars) as Nuclear Plant X had been placed in the rate base 8 years before, in 1974. By1982, the capital charge for Plant Y in the eighth year of operation would have diminished so much(using the same schedule of capital charges) that it would cost $0.03/kWh while the first-year capitalcharge for Nuclear Plant X, put in service in 1982, would be $0.09/kWh. Part of the reason for the dif-ference is that the book value (see explanation below of Plant Y is only $1.1 billion, the equivalentin 1982 dollars to the $2 billion cost of Nuclear Plant X. The rest of the difference is that the capitalcharge for the eighth year is only 0.15 of the original cost; compared to 0.24 in the first year.

In still another example, if Nuclear Plant X is replaced in its 30th year of operation by another identi-cal plant, the first-year capital charge for that plant will& $3.6 billion, more than 20 times the capitalcharge of $170 million in the 30th year of Nuclear Plant X.

Why Utility Accounting Practice Produces This Result. T he main reason for this result is that thevalue of a plant is carried at original cost (book value) not at replacement cost (market value) on a util-ity’s books.The annual capital charge used in computing rate requirements has a series of components,

‘The analysis in this box is bed on two ankles by My Hunt Waiter, ‘~rendhgthe Rate be;’ PuLrk Utilities Fbttnightiy, May 13, 1982, and “Avokiingthe h40neySaving Rate Increase,” Public Uti/ities hrtnighdy, June 24, 1962.

25-450 0 - 84 - 5 : QL 3




143

2,263

8 , 7 7 9



Ch. 3—The Uncertain Financial and Economic Future . 55

Figure Cl .—Rate Base and Revenue Requirements Under Original Cost Accounting

Current earningsDepreciated rate base x 0.12

Rate base under original cost accounting Straight line depreciationbeginning of each year Original rate base x 0.05 —

—

—General price levelIncluding fuel costs

It--k —

To ta l revenue requ i remen ts =.ea rn ings p lus dep rec ia t ion

—

$5 6

7 8 9 10 12 14 16 18 203 5 64

Year

Figure C2. -Rate Base and Revenue Requirements Under Trended Original Cost Accounting

Curren t ea rn ingsTrended deprec ia tedrate base x 0.05

Deprec ia t ion fo rmu la tha tIncreases deprec ia t ion

—

-—

m1

charges wi th nomina lInterest rate

Rate base under trended

koriginal cost accounting

Genera l price levelIncludlng fuel costs

Total revenuerequl rements =earn ings p lusd e p r e d a t i o n

18 20

I I l-r

iiii3 4 5 6 7 9 1 0 12 14 16

Year




plants which w ould c ause signific ant ea rly yea rrate increases under conventional rate treatments(75). Two of these (New York and Illinois) havedeveloped plans for “negative CWIP.” Underthese schemes, some CWIP is allowed in the ratebase for several years before a plant comes on-line and then an equal amount is subtracted from

the rate b ase for the first few yea rs a fter the p la ntcomes online. After the first few years the rateba se returns to w hat it wo uld have be en in theab sence o f any CWIP.

Pay-As-You-Go for Inflation Schemes.—Re-cently, a series of proposals have been made toad ap t the seq uenc e o f rate req uireme nts for a

capital-intensive plant, (e.g., a nuclear plant)mo re explic itly to inflation (54,88). In effect, theseproposals would eliminate the front-end loadingof capital return for capital-intensive plants in timeof inflation and replac e the “ do wnwa rd slop e”of annual rate requirements in constant dollars

(shown in fig. Cl in box C) with a horizontal orgentle upward slope more like the sequence ofannual rate requirements for an oil plant (shownin fig . C2 in box C).

Some of these proposals would do this direct-ly by using a rate o f return net o f inflation and

adjusting the rate base for inflation (this is called“ trend ed original co st” rate ma king in c ontrast to“original cost” ratemaking), Others would do itindirec tly b y d eferring c ertain o pe rating o r de-preciation expenditures until later so as to ap-proxima te a n up wa rd slop e o f rate req uireme nts.

The Obstacles to a Long-Term CommissionPerspective.– In principle, these latter proposalswould all make it easier for PUCs to increasethe authorized real return on equity because rateincreases for new powerplants are less likely.Increasing the authorized return shouldin turn improve the incentives for constructingnew po we rplants whe n it is in the long-te rm in-terests of the ratepayers. All these proposals, how-ever, rely on an implicit agreement between in-vestors and PUCs that a particular way of deter-mining revenues will be ma intained over de c -

ad es. Und er Trend ed Original Co st sc heme s, utili-ty investors accept a lower rate of return in earlyyears with the promise that the rate base will befully adjusted for inflation. Under indexed rate

of return schemes, investors accept a somewhat

lower than market rate of return as interest ratesare going up, in return for enjoying a somewhat-higher-than-ma rket rate of return a s interest rate sare coming back down.

It is just this implicit agreement that seems tobe missing from today’s rate regulation proce-dures. in some cases, the PUC may be willing to

work out a sensible approach to rate determina-tion over the long term, but is blocked by the

Sta te leg islature. The Ind iana PUC, fo r exam p le,introduced a graduated rate increase incor-

porating trended original cost principles to bringMa rb le Hill, a large nuc lea r plant, into the rateb ase o f Ind iana Pub lic Service . The p lan w a s ex-p lic itly bloc ked b y the Sta te leg islature. Eight

Sta tes, by vote of the Sta te legislature or b yreferendum, have banned CWIP inclusions inutility rate bases for just the reasons described

above (41).

Furthermore, commissioners may lack the timeor motivation to grasp the long-term view. Penn-sylvan ia is one o f the fe w Sta tes with 1()-yea rterms for its appointed commissioners. ManyStates have reduced PUC terms of 6 or longer to4 or 5 years. An increasing number of States haveelected rather than appointed commissioners. Formost commissioners, electric utility rate cases areonly a few among hundreds or thousands of casesfrom local as well as statewide utilities, that pro-vide wa ter, sew er, telephone , and ga s as well aselec tric ity. Often, elec tric utilities and their con-sumers must take time to educate commissioners

about the issues surrounding electricity supply,de ma nd a nd rates over the long te rm.

It is interesting to note (see ch. 7) that the

United States is the only one of all the major de-veloped countries with a Federal system in whichretail electricity rates are regulated at the Statelevel. In many countries, electricity rates are un-reg ulated . In West Ge rma ny, Sta te e lec tric a u-

thorities set their own ra tes sub jec t to Fed eral a p-proval. In the United States, State regulation leadsto the result that the cost of utility capital (returnon eq uity) varies a mo ng the different State s from

12 to 17 percent, even though the market for cap-ital generally is recognized as national. Becauseof the strong U.S. Federal tradition, however, anyprop osa ls for reg iona l or Fed eral dete rmination

of the c ost of c ap ital or other reg ulation on Sta te



Ch. 3—The Uncertain Financial and Economic Future q 5 7

regulation must be developed in the context oflongsta nd ing leg a l trad itions about the Fed eralregulat ion of commerce.

The Impact of Changes in Rate Regulation inElectricity Prices.–Changes in rate regulation toincrease the return to capital, utility cash flow,and/or quality of earnings, in turn would increase

electricity rates. How much rates would increaseis important to know for two reasons. It wouldhelp to weigh current consumer interests againstfuture consumer interests. It would help also toidentify the likely future course of electricitypric es and the resulting imp ac t on electric ity d e-mand. Uncertainty about future electricity priceincreases is a key source of uncertainty abouthow muc h elec tric ity de ma nd will increase.

Most of the a ttemp ts to estima te the impa c t ofchanges in rate regulation on electricity rates havefoc used on reg ions (48,85). There a p pea rs to b e

minima l imp ac t on a verag e regional elec tric ityp ric es from inc rea sing the return to cap ita l andincluding CWIP in the rate base—an increase inaverage regional electricity prices of 2 to 3 per-cent. Regional analysis of rate impacts, however,c omb ine the experienc es of q uite different util-ities.

For two individual utilities, examined as casestudies in a recent analysis, the impacts of ratereguIation changes wouId be significant but fair-Iy short-lived (2). Increasing the average rate ofreturn in 1982 from 12 to 16 percent, for exam-ple, would ha ve c aused a 1 -year increa se o f 3.3

percent in the rates of a Southeastern utility anda 6.2-percent increase in rates for a Midwesternutility. Rates would have stabilized in the follow-ing years.

The imp ac t o f CWIP on ra tes is estima ted tobe greater but also fairly short-lived. IncludingCWIP in rate base in 1982 would have increasedrate s 5 percent fo r the Southe astern utility and

14.2 percent for the Midwestern utility. Eight

years later, however, in 1990, the rates would beonly 0.4 percent higher than without CWIP forthe Southe astern utility and wo uld ac tua lly belower for the Midwestern utility. Although short-Iived , the inc rea se in rates is large eno ugh tha tthere w ould b e a substantial imp ac t on electric i-

ty demand, spread out, to be sure, over a numberof years.

Before PUCs can tackle fully the long-term im-plications of possible rate regulatory changes, itwould be useful to have a more complete under-standing of the impacts on rates and potential de-mand responses.

Conclusion.– Because of the financial deteri-

oration experienced in the 1970’s, utilities do nothave the financial reserves tha t the y ha d in the

late 1960’s and must therefore pay more atten-tion to the impact of their future construction pro-

gram s on their future financ ial health. Althoug htheir finances are improving, utilities are likelyagain to f ind themselves in weakened conditionsimila r to t ha t e xperienc ed in the 1970’s if theyembark on another round of large-scale construc-tion later th is c entury. This is espec ially true if in-f lation increases again and exacerbates the im-pa c t of AFUDC and the front-end load ing o f raterequirements for such capital-intensive projectsas nuclea r plants.

The last section of the chapter discusses utilitystrategies. One element of choice for both utilities

and PUCs is the tradeoff between short-term priceincreases from rate regulation policies designedto be more attractive to capital and longer termprice decreases projected to come about fromconstruction of central station powerplants (in-

cluding nuclear) which are expected to be thelowest cost source of baseload power over theirlifetimes.

THE COST OF BUILDING AND OPERATING

NUCLEAR POWERPLANTSIn addition to the bias against capital-intensive ma jor financ ial rea sons to b e w ary of investing

generation caused by current ratemaking prac- in nuclear powerplants. First, the cost of buildingtice, investors and utility executives cite several a nuclear powerplant has increased rapidly over




the past de c ad e. The estima ted c ost of the a ver-age nuclear plant now being completed is so highthat the average coal plant, in most cases, would

produce electricity more cheaply over a l ifetime(although in most regions the lowest cost nuclearplants are still competitive with coal plants). Sec-ond, the average construction time* of a nuclear

plant has increased much faster than the averageIeadtimes for a coal plant, and this makes it hard-er for nuclea r plants than c oa l plants to m atc hdem and . Third , since the Three M ile Island ac -cident, it has been widely recognized that a ma-

jor accident can disable an entire plant for an ex-tended period of t ime and require more than $1billion in cleanup costs, as well as other expensesto restart the plant and pay for replacement pow-er. Ma jor d isabling a c c idents a t coa l p lants a reboth less likely and less costly to cleanup andrepair. Finally, the current political climate fornuclear (described in ch. 8) is a major source of

financ ial risk. The o utp ut o f a seve ral billion d ollarinvestme nt in a p lant c ould b e lost for a yea r ormore if regulatory commissions refuse to put itin the rate base, or if a statewide anti-nuclearrefe rend um p asses and the plant is shut d ow n.Nuclear accidents or near-accidents in plantsow ned by other utilit ies c an lea d to a new seriesof sa fety regu lations (d isc ussed further in c h. 6)req uiring b ac kfits tha t ma y c ost a sizab le frac tionof the original cost of the plant.

The Rapid Increase in Nuclear

Plant CostIn the early 1970’s, nuclear powerplants were

completed for a total cost of about $150 to $300/ kW. ** As of 1983, seven nuclear powerplants al-most complete and ready to come online will cost

from $1,000 to $3,000/kW, an increase of 550to 900 percent. General inflation alone would ac-c ount o nly for an increa se o f 115 percent from1971 to 1983. Inflat ion in co mp one nts of (lab or

and materials) used to build nuclear power-plants*** would a cc ount for a further increaseof about 20 percent.

*Defined as follows: for nuc/carp/ants, issue of construction per-

mit to commercial operation; for coal plants, order of boiler to com-

mercial operation.**In “mixed” dollars, see explanation below.* **As measured by the Handy-Whitman index. See below.

Several attempts have b een ma de to do cume ntand unde rstand the increase in co st of nuc lea rpower over the past decade (6,7,1 7,37,55,61 ,76).

The ta sk is d iffic ult bec ause the c ost d a ta c itedabove cannot be used for comparing plants overtime. The a bo ve estima tes a re c om po sites of c on-struction expenses paid in different years with dif-

ferent d olla r va lues, referred to a s “ mixed ” dol-lars. Most also include some interest that has beendeferred during construction, capitalized, andadded to the total capital cost (see box B in theprevious sec tion on CWIP and AFUDC). Theamount of interest that is capitalized varies fromSta te to Sta te and interest rates va ry from yea rto year.

The increase in c osts of nuc lear pow erplantsthrough the 1970’s was ana lyzed in a c a refullydoc ume nted stud y by Cha rles Kom anoff (55). Thecosts exclude interest during construction andwere adjusted for inflation, permitting com-

pa risons from yea r to ye a r. * Figure 15 is a p lot

of the costs per kW (expressed in 1982 dollars)

of individual powerplants with construction per-mits issued from 1967 to 1971. (It is mo re ac c u-rate to group the different generations of nuclearpo werplants by start da te than b y co mp letiondate. Later comp letion d ates, by de finition, willhave a disproportionate share of the delayed, andtherefore p rob ab ly mo re e xpensive, plants.)

For plants with construction permits issued

around 1967 (and generally completed in 1972-74), the d irec t c osts in 1982 do lla rs rang ed from

$400 to $500/kW. For plants with constructionpermits issued 3 years later, in 1970-71 (and com-pleted in 1976-78) the direct cost in 1982 dollarshad more than doubled to $900 to $1,300/kW.

A comparable analysis by Komanoff of the costsof plants currently under construction has beenc omp leted but will not b e pub lished until early1984 (56). Preliminary results show that the cost

of a typical plant continued to increase, and therange of cost experiences also has increased sincethe ea rly 1970’s. Figu re 16 co mp ares “ typic a l”plants completed in 1971 and 1978 (these are

*As described in Komanoff’s Powerplant Cost Escalation (55) app.

C, a standard pattern of cash payments was assumed for each plant

and then deflated using the Handy-Whitman index developed tomake inflation estimates of components used in powerplants.




Figure 15.–Costs of Nuclear Units WithConstruction Permits Issued, 1966-71

(without interest during construction)

q *

0

1967 1968 1969 1970 1971Construction permit issue date (by midpoint of year)

NOTE: Plant costs in mid-1979 dollars were escalated to mid-1962 dollarsusing the Handy-Whitman index for nuclear plant components (multi-plying by a factor of 1.276).

SOURCE: Updated by OTA from data In Charles Komanoff, Power Plant Cost Escalation, Komanoff Energy Associates 1961, republished by VanNostrand Reinhold, 1962.

constructed from a composite of characteristicsassociated with average and not high or lowcosts) with the full range of costs estimated fora group of 32 plants under construction for com-p letion in the 1980’s. The c ost (in 1982 do lla rs)

of a typical plant increased from about $430/kWin 1971 to $1,020/kW in 1978, to a range of $840to $3,540/ kW for plants und er c onstruction in1983. The median plant of this group is expectedto cost $1,725/kW and the average cost is some-what higher ($1,880/kW) reflecting the wide vari-

ation in costs at the upper end of this wide range.

The same inc rease a lso is ev ide nt in p a irs ofplants built by the same company and intendedto be identical except for regulatory changes andsome construction management improvements.The c ost of Florida Pow er & Light ’ s St. Luc ie 2when completed in 1983 ($1,700/kW in 1982 dol-

lars) wa s ab out 50 perce nt mo re tha n tha t of St.Lucie 1 completed in 1976. Commonwealth Edi-son’s Byron 1 and 2, to be completed in 1984and 1985 at an estimated cost of $1,100 to

Figure 16.—Total Capital Costs for Nuclear PlantsCompleted in 1971, 1978, and 1983=87 (estimated) in1982 Dollars Without Interest During Construction

4,000

3,000

2,000

1,000

0

Typicalplant1,020

3,540 highest

3 plants

2,770

5 plants

‘ 2,340

8 plants

1,725

8 plants1,350

5 plants , ,2 0

3 plants ‘840 lowest

1971 1978 1983-87

Date o f comp le t ion

NOTES: “Plant” is defined as a single nuclear site or station with one or morereactors, The plant costs in-mid-1979 dollars from Komanoff’s book for 1971 and 1978 plants were stripped of interest and escalated to mid-1962dollars using the Handy-Whitman index for nuclear plant components(multiplying by a factor of 1.276). The costs for the plants to be com-pleted in 1963-87 are based on mid-1963 utility estimates for a group of32 sites (stations) with a total of 50 reactors and excludes: Marble Hill,Waterford 3, Susquehanna and all Washington Public Power System(WPPS) plants except WPPS 2.

SOURCE: Charles Komanoff, Power Plant Cost Escalation, Komanoff EnergyAssociates, 1981, republished by Van Nostrand Reinhold 1982; un-published analysis from forthcoming report by Charles Komanoffand Irving C. Bupp to be published in the winter of 1984, and theOff Ice of Technology Assessment.

$1,150/ kW (in 1982 dollars) w ill co st a bout 90percent more than the company’s Zion 1 and 2,completed in 1973 and 1974.

Until the early 1980’s, nuclear plant costs in-

creased steadily with each generation of plants(55,61,75). How ever, Kom anoff’ s ana lysis showsno tendency for plants scheduled for completionlater in the 1980’s to have significantly greaterexpected costs than plants being completed in1983-84. In part, this may be due to underestima-tion o f c osts for p lants still far from c om p letion,but it also is probable that factors other than time

now are more influential on powerplant cost. (Ac om plete list of the mixed -dollar c ost for plantsin various sta ge s of c om p letion is g iven in ap p .table 3A.)




The va riation from lowe st to highest c ost nuc le-ar pow erp lants in the c urrent genera tion is strik-ing. Construction costs per kilowatt are expected

to be o ver four time s highe r (in 1982 do lla rs) forLong Island Lighting Co .’s Shoreham a t $3,500/

kW tha n the y are fo r Duke Pow er’s Mc Guire 1

and 2 at $840/kW. For the current generation of

pIants, Komanoff found some variables that ex-plain much of this large variation (56). For exam-ple, estimated plant cost decreases about 15 per-cent for every doubling of the number of mega-watts at a single site. Estimated plant cost alsodecreases 8 to 10 percent for each previous plantsite built by the same utility. Based on theseresults, a utility that had built on 5 previous plantsites should be able to construct its next plant sitefor 30 to 40 percent less than a utility with noexpe rience. Plant c ost a lso va ries by m anufac -turer (as much as 15 percent) and is significantlyhigher (30 to 40 percent) for plants located in the

Northeast region due primarily to higher laborcost and shorter construction seasons.

The imp ortanc e o f utility expe rienc e a nd site-related experience for this latest group of pIantsis evidenc e o f the impa c t of some of the utilitymanagerial experience described in chapter 5.There see ms to b e a site-sp ec ific an d c om p an y-specific learning-curve for bringing plant costsdown.

Reasons for IncreasedConstruction Cost

Several different kinds of increase contributedto the dramatic increase in average cost describedab ove. To be gin with, these c omp arative c ost e sti-mates exclude the influence of nuclear-compo-nent inflation tha t c omp ares the c ost of eq ual-quality nuclear components and materials overtime. Nuclear component prices increased 1 or

2 perce nta ge po ints faster than infla tion. *

Seve ra l c ha nge s ac c ount fo r the increa se in

constant dollar cost. According to several relatedDOE studies, materials used in nuclear plantshave increased, e.g., from an estimated 2,000ft/MW of cable for a typical plant to be con-

structe d in 1971 to a bout 5,000 ft/ MW of c able

*This is measured by the Handy-Whitman nuclear index (55).

Photo credit: Duke Power Co.

Capital costs per kilowatt of identical plants at a singlesite are usually lower than average. This photo showsCatawba nuclear station, owned by Duke Power Co.,which is expected to produce among the lowest cost

electricity of any plant in its timeframe whenit comes online in 1985

for the ave rag e o f eight p lants unde r construc -tion in 1982-85. Figure 17 shows similar inc rea sesin the use of concrete, piping and cable raceway

(supports for electric cable) (1 7). Increased ma-terial requirements are due both to direct in-creases in structural and electrical complexity andto the inc rea sed rew ork nec essary to m eet mo restringent quality-control requirements.

Materials also have become more complex. Awhole set of seismic requirements to restrain pip-ing systems during earthquakes was introducedin the la te 1970’s. Simple cast or machined pipesupports (costing several hundred dollars) have

been replaced with very sophisticated restraintscalled “snub bers,” with shock-absorbers, costing

many thousands of dollars. Pipe supports havebecome more massive and have had to be fitted




Figure 17.—Trends in Material Requirements in Estimates of PWR Construction Cost

PWR Wash 1230 (1971)

m PWR Nureg-0241 (1976)Data of 8 PWR’s (1982-85 comm. O P)

147 I

Concrete( y d3 per MW)

Highest

Average

Lowest

N A

570.

340

263

7323

4889

3966

Piping Ca b l e(ft. per MW) (ft. per MW)

225

151

70

Raceway(ft per MW)

NOTE The first two columns come from engineering estimates prepared for the NRC (formerly the AEC) based on construction data from plants under construction orcomplete at that time The last column comes from a survey of eight plants under construction See source article for references and more description

SOURCE John H. Crowley and Jerry D. Griffith, “’U S Construction Cost Rise Threatens Nuclear Option “ Nuclear Engineering International, June 1982

with muc h tighter toleranc es to the pipes theysupport.

Quality-control procedures and paperworkhave a dd ed to the c ost of materials and co m-po nents. Although the re ha s be en no c omp re-hensive study, there are individual examples andanecdotes to illustrate the claim that quality con-trol represents a bigger and bigger share ofnuclear materials cost. In one such example (86)structural steel supports now required for nuclearplants cost between two and three times the cost

of the same quality steel supports that are still

used for general construction projects and thatwere permitted on nuclear projects until 1975.Of this am ount, the q uality c ontrol proc ed ures

account for virtually all the increased cost.

Fina lly, there ha s be en a stea d y increa se in theam ount of labo r req uired pe r kW, both ma nual(craft) and nonmanual. For a series of typicalplants costed out over 15 years in a study for

DOE, craft labor requirements increased from 3.5workhours/kW for a plant starting constructionin 1967 to 21.6 workhours/kW for the averageof 16 plants under construction for completionin 1982-85 (17). Nonma nua l field and eng inee r-ing services also have increased dramatically. Fora slightly different series of typical plants,estimates of field and engineering services in-

creased from 1.3 workhours/kW in 1967 to 9.2wo rkhours/ kW in 1980 (16).

The increa se in lab or per kilow at t of c ap ac ityis the result of complex interactions resulting from




increasingly demanding regulations, quality-assur-ance requirements and the subsequent utilityma na ge me nt respo nse to these. These a re d e-scribed in more detail in chapters 4 and 5 andin several case studies (92).

There is la rge variat ion in ma terial a nd labo rrequ irem ents from p lant to p lant , just as the re is

la rge va riation in overall c a pita l co st. For a g roupof 16 plants scheduled to be completed in1982-86, craft labor varied from a low of 15workhours/kW to a high of 33 workhours/kW.Simila rly, for a g roup of e ight p lants, linea r fee tof cable varied from a low of about 3,300 ft/MWto a high of about 7,300 ft/MW (see fig. 17).

The Increase in NuclearConstruction Leadtimes

Nuclear construction Ieadtimes also increased

over the decade, making it increasingly difficultto match nuclear plants to demand, adding to in-terest a nd esc alation c osts, and exac erbating

problems with cash flow. At the same time, lead-times for coal plants increased very little (froman ave rag e o f 58 to a bo ut 60 months) (1).

Documenting the increase in Ieadtimes for nu-clear plants is made difficult by the fact that someplants have b een d elayed d elibe rate ly by their

utilities because of slow growth in electricity de-ma nd and financ ing diffic ulties. There a lso a p -pears to be important differences in the regulatoryenvironments for different generations of plantsthat m ust also b e ta ken into a c c ount.

A recent study of Ieadtimes for EPRI took bothdeliberate delays and regulatory stage into ac-

c ount* (1). The stud y ident ified from p ub lishedsources those plants that had been delayed delib-erately more than a year by their utilities andanalyzed their Ieadtimes in a separate group. Ina more d eta iled c ase study o f 26 of these p lantsEPRI found that 8 had been delayed significant-ly (averaging 27 months) while 22 had only been

delayed an average of 2.5 months.

Grouped by date of permit, it is plausible toidentify three g ene ra tions of nuc lear p lants. Forthe first g ene ra tion, for whic h c onstruction p er-mits were issued from 1966 to 1971, leadtimes* *increased stea d ily from about 60 to 80 months.

This ap pe ars to reflec t a n increa se in the designedcomplexity of nuclear powerplants and possiblythe strains of rapid growth as well.

A sec ond group of p lants had their c onstruc-

tion permits issued from 1971 to 1974. Leadt imesfor that group were much higher than the first,averag ing 120 months and ranging from 100 to160 months. Leadtimes for plants without signifi-cant deliberate delays averaged about 10 monthsless tha n tho se with sign ific a nt d elibe rate de lays.It appears that this group of plants suffered a ma-

jor increase in reg ula tory c om p lexity (including

the 1974 Calvert Cliffs decision, the regulationsfollowing the 1976 Browns Ferry fire and the 1979a c c id ent a t Three Mile Isla nd ) without the op po r-tunity to develop construction and regulatoryplanning techniques to handle the increased

c om plexity. They m ay ha ve suffered as well fromsome of the effects of rapid growth in the indus-try, such as incomplete designs and inexperi-

enc ed sup ervisors.

Although the data are sketchy, it is possible that

a third generation of nuclear plants is now emerg-

*The study grouped the plants by date of construction permitto avoid the obvious problem that later completion dates, by defini-tion, include a larger proportion of Iong-leadtime plants.

* *Leadtimes for this analysis are defined as time from date of con-

struction permit to commercial operation.




ing, with construction permits issued later, in1975-77. After adjusting for deliberate delays andexcess optimism in time estimates, EPRI foundtha t this la test group of p lants ap pea rs to ha ve

somewhat shorter Ieadtimes than the 1971-74group . Lea dt imes for all p lants in the g roup av-erage ab out 100 months and rang e from 65 to120 months. Plants without significant announceddeliberate delays averag e 10 to 15 months lessthan the a verag e. The plants with shortest lead -times in this group are already in operation andwere completed faster than the shortest Ieadtimep lants in the ea rlier group (see fig. 18). The n um-bers are so sma ll, how ever, tha t it is too e arly totell if these p lants a re a nything b ut a nom a lies.Those p lants with long er Iead times in this late stgroup still may experience significant delays be-yond the adjusted estimates calculated by EPRI.At the same time the re is som e c ase stud y evi-dence that the plants that were started later were

able to compensate for increased regulatory com-plexity in the plant design and construction man-agement and were also able to plan systematically

their dealings with the NRC. (Case Study 2 in ch.6.)

The Impact of Delay on Cost

In a period of substantial general inflation char-ac teristic o f the last 5 yea rs, a d elay in nuc lear

plant construction can cause an alarming increasein the c urrent d olla r c ost of the plant. Inc rea sesin the current dollar cost, however, must be dis-tinguished carefully from increases in the real orconstant dollar cost of the plant (after the impact

of g ene ral inflation has be en eliminate d). Thesein turn must be distinguished from increases dueto changes in regulations or other external influ-enc e during the p eriod o f delay.

For a hypothetical plant that has been expectedto b e c om pleted in 8 years but instea d ha s be endelayed to 12 years, with no increase in complex-

ity or sc op e, the re a re tw o source s of inc rea sesin total capital cost in constant dollars. One is that

nuclear components, materia ls and labor may

Figure 18.—Construction Leadtimes for Nuclear Powerplants

I I

q

q

q

q

U

I

q I

q q

q

q

q

q

1965 1970 1975 1980

Construction permit issue date

NOTES. The Ieadtimes are based on estimated times to commercial operation for those plants not yet in service. The gapscorrespond to periods of Iicensing inactivity in the industry Leadtimes are calcuIated from construction permitissue date-to-date of commercial operation

SOURCE Applied Decls!on Analysis, Inc An Ana/ys/s of Power Plant Construction Lead T/rrres, Volume 7” Analys/s andResults, EPRI. EA.2880 February 1983 Graph based on Nuclear Regulatory Commlsslon data.



64 Ž Nuclear Power in an Age of Uncertainty

have increased 1 or 2 percent faster than generalinfla tion (esc a lation). The sec ond is real inte restduring c onstruction* that is ca pitalized andadded to general total plant cost as AFUDC (Al-lowance for Funds Used During Construction).

(See bo x B ab ove .)

Ta b le 9 show s the increa ses in co nsta nt d ollars

and in current dollars of seve ral different c ases:5, 7, and 9 percent general inflation with no realescalation in nuclear components and with 2 per-c ent rea l esc alat ion a nd rea l interest rates of 3percent and 5 percent above general inflation(1 3). Several of the examples in table 9 can serveas illustrations of the difference between increasesin current and constant d ollars. For exam ple, incase 3, if a plant takes 12 years to build duringa period of general inflation of 7 percent, escala-tion of nuclear components of 9 percent (2 per-centage points faster than general inflation) andan interest rate of 12 percent (a rather high real

interest rate of 5 percent), the “mixed currentdollar cost” of the plant will be 233 percent high-er tha n its overnight c on struc tion c ost. Two -third sof the increase, however, is general inflation. Therea l co nstan t d ollar increase in the p lan t c ost isonly 48 percent. Construction of the same plantin 8 yea rs time w ould c ause a c urrent d olla r in-c rea se o f only 123 pe rc ent a nd a co nstant d ollar

*Real interest is the nominal rate of interest less the rate of general

inflation, e.g., real interest is 5 percent for nominal interest ratesof 12 percent when general inflation is 7 percent.

increase of 30 percent. For this case, shorteningthe plant’s Ieadtime would save about a third ofits current dollar cost but only about 12 percentof its constant dollar cost. *

The Cost of Electricity From Coal andNuclear Plants

The stead ily inc rea sing c ap ital co sts of nuc lea rpower (including the increasing costs broughtab out b y increasing Iea dtimes) lea ds to a c rucialquestion: at w hat po int d oes the increasing c ap -ital cost of nuclear plants make nuclear power

a more expensive source of electricity comparedto alternative generating sources, especially coal?As long as it is likely that utilities will avoid theuse of oil and gas for base load electricity gener-ation, the chief competitor to nuclear is coal.

Initially (for most nuclear plants completed bythe early or mid-1970’s) there was no doubt that

electricity generated from these plants was sub-stantially cheaper than coal-generated electrici-ty. Because of the way capital charges are recov-ered in the rate base (see box C above), the costof electricity from these plants has become steadi-ly c hea pe r relative to electric ity from c oa l plantsbuilt at the same time. As the capital cost ofnuclear plants has risen, however, the relative ad-

*In this case 2.23 (8 years) is about 67 percent of 3.33 (12 years)and 1.30 (8 years) is about 88 percent of 1.48 (12 years).

Table 9.—Additions to Overnight Construction Cost Due to Inflation, Escalationand Interest During Construction (in constant and current dollars)

Percent increase in Percent increase inconstant do llars current dollars

8-year 12-year 8-year 12-yearIeadtime Ieadtime Ieadtime Ieadtime

Case 1: 7% general inflation,0 escalation, 100/0 interest rate . . . . . + 12 + 19 +92 + 167Case 2: 7% general inflation,0 escalation, 12% interest rate . . . . . +21 +33 + 107 + 200Case 3: 7% inflation, 2% escalation,12% interest rate. . . . . . . . . . . . . . . . . +30 +48 + 123 + 233Case 4: 5% inflation, O escalation,100/0 interest rate. . . . . . . . . . . . . . . . . — +34 — + 140Case 5: 90/0 inflation, O escalation,14% interest rate. . . . . . . . . . . . . . . . . — +3 3 — + 273

NOTE: “Escalation” is defined as the increase in the unit costs of labor and components used in nuclear plants (with no changein quality) above the rate of general inflation. For inflation of 7 percent, an interest rate of 10 percent corresponds toa real interest rate of 3 percent, an interest rate of 12 percent corresponds to a real interest rate of 5 percent.

SOURCE: For the calculation, Wilfred H. Comtois, “Escalation Interest During Construction and Power Plant Schedules,”Westinghouse Power Systems Marketing, September 1975.




vantage of nuclear power has diminished. Forplants presently under construction, the averagecapital cost is now so high that the typical nuclearplant probably would produce more expensiveelectricity over its life time than the typical coalp lant. Only elec tricity from the lea st expensive

nuc lea r plants still ma y be c om petitive with av -erage cost coal-generated electricity. Average costnuclear plants, however, still can compete withmo re expensive co al plant s.

Com pa ring the c osts of nuc lea r and c oa l-firedelectricity is made difficult by the different im-pa ct o f fuel and c ap ital cost co mpo nents for eac htype of plant. Capital cost is a far more impor-tant component of nuclear-generated electricitytha n for coa l plants. While the Ieve lized c ost offuel (uranium ore, enrichment, storage shipmentand disposal) and operations and maintenanceea c h run a bo ut $0.0075/kWh, the c ap ital c harge

(Ievelized charge* over the life of the plant) perkWh for new p lants ma y range from as little as$0.01/kWh for older reactors to $0.10/kWh or

even more for the most expensive of today’s reac-to rs. The c ap ital cha rge p er kWh increa ses with

higher total construction cost (including the im-pact of longer Ieadtimes), with higher interestrate s, and with a shorter c ap ital rec overy pe riod .The c apita l cha rge (as we ll as op erations andmaintenance) per kWh also increases as the plant

capacity factor* * is reduced because there is lessoutput among which to apportion the annualcapital cost. Since the earliest nuclear plants were

built, there have been significant increases in allthe categories that increase annual capital

c harges. The c ap ac ity fac tors of nuc lea r plantsalso have been less than expected. (See ch. 5 formo re d isc ussion of nuc lear ca pa c ity fac tors.)

For coal-fired electricity, on the other hand, thecost of the fuel is at least as important as thecapital cost in determining the price of electrici-ty over the life of the plant. Operations and main-tenance of coal plants cost somewhat less than

*Various techniques are used to “levelize” costs over a plant’s

lifetime. One simple method, used in the EIA study of coal andnuclear costs, is to take the present discounted value of the streamof costs and divide it by the number of years to get an annual level-ized cost (36).

**Capacity factor equals the number of hours of actual opera-tion divided by the hours in the year.

for nuclear plants. Fuel cost, however, may rangefrom less than $0.01/kWh in regions where plantsc an b e b uilt nea r the c oa l mine to almost fourtimes that in regions located far from coal fields(assuming rapid increases in coal prices). The rateat which c oa l pric es are likely to e sc alate overseveral decades has a significant influence on theforecast average cost of electricity from the plant

over its lifetime. Levelized electricity prices willbe about $0.015/kWh (in constant dollars), high-er, on a verage , if co al pric es esc alate at a rea lannual rate of 4 percent than if they don’t escalateat all in real terms (36).

Unfortunate ly, there is no stud y of rec ent plan tsusing ac tual rep orted c ap ital co st of c oa l plants.In a DOE study (36) using coal and nuclear plantmodel data on capital cost, the cost of electrici-ty is about eq ua l in five o f the ten DOE reg ions,slightly lower for nuclear in two of the regions

and considerably lower for coal in two of the re-gions. There is reason to believe, however thatnuclear capital costs are higher and coal capitalcosts may be lower than the study results. Capi -tal costs of the typical nuclear plant reported inthe study a re a bo ut 15 pe rc ent lower than theaverag e (in constant dolla rs) of the p lants now

unde r construc tion. On the o ther hand, the c ap i-tal cost of the typical coal plant reported in theDOE stud y is mo re t han 40 pe rc ent higher tha nthe c ap ital co st of the typica l 1978 co al plant (in-c lud ing a flue-ga s desulphuriza tion sc rubbe r) inthe 1981 Koma noff study up da ted to c onstant

1982 dollars. While it is possible tha t the c ap italc ost of c oa l plants ma y have inc rea sed 40 pe r-

cent since 1978, several factors make it unlike-ly. Coal plant construction Ieadtimes (unlike nu-clear plant Ieadtimes) have not increased since1978. Since the cost of scrubbers already is in-cluded in the 1978 typical coal plant capital cost,it is unlikely that further pollution control im-provements and design improvements would addmore than 20 percent. If, indeed, actual nuclearconstruction costs are higher and actual coalplant construction costs are lower than the plantmodel results, the typical nuclear plant would beexpected to produce more expensive electricityin all regions.

Low-cost nuclear plants, however, still wouldbe competitive with the average coal plant. Com-




Photo credits: Atomic Industrial Forum

Nuclear fuel comes in pellets and is assembled intofuel rods and then into fuel assemblies. Fuel for anuclear powerplant is compact and only transportedevery 12 to 18 months. It is inexpensive compared tothe cost of alternative fuels such as coal, natural ga s,

and oil. Each Vi-inch-long pellet of enriched uraniumshown here can generate approximately the same

amount of electricity as 1 ton of coal

monwealth Edison Byron plants are expected tocost $1,100 to $1, 150/kW in 1982 dollars (withoutinterest during construction)* and Duke Power’sMcGuire and Catawba plants are expected to cost

$900 to $1,200/kW in 1982 do lla rs. When thec onstruction c ost of a nuc lear plant is no m orethan 20 to 40 percent above that of a coal plant,i t can be expected to produce electrici ty morecheaply, sometimes substantially, over the plant’s

lifetime.

*Costs of the Byron plants may go up, however, following aJanuary 1984 NRC decision to deny the plants an operating license.

However, there is a further element of the com-petition between the costs of electricity from

nuc lear and from c oa l. The p a ttern of co sts ove rthe life of t he p lants a re substantially d ifferent.Under current accounting rules capital chargesare highest in the early years and decrease as de-preciation charges are deducted from the assetbase. Coal costs on the other hand will increaseat, or faster than, the rate of general inflation overthe life o f the p lant. Thus for p lants with the sameIeve lized c ost o f elec tric ity, elec tric ity from thenuclear plant will cost more in the early years andelectricity from the coal plant will cost more inlater yea rs (see b ox C a b ove ). The highe r co st ofnuclear plants in the early years could be dis-couraging to an electric utility that had facedmuch opposition to rate increases.

Future Construction Costs of

Nuclear PowerplantsIt is very unlikely that there will be any future

for nuclear plants of current average capital cost.Nuclear plants, on average, are now so costly thatthey a re no long er likely to produc e electric itymore cheaply than coal over their lifetimes. Giventhe p attern of front-end loa ding o f c ap ital c osts,even with equal lifetime costs, nuclear-generatedelec trici ty would not be chea per than c oa l-gen-erated electricity for 10 to 15 years. For thisreason, utilities are not likely to order morenuclear plants if the capital cost of newly ordered

plants is expected to repeat that of the currentaverage plant.

There is c onsiderab le evide nc e that , with ef-fort, the cost of an average nuclear plant can be

red uc ed sub sta ntially from the c urrent level. Thelowest cost nuclear plants already cost substan-tially less than the average. Within the presentframewo rk of regulato ry req uirem ents for plant

design and quality control, a few utilities havebuilt enough plants to take advantage of a con-struction learning curve and have developedtechniques for minimizing delays, rework, andworker id leness. These tec hniqu es a re de sc ribedin more d eta il in c ha pte r 5. They involve c a refuland complete engineering, careful and thoroughplanning and project management (including theuse of sophisticated computerized tracking and




inventory planning), and attention to motivationfor prod uctivity and c oop eration. If a sep aratec omp any is used to ma nag e the p roject, theremust be explicit incentives to complete projectson time and at budgeted cost. A reasonable goalfor such efforts could be to equal the capital costof Commonwealth Edison’s Byron and Braid-wood plants and Duke Power’s McGuire/Cataw-ba plants of $1,100/kW (1982 dollars) direct con-struction cost plus another $200 to $250/kW forinterest during construction.

Looking overseas, there is evidence that a dif-ferent approach to certain aspects of safety regu-lation a nd qua lity c ontrol c ould b ring c onstruc-tion costs down still further. Constructing a Frenchplant requires about half as many workhours/kWas constructing an American plant (86). As is de-scribed in more detail in chapter 7, the French

have standardized their plants and have built twobasic types of reactors (925 MW and 1,300 MW).This avo ids muc h of the rewo rk that occurs in

American plants because the engineering is es-sentially complete before each plant is started,and because the first plant of each type functionsas a full-sc ale mo de l to help a void p iping a ndc ab le interferenc es and other problems of twodime nsional d esign . Within a far m ore c entralizedand controlled approach to safety regulation (see

ch. 7) the French have also taken an approachto earthquake protection that minimizes the im-pa c t on c onstruc tion. They a lso ha ve a d ifferent

ap proac h to qua lity c ontrol that m inimizes de laysduring c onstruction (d esc ribed in chs. 4 and 5).

A specific estimate of possible reductions in av-

erage plant cost was made in a recent study ofthe U.S. nuclear industry viability (87). Accordingto this estimate, typical plant costs (including in-terest during construction) could be reduced fromabout $2,220/ kW (in 1982 do lla rs) to $1,700 to$1,800/kW (20 to 25 perc en t less). This estima teassumes that the United States can go only partway towards constructing plants with as fewworkhours as is now done in France. The French

regulatory environment, and utility management,and construction tradition are sufficiently different

that much is unlikely to be duplicated in the

United Sta tes. The p roposed steps to b ring c on -struction costs down would include:

q

q

q

Reduction in Construction Workhours. Afully sta nda rdized p re-certified design a ndemphasis on multi-unit sites could reduceconstruction workhours from 14 million to12 million per 1,300-MW plant, a numberwhich is still about 25 percent higher thanestimated construction workhours for a com-parab le Frenc h p lant. This wo uld c om eabout through progress up a learning curveof construction management techniques.Reduction in Engineering Workhours. Stand-ardization and regulatory predictability alsoc ould c ut eng ineering wo rkhours pe r plantroughly in half from about 9 million to about4.5 million, by reducing construction engi-neering support by more than 80 percentand quality assurance workhours similarly.Engineering workhours would still be about60 perce nt highe r than tho se in Franc e, re-

flec ting the d ifferenc es in U.S. co nstruc tionproject organization.Eight-Year Project Schedule. Reducingaverage project schedules from 11 to 8 yearswo uld red uc e interest a nd rea l esc alationc osts. Tota l co nstruction c ost in c onstant d ol-lars wo uld b e reduc ed by 7 to 15 pe rc ent(see table 9).

The d esirab ility of sta nd a rdiza tion in nuc learplant design and construction is a complex ques-tion that would affect far more than the capitalcost of nuclear plants. It was the subject of ap rev ious OTA report, Nuclear Powerplant Stand- ardization (April 1981 ) and is discussed furtherin chapter 4. One issue is whether standardiza-t ion c an b e a chieved w ithout sac rificing the a da p-tabil ity of nuclear technology to new informationab out d esigns that w ould b enefit nuclear plantsafety or operation. A second issue is the institu-tional obstacles to standardization in an industrywith more than 60 nuclear utilities, 4 reactor ven-dors, and more than 10 AE firms.

While op po rtunities exist fo r red uc ing nuc lea rc onstruc tion c osts, it should be rec og nized tha tcosts might also increase. Further serious ac-

cidents could lead to a new round of majorchanges in regu lation. There are still imp ortant

unresolved safety issues (discussed in ch. 4) that

could lead to costly new regulations. utility ex-

ecutives are well aware of this possibility.




The Financial Risk of Operating aNuclear Powerplant

The Risk of Property Damage.–The accidentat Three Mile Island was a watershed in U.S.nuclear power history because it proved that

serious accidents could indeed occur and causeenormous property losses even without causingany o ffsite da ma ge. The tota l cost of the c leanupis estimated at $1 billion, not counting the carry-ing costs and amortization of the original capitalused in building the plant nor the cost of restart-ing the p lant. O f this $1 billion , $300 million wa sc overed by insuranc e from the insuranc e p oo l

(see tab le 10 for explana tions). Gene ral p ub lic

Utilities (GPU) is now in the process of negotiatingthe financing of the rest from various sources in-cluding the utility industry through the EdisonElectric Institute, the rate payers as approved by

the Pennsylvania and New Jersey PUCs, the Statesof Pennsylvania and New Jersey and the FederalGovernment .

Since Three Mile Island , prop erty insuranc ecoverage for nuclear pIants has increased to $1billion, a bout ha lf in p rimary insuranc e a nd therest in excess insurance (once a $500 million ac-c ide nt c ost has be en reac hed ). Some of the p ri-ma ry insuranc e a nd mo st of the e xcess insura nc ehave b een p rovided by two new m utual insur-

ance companies formed by groups of utilities,Nuclear Mutual Ltd. (NML) and Nuclear EIectricInsuranc e Ltd . (NEIL) –(see ta b le 10). NEIL a lsoprovides almost $200 million in insurance for pur-chases of replacement power while a plant isdisabled.

Despite this threefold increase in insurance,how eve r, som e o f the expe nses inc urred in the

Three M ile Isla nd a c c id ent still have not be en in-sured; namely, maintenance of the disabled plantand carrying costs and amortization of the capital

tied up in the disabled plant. For the moment

these a re b eing p a id by GPU stoc kholders whohave not received a dividend since the accident(44).

Table 10.—Nuclear Plant Property and Liability Insurance

Description Coverage

ANI-MAERP:Commercial insurance consortium Reactors at 34 sites. “Primary”

of about 140 investor-owned insurance (responds initiallycompanies (American Nuclear to a loss) $500 million/siteInsurers - ANI) and 120 mutual $68 million excesscompanies (Mutual Atomic EnergyReinsurance Pool — MAERP)

NML:Nuclear Mutual Limited is a Reactors at 27 sites, $500

mutual insurance company created million primary insurance/siteby several investor-ownedutilities and located in Bermuda

NEIL-I:Nuclear Electric Insurance Limited — Reactors at 36 sites, $2.3 million/

extra expense insurance to pay week 1st year; $1.15 million/weekfor replacement power from an 2nd year up to $195 millionaccident covered in primaryinsurance

NEIL-II:Nuclear Electric Insurance Limited — Reactors at 32 sites. “Excess”

property damage excess damage insurance (covers damage aboveabove limit of primary insurance limit of primary insurance) $415

Liability insurance: million/site

ANI-MAERP:Liabil ity insurance required by All reactors. $160 mil lion available

Price-Anderson from premiums, plus $400 million/ accident available from retroactiveassessments of $5 million/reactor/ accident

SOURCE: John D. Long, Nuclear Property Insurance: Status and Outlook, report for the U.S. Nuclear Regulatory Commission,NUREG-0891, May 1982; papers presented at Atomic Industrial Forum Conference on Nuclear Insurance, Feb.14-16, 1983.




Further increases in insurance capacity are de-sirable, given the increasing replacement valueof plants. However, they are limited by the assetsof the insurance companies and the utilities inthis country and by the reluctance of reinsurers,suc h a s individua l synd ica tors in Lloyd ’s of Lon-don, to assume any more American nuclear risk.

The a deq uac y of c urrent insuranc e is threat-

ened further by several other issues. Under publicpressure, Congress could stipulate that all prop-erty insurance be used first to pay cleanup costsand only then to pay carrying costs and the costsof restarting the p la nt. This wo uld m ea n tha t theutility would have to turn to the PUC to obtainhigher electricity rates to cover all the other costs

of an accident or obtain the funds by withholdingdividends from shareholders. Another source of

uncertainty is the heavy reliance of the utility-run

mutual insurance systems on retroactive assess-me nts. These are p remiums tha t the utility co m-mits itself to pay in the event of an accident. NML,for examp le, may c all on retroa c tive p ayme ntsup to 14 times annua l p rem iums in the e vent o fa serious accident that depletes existing insurancereserves. The willing ness and ab ility o f ut ilities

to pay these assessments has not yet been tested.Som e ob servers have expressed the fea r thatPUCs may balk at allowing utility insurance ex-penses “to pay for the other guy’s accident” (57).

The Risk of p ub lic Liab ility .–Sinc e 1957, the

Price-Anderson Act has limited public liability, inthe eve nt o f a serious ac c ident , to $560 million.Pressure to increase this limit has been mounting.Inflation alone would justify raising the limit to

about $1.9 billion assuming $560 million wa s anappropriate figure in 1957. Pressure, however,to go beyond keeping up with inflation, or evento eliminate the limit altogether, arises fromseve ral stud ies pub lished over the last 10 yea rs,the Rasmussen Report (WASH-1900) in 1974 andthe 1982 Sandia siting study. Both analyze theconsequences from low probability accidents andare described more fully in chapter 8. Part of the

Price-Anderson Act is due to expire in 1987. Thefirst round of debate in Congress on this issue may

be gin soon, stimulated by the rece nt p ublic ationof an NRC report on public liability.

Ironica lly, this p ressure to increa se the limit fo r

p ub lic lia bility c om es just a s the priva te insuranc eresources of the industry have increased enoughto cover the current statutory limit fully. About$160 million is available from the insurancepoo ls, and the rest (a bout $400 million) is ava il-able from a $5 million per reactor retroactive as-sessment required in the Price-Anderson Act.Under current law, the Federal Government hasguaranteed to provide any liabil ity damages be-yond the nuclear industry’s resources and up tothe statutory limit. Currently, because of the avail-ability of private insurance funds and the increasein the numb er of plants ava ilab le to pa y the $5million assessment, the Federal Government hasno liab ility. if the limit we re raised or elimina ted ,

there would be pressure for the Federal Govern-me nt to aga in assume the excess liab ility.

From the standpoint of the insurance industry,

the most serious issue is the growing pressurefrom c itizen’ s group s to a llow dama ge from nu-clear accidents to be covered in homeowners’policies. Currently, by consensus of all insurers,a ll hom eo wne rs’ insuranc e polic ies spe c ific a llyexclude a nuclear accident as an insurable risk.Homeowners, in effect, may make claims only

aga inst the responsible utility and be p a id fromthe utility’s ow n insurance resources. For insurers,this characteristic of nuclear insurance channelsthe risk into a single c ate go ry which c an be ide n-tified and assessed. Since the potential damagesare both large and of unknown probability, insur-

ers are m uc h m ore w illing to provide fairly largesums if the structure of risk is simple because agiven a c c ident w ill result on ly in a c laim from asingle utility, and not in hundreds of individualhomeowner claims through multiple insurancecompanies. Were the homeowner’s insuranceexclusion to b e removed by law, one p rob ab leresult would be a reduction in total private in-surance resources available for a single accident.This is bec ause the resulting multiple sources ofliab ility (from millions of hom eo wne r po lic ies) in-creases the perceived risk to the insurers, whoin turn respond by reducing the total amount

available.

Another financial risk of unknown size is thepossibility that workers exposed to radiation may

25-450 0 - 84 - 6 : QL 3




file occupational health suits in future years,based on a statistical link between low-level radia-tion exposures and various diseases, such ascancer, with long periods of development. Insome respects the nuclear power industry is bet-ter p repa red for suc h suits tha n industry ha s b ee nfor comparable suits arising from exposure to

asbestos because detailed records are kept onevery worker’s exposure to radiation. Current-ly, any c ourt settlem ents due to wo rkers’ expo -sure to radiation would be paid out of ANl-MAERP pooI insuranc e. If the size o f suc h set tle-ments becomes at all substantial, however, utili-ties and their insurers may move to establish af ixed compensation program, comparable to thebasic workman’s compensation program, which

Photo credit: Westinghouse

One source of uncertainty about the future costof nuclear power is the possibility that workersexposed to radiation may sue the nuclear plantowners to recover damages from health effects of

radiation exposure

pays fixed sums for each type of injury. Since injury in this c ase is based on a p robab ility link tolevels of radiation exposure the level of prob-ability would be included in the program, muchas has been proposed for compensation for vic-tims of nuc lea r wea po ns testing.

Impact of Risk on the Cost of Capital

As previously noted, from an investor’s pointof view the re a re seve ral rea sons to b e w ary ofutilities with substantial nuclear operations:

q

q

q

since nuclear-generating plants take longerto build, it is more likely that they will be

poor ly matched to ac tual dema nd;the cost of constructing them is harder toestimate and control than it is for coal plants;a ndthere is a small but finite risk of a major

disabling accident. Under current insurancecoverage and PUC rate decisions, a largefraction of the many costs of such an acci-dent would have to be borne by the stock-holder.

A few a ttempts have b een ma de to estimate theimpact of these three elements of risk on the cost

of capital to nuclear-owning utilities but theresults a re no t c lea r-c ut. As of 1981, the h ighe stbond ratings belonged to utilities operating nu-c lea r plants, although the lowe st b ond ratingsbelonged to utilities with nuclear pIants under

c onstruc tion. Afte r Three M ile Island , there wasan imme diate e ffec t on the relative stoc k marketprices of nuc lear and non-nuc lear utility stoc ks.The p rice o f non-nuc lea r stoc k inc rea sed ove r 50c ents a sha re relative to nuc lear stoc k. The e f-fec t p ersisted for a t lea st 2 yea rs (46). Financ ialexperts and utility executives agree, however,that ano ther serious ac c ide nt c ould ha ve very

serious financial consequences.

In the yea r follow ing the Three M ile Island in-

cident, a study of investor attitudes towardsnuclear utilities (1 1,46) showed that investorsranked the risks assoc iated with nuc lear po we r

as a serious problem but less than problemscaused by regulation, high interest rates and in-flation. Twe nty-five p erc ent o f institutiona l inve s-tors said the Three Mile Island ac c ident ha d a neg -ative impact on the weighting of electric uti l i t ies




in their investment portfolios. I n general, port-

folio managers showed increased concern if utili-t ies had a high dependence on nuclear, but over

82 percent said they recommend companies withsome nuclear component in their fuel mix. Itwould seem that investors are more concerned

currently about the risk of construction cost over-runs and delays and the financial strains of plantconstruction than they are about the financial riskof a disab ling ac c ide nt (47). The rec ent indica -tions that several nuclear plants such as Zimmer,Midlands, and Marble Hill may never be completedand licensed to operate has caused anotherround of investor co nce rn. * Betwe en Oc tob er

and late November 1983, stock prices droppedin 36 out of 50 companies with nuclear plantsunder construction.

In summary, utilities assume greater risks whenthey build nuclear pIants than when they build

coal plants. Even if, due to standardization and

c areful c onstruction, the c ost of nuc lea r po werover 30 years is estimated to be substantially lessthan coal-fired electricity, utilities still might hesi-tate to order more nuclear plants unless they arecompensated in some way for the additional risks.

*See Nuc/ear phobia, a research report by Merrill Lynch, Pierce,Fenner & Smith, Dec. 15, 1983.

NUCLEAR POWER IN THE CONTEXT OF UTILITY STRATEGIESThe d ec ision to o rder a nuc lea r po we rp lant is

only one of many choices utility executives canma ke given their c om pa nies’ loa d forec asts andpresent and future financial situations. They couldinstea d order one o r more c oa l plants; co nvertan oil or gas plant to coal; build a transmis-sion line to facilitate purchase of bulk powerfrom Canada or from elsewhere in the UnitedSta tes; d eve lop sma ll hyd roe lec tric source s, w indsources, or other small-scale sources of power;or start a loa d-ma nag eme nt, c og eneration, or

energy conservation program (or some combina-tion of a ll of these).

What utility executives choose depends on thereliab ility o f the ir loa d forec asts, the op tions forretiring oil and natural gas plants, the availabili-ty of reliable sources of purchased power, the na-ture of rate regulation in the State in which theyoperate, and their companies’ abilities to managelarge c onstruction p rojects on the one ha nd o rsuccessful load management and conservationprograms on the other.

From a rec ent survey o f utility exec utives (90)

and the results of two OTA w orkshops, it is c lea rthat utility executives are now considering amuch wider variety of alternatives to construc-tion of ne w large ge nerating plants. Although

utilities do not seem to be avoiding capital invest-

ment at the risk of providing inadequate electricsupplies, none theless they a re ta king financ ialc onsiderations hea vily into a c c ount, espec iallythe ab ility to ea rn a return on CWIP. Som e ex-ecutives say their companies have deliberatepo lic ies of p roviding g enerating ca pa c ity witheither minimum capital cost or short Ieadtimesor bo th.

One possible option is to delay any powerplantconstruction as long as possible and then meetany need for new capacity with combustion tur-bines which c an b e c onstruc ted in 3 to 4 yea rsand cost only $200 to $300/kW (in 1982 dollars).Such a choice is now less risky for future elec-tricity rates because of apparent softening of na-tural gas markets. Combustion turbines cost solittle that they could in theory be written off

quickly and replaced by longer Ieadtime plantsif electricity demand were increasing enough towarrant longer Ieadtime generating capacity withlower fuel costs.

In a recent study six utilities described two alter-

na tive sets of c onstruction p lans: one p lan tha tthey would follow under financially generous ratereg ulation a nd the o ther under financ ially c on-strained rate regu lation (64). The re is a some wh a t




exaggerated difference between the two sets ofcircumstances*–but they do illustrate the rangeof utility choice.

Under financially constrained circumstances,the six utilities expect to rely on more use of pur-c ha sed pow er. One w ill invest in more t ransmis-

sion lines. The utilities also expec t to keep oldplan ts on line an d will defe r the retiring of o il a ndna tural gas p Iants. It is interesting t ha t none ex-

pect to build combustion turbines to catch up todem and growth.**

The preferred ge nerating c hoic e o f the six utili-ties under financially generous circumstances ismedium-sized coal units of 500 to 600 MW whichthe six utilities plan to build in substantial num-bers, over 17 GW by the year 2000. Although sev-eral utilities expect to finish nuclear powerplantsund er co nstruction, only one o f the six expe c tsto start a new nuclear plant. Utility executives atthe OTA w orkshop s and in the survey now per-c eive nuclea r powe r to be too risky to inc ludein future construction projects even under a lessfinancially constrained future. Even for those

utilities that have experience in keeping construc-tion costs under control, there are important

perceived risks from lack of public acceptanceand the possibility of one o r mo re Three M ileIsland types of accidents. Utility executives saidthey expec t to ma ke use of “ co okie-cutter” co alplants because of the greater predictability of theiroperating and construction costs.

It is c onc eivable tha t othe r type s of nuc lea rplants than those currently available in the United

Sta tes wo uld be m ore attrac tive to utility exec u-tives, Executives reported in an EPRI survey of

utility executives’ attitudes towards nuclear pow-er that smaller nuclear plants would be desirablebecause they would require a smaller total capital

commitment and could more easily be fit to un-c ertain load growth (22).

Som e exec utives a lso believe tha t significa ntsafety improvements would make nuclear plants

*For example, the cost of capital is assumed to differ by 300 basispoints between the generously treated case which results in an AAA

bond rating and the constrained case which results in a BBB bondrating.

* * Building a combustion turbine for use more than 1,500 hours

a year is still prohibited under the Fuel Use Act. In practice,however, an increasing number of exemptions are being allowed.

less vulnerable to changes in regulation and ad-verse public reactions and thus more attractive.Several utilities are members of a gas-cooled reac-tor council that supports research and develop-ment o f high tem pe rature ga s-coo led rea c tors(HTGR) (desc rib ed inch . 4). One e xec utive te sti-fied for Florida Power & Light Co. (FP&L) in March

1983 that for FP&L’s crowded site in Dade Coun-ty, an HTGR is the on ly op tion. Sha llow wa terand delicate ecology hinder coal transportationand the c loseness of the City of M iam i and prob-lems with raising the water temperature rule outa light water reactor. FP&L does not want all thepossible hea dac hes of b uild ing the lea d HTGRbut might be willing to build the second plant(91).

If utilities wish to avoid building powerplantsa ltoge ther they ha ve seve ra l op tions (14). Oneof these, featured in several of the six utility strat-

eg ies d esc ribed in tab le 11, is to m a ke b ette r useof existing powerplants. Ironically, the financialweakness and excess capacity that had led utili-ties to cancel new construction has also fosteredneglect of maintenance of existing powerplants.In one recent survey of 80 GW of coal power-

plants there ha d b een a de c line of mo re than 12percent in availability from 1970 to 1981 (1 5).Pennsylvania is one of several States investigatingchanges in regulatory policies that would en-courage more efficient use of existing power-plants (78).

Major investment may be needed to extend thelife of a n existing plant w ell beyond the norma lretirement age of 30 to 35 years. A plant that ha sbeen operating effectively for 40 to 50 years maynot have any of the same components as the orig-

inal plant. Nonetheless, substantial investment toupgrade an existing plant will in almost all casesc ost fa r less tha n b uild ing an ent irely new plantof the same capacity.

Utilities can also substitute programs to reducepeak demand for building new capacity. A recentEPRI survey identified over 200 utilities that wereworking with their customers on conservation

and load management programs (23). Whilesome of the programs were demonstrations,others represented major corporate commitments

to loa d c ontrol. For examp le, the New Eng landElectric System’s successful experiments with on-




Table 11 .—Six Sets of Alternative Utility Construction Plans

ProjectedUtility and type growth in Plan A Plan B(timeframe of plan) load (%/yr.) capital discouraging capital attraction

Utility A. Coal conversion 1.5(1982-2000)

Utility B. Coal/nuclear(1982-2005)

Utility C. Gas/coal(1982-2000)

Utility D. Gas displacement(1982-2001)

Utility E. Oil displacement(1982-2000)

2.0

4.0

3.0

1.5

Utility F. Purchase/coal 3.0

(1982-2001)

Cost $4.1 billionq Sell part of share of nuclear plant Convert 200 MW oil to coal• Reduce sales to outside

customers

Ž Purchase 175MW. Retire no old plantsCost $23.9 billion. Delay two-thirds of a new nuclear

plant for 4 years. Double the amount of purchased

power Consume more oil and gas. No existing plants retired

Cost $64.8 billionq 3,000 MW coal capacity 1982-97q 6,000 MW coal capacity

1998-2009• Reserve margins 13°/0 1990; 9.60/0

in 2000

Cost $8.9 billionq Finish large nuclear plants near

completion. No plants retired. Meet incremental demand

through purchased power

Cost $12.9billionqNo construction projectsq Defer 2,000 MW of natural gas

and oil capacity

Cost $4.8 billion

No constructionq Increase purchased power to36°/0 of total

q Spend $1 biIIion on transmissionlines to wheel in power

Cost $9.9 billion. Finish nuclear plant on schedule Convert oil plant to coal (800 MW)q Build four 600 MW coal plants (two-

thirds ownership)

Cost $40.6 billionq Complete several nuclear plants

without delay. Build four 500 MW coal plants in

1990’sq 75 ‘/0 share i n two 1,100 MW nuclear

units in 2000q All plants retired on scheduIeq Intermediate load coal plantCost $62.8 billionq 5,000 MW new coal capacity in 600

MW increments 1982-97q 5,000 MW more coal 1998-2009. Reserve margin over 20°/0

Cost $30.5billionq 2,500 MW of coal capacity 1988-97

to displace gas. Three large coal plants in

mid-1 990’s to meet additional loadq Oil and gas capacity retired on

scheduleq Finish large nuclear plants near

completionCost $22.1 billion. Purchase share in large coal project

under constructionq Joint owner of coal plant online

early 1990’s. Build transmission lines to

purchase powerCost $7.6 billion

. Four new coal units of 500 MW1988-97q Purchased power shrinks to 7 ‘/0

of total capacity

SOURCE Peter Navarro. Long Term Impacts of Electricity Rate Regulatory Policies for DOE Electricity Policy Project, February 1983

site thermal storage of elec tric hea t a nd home management and energy conservation by theirenergy conservation led it to develop a 15-year customers. As described in the EPRI report andplan aimed at reducing peak demand by over 500 others (14,23,42), these include programs to pro-

MW and average demand by another 300 MW vide rebates for the purchase of energy-efficient

over the 1980-95 period . The p lan is expec ted refrigerators, air-conditioners, or heat pumps, low

to save utility customers about $1.2 billion over interest or interest-free loans and energy index-

that period (65). ing p rograms. For a utility inte rested in conserv-

ing c ap ital, som e utility-c ontrolled loa d ma nag e-Utilities have suc c essfully used a wide va riety ment technologies offer strikingly low capital cost

of techniques to encourage investment in load ($110 to $200/kW). It takes thousands of installa-




tions of such devices in buildings owned by hun-dreds of different customers to equal one 100MW powerplant (see table 12).

Utility executives interviewed for the TheodoreBarry survey cited above report that they are rely-ing more and more on non-powerplant optionsto meet the needs of future growth but they stillhave concern about their long-term effectiveness(90). Without extensive metering of componentsof individual buildings, and extensive data collec-tion on oc c upa nc y and pa tterns of use, it is d if-ficult to d etermine how muc h of a given build-ing’s change in electricity use is due to load man-agement devices and how much is due to otherreasons that may be short-lived or unpredictable.

Utilities seeking to avoid costly capacity addi-tions might a lso wo rk to enc ourag e p ow er pro-duction by cogenerators and other small powerproducers in their service territories. (The poten-tial for cogeneration to reduce electricity demand

was discussed above in the section on electrici-ty demand.)

How ever, current e stimate s of m arket p oten-tial for sma ll pow er prod uc ers are seve ra l timeshigher than estimates of the central generatingca pac ity that could be displaced by small produc-ion. This is bec ause utilities can use small pow erproduc tion to displac e ad ditional c entral station

generating capacity only if it can be counted onto oc cur at times of pe ak de ma nd. A rec ent CRSstudy estimates the total capacity displacementpo tential from c og eneration, wind, and sma ll hy-

droelectric as ranging from about 5 GW to about22 GW, even though the m arket p otential for co-

generation and wind totals about 63 GW and thetechnical potential for small hydroelectric isestima ted a t over 45 GW (14). OTA rec ent ly es-timated the full technical potential for cogenera-tion as even higher, 200 GW (72).

It is wo rthy of no te tha t the estimated ma rketpotential for cogeneration alone of about 42 GWis one m ea sure o f the ma rket for using HTGRsfor c oge neration. As is d isc ussed further inc hap ter 4, HTGRs op erate at far highe r temp er-atures than d o light w ater rea cto rs and ca n beused to supply steam and pressurized hot water.The resulting high e ffic iency o f op eration a nd p ro-duction of both electricity and steam for sale canoffset the ir som ew hat higher ca pita l cost.

Implications for Federal Policies

As long as electric utilities are regulated thereis great potential to influence the strategic choices

they make by adjusting the way expenses and in-vestments a re ha nd led in elec tric ity rate d ete r-mination. This repo rt d oe s not ana lyze a ll thepossible ways in which utility strategies other thancentral station powerplant construction can beinfluenced, but it should be recognized thatFed eral and Sta te regulation c an b e structuredto encourage cogeneration, conservation andload management, and upgrading of central sta-tion po werplants.

Utilities have several reasons to wait beforeorde ring m ore pow erplants. The c urrent highreserve margins are one reason, and uncertain-ty about the future growth of the economy and

Table 12.-Cost and Volume of Various Load Management Devices(controlled by utilities)

Estimated numberof installations

to equal 100 MWreduction in peak Cost/ Approximate

Device demand installation cost/kWWater heater time switch , . . . . 91,000 $130-$240 $118-$218Radio and ripple control 71,000 Radio $95-$108(cycles water heater, air- (water heaters)conditioners) . . . . . . . . . . . . . . . $67-$107

93,000 Ripple $100-$115

(air-conditioners)SOURCE: Table published in OTA study Energy Efficiency of Buildings in Cities; based on John Schaefer, Equipment for

Load Management 1979; and other sources in contract report to the Office of Technology Assessment by TempleBarker & Sloane.



Ch. 3—The Uncertain Financial and Economic Future q 7 5

its impa c t on e lec tric ity de ma nd is anothe r. Al-

though electricity demand forecasting is a trickybusiness at best, there is reason to believe, fromboth a “ top-down” and a “bottom-up” approachto demand forecasting, that electricity demandgrowth will be slower in the 1980’s than it willin the 1990’s.

Each of the utilities, influenced by its State PUCwill make decisions about the proper rate andtype of g enerat ing c ap ac ity to ad d. What theseindividual decisions add up to in terms of a na-

t i ona l e lec t r i c g r id depends on the ba lance

among individual uti l i ty construction decisions.

For example, i f al l uti l i ties choose to minimize

capital requirements and depend on purchase

power arrangements, the national reserve margin

would drop dangerously low and there could be

a scram ble to b uild plants quickly. There c ould

be a short period of unreliable electricity supply.

On the other hand, if all utilities chose to build

long Ieadtime generating capacity to meet fore-c asts of rapid growth in electric ity de ma nd, a nddemand growth failed to increase as forecast, thecurrent high reserve margins might reappear.From a national point of view, it might be bestif the different States and utilities adopted a mix-ture of these approaches.

From a perspective of long-range industrial poli-

c y, there ma y b e reason for the Fed eral Govern-ment to encourage steps that make electricity rateregulatory policies handle inflation be tter, and in-sure stab le elec tric ity pric es ove r the long term

(see box C). Although average electricity rates areforecast to increase smoothly and slowly over thenext one or two decades, this masks a set of off-setting roller coaster rides for individual utilities,that is reflected clearly in the differences amongforecasts of regional electricity rates (mentionedin the section on electricity demand above). Intimes of high inflation, utilities bringing newpowerplants online have rapidly increasing ratesfor several years and then slowly decreasing rates.The increa sing ra te phase ma y disc ourag e the lo-

cation of industries, which might benefit over thelong run from the declining rate phase.

At the moment, if rate regulatory policies acrossthe country were to shift to favoring longer lead-time, capital-intensive technologies, coal-firedgenerat ion wo uld be enco urag ed far more thannuclear powerplants because utility executives

seem to prefer the smaller size, shorter Ieadtimes,lowe r financ ial risk, and greate r public ac c ep t-

an c e o f co a l. The imp lic a tions for the nuc lea r in-dustry are bleak, and read as such by the industry

(87). Only a handful of orders for central stationpowerplants of any kind is likely before 1990.After that, if a modest number of powerplants areneed ed , 10 to 15 GW/yr, c oa l ma y seem ad e-

quate (unless there is a dramatic change in atti-tudes and public policy a bout the impa cts of c oal-burning on acid rain and carbon dioxide buildup,and no significant improvement in coal-burningtechnology).

If the amount of new capacity needed is muchlarger, however (up to 30 GW/yr), utilities maylook to nuclear again as a way of diversifying theirdependence on a single technology (coal). In ad-

dition, now that natural gas shortages appear lesslikely ove r the ne xt de c a d e, it is also p ossible tha tcombustion turbines, particularly high-efficiencycombined cycle plants wi l l seem to be accept-able sources of diversity. For those reluctant tocontinue such reliance very long, or to dependheavily on so-called new technology there couldbe renew ed interest in new nuclear p lants begin-

ning in the 1990’s. By that time if the construc-tion and operating risks of nuclear are significantlybetter than they seem to utilities now, or if alter-na tives, na me ly c oa l, a re sign ifica ntly wo rse, util-ities ma y plac e o rders for more nuc lear plants.In p artic ular, if nuclea r plants c an “ ma tc h the

ma rket mo re” (in the wo rds of o ne O TA wo rk-shop participant) and come in smaller sizes, withshorter Ieadtimes and predictable costs, theymight be a competit ive option again for supply-ing electric-generating capacity.



Ch. 3—The Uncertain Financjal and Economjc Future q 7 9

79,

80.

81.

82.

83 .

84<

85.

Resources for the Future, Inc., F%ce Elasticities of Demand for Energy: Evacuating the Estimates, re-port for the Electric Power Research Institute, prin-

cipal author Douglas Bohi, September 1982.Resources Plann ing A ssoc iate s Co .,The Effects of Three Nuclear Powerplants on Future Electricity Prices, a report for the DOE Electricity Policy Proj-

ect, February 1983.Schmidt, Philip, analysis for forthcoming report,Electricity and Industrial Productivity: A Technical and Economic Perspective, to be published byEPRI in 1984.Sc ience Ap plications Inc., Case Studies of E/ectric

Generating Plant Demographics: Efficiency and Avai/abi/ity Enhancements, for the Depa rtment o fEnergy, January 1983.Searl, Milton, and Starr, Chauncy , “U.S. Generat-ing Capacity Requirements for Economic Growth,

1990-2000,” Pub/ic Uti/ities Fortnight/y, Apr. 29,1982.

Solar Energy Research Institute, A New Prosperi-

ty: Buildinga Sustainable Energy Future, Commit-tee Print of the Com mittee on Energy a nd C om-merce, April 1981.Southwest Energy Associates, Inc., The Financia/Condition of the Electric Utility Industry and Pos-

sibilities for Improvement, a report for the EdisonElectric Institute, March 1982.

86,

87.

88.

89.

90.

Stevenson, J. D., and Thomas, F. A., Se/ected Re- view of Foreign Licensing Practices for Nuclear Powerp/ants, for the Nuclear Regulatory Commis-

sion (N UREG/ CR-2664), Ap ril 1982.S. M. Stoner Corp., Nuclear Supply Infrastructure Viabi/ity Study, for the Argonne National Labora-tory, contract No. 31-109-38-6749, November1982.

Streiter, Sally Hunt, “Trend ing the Rate Base” and“Avoiding the Money Saving Rate Increase,” two

of a three p art series in Pub/ic Uti/ities Fortnight- /y, May 13, 1982, and June 24, 1982.Strom, David, The Potentia/ for Reducing E/ectricUtility Oil Consumption During 1985, u n p u b -lished working draft for OTA rep ort on Tec hnol-

ogies for Response to a Major Oil Disruption, Dec.

6, 1982.Theo do re Barry & Assoc iate s, A Review of Ener-

gySupp/y Decision Issues in the U.S. Electric Uti/i-ty Industry, 1982.

91. Uhrig, Robert, testimony in March 1983 before the

92

House Co mmittee o n Scienc e a nd Tec hnologyand telephone co mmunication with OTA staff, Oc-

tober 1983.United Engineers & Constructors, Inc., “Regula-tory Reform Case Studies,” August 1982.



Tennessee Authority

Boiling water reactor vessel being hoisted into a containment building at Browns Ferry nuclear plant



Contents

Page

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83

Som e Basics in Nuc lea r Powerp lan t Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84

The Safety and Relia b ility o f Ligh t Wate r Rea c tors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87Overview of U.S. Reactors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87Safety Concerns . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87Reliability Concerns . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88Exa mp les of Spec ific Conc erns . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90Lessons Lea rned From Op era ting LWRs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Advanced Light Water Reactor Design Concepts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Hea vy Wa te r Rea c to rs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

The High Temperature Gas-Cooled Reactor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Inherent ly Sa fe Rea c to r Co nc ep ts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

The Sma ll Rea c to r . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

The Sta nd a rdized Rea c to r . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Co nc lusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Ch a p te r 4 Refe renc es . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Tables

Table No.

13. Comparisons of Lifetime Capacity Factors for U.S Reactors . . . . . . . . . . . . . . . . . . . .14. Unresolved Safety Issues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .15. World Comp arison of Rea c tor Typ es . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .16. World Power Rea c tor Performanc e . . . . . . . . . . . . . . . . . . . . , . . . . . . . . . . . . . . . . . .

Figures

Figure No.

19. Pressurized Water Reactor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .20. Boiling Water Reactor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .21. Comparison of Fossil Units (400 MWe and Above) to All Nuclear Units . . . . . . . . .22. Heavy Water Reactor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .23. Schedules for Alternative Reactors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .24. High Tem pe rature Gas-Coo led Rea c tor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .25. Com para tive Fuel Respo nse Time s. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .26, Proc ess-Inherent Ultimately Safe Rea c tor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .27. Nuclea r Rea c tor Availa b ility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

93

94

96

99

102

105

107

107

109

Page

8991

96

97

Page

85 86 89 98 99

100

101

104

106



Chapter 4

Alternative Reactor Systems

INTRODUCTION

Nuclea r pow er in the United Sta tes ac hieved

some remarkable successes in its early years and

experienced dramatic growth in the late 1960’sand early 1970’s. While this rapid growth wasseen as a measure of the success of the technol-ogy, in retrospect it may have been detrimental.As discussed in chapter 5, the size and complexity

of reactors increased rapidly and there was littleopportunity to apply the experience gained fromolde r p lants to the design of newe r one s. In ad -dition, the regulatory framework was incompletewhen ma ny of the plants were d esigne d. As newregulations were formulated, the designs had tobe ad justed retroac tively to ac co mmod ate toc hanging c riteria . With the rush o f c onstruc tion

in the mid-l 970’s, it was difficult to fully integratethese new requirements into the original designs;hence some portions of the reactor designs

emerged as a patchwork of nonintegrated andoften ill-und erstoo d piec es.

Several changes in design requirements havehad far-reaching effects in today’s reactors, eventhough they were not originally expected to havesuch an impact. For example, new criteria on fireprotection in nuclear powerplants have spawnednew features and systems to prevent, contain,

a nd mitiga te fires. This led to grea ter sep a ra tion

of safety systems, changes in cable-tray design,requirements for more fire-resistant materials, andchanges in civil structures to prevent the spreadof fires. Clearly, these modifications can haveramific a tions for other plant systems. Other reg-ulatory actions concerning seismic design, decayhea t rem ova l system s, and p rote c tion of safe ty

systems from other equipment failures have alsohad extensive impacts.

A fresh loo k at the d esign o f light wa ter rea c -tors (LWRs) could be useful if it more fully in-

tegrated the cumulative changes of the past and

r e e xa m in e d th e c r i t e r i a th a t s t imu la te d th o sechanges. In addition, a new design c o u l d i n c o r -porate analytical techniques and knowledge thathave been acquired since the original designs firstwere formulated . In fac t, it c ould be be nefic ial

to investigate designs of alternative reactors thathave different and potentially desirable charac-

teristics. It is possible that a new design could im-prove safety and reliability at an acceptable costand within a rea sona ble timefram e.

This p rovide s the b a sic te c hnica l reason fo r re-evaluating current nuclear technology as em-

bodied in LWRs. It is important to question,however, the justification for actual changes tothe current system. Are there any indications thatthe current generation of LWR is less than ade-qua tely safe or reliab le? The p ublic ap pe ars to b eincreasing ly skept ica l that nuc lear rea c tors are

good neighbors. As discussed in chapter 8, morethan half of those polled expressed the belief thatrea c tors are d ang erous. The sam e p ercenta ge ofthe public opposes the construction of newplants. While this is not an absolute measure of

the adequacy of today’s reactors, it does reflecta growing concern for their safety.

The nuclear utilities also have assessed the cur-rent rea c tors in view of the ir spec ial need s andinterests. While they do not believe that LWRsare seriously flawed, the utilities have expresseda desire for changes that would make plants eas-ier to operate and maintain and less susceptibleto ec onom ically da ma ging ac c idents (1 3). Some

movement has already been init iated within thenuclear industry in response to utility needs. Mostof these efforts focus on evolutionary changes tothe current designs and thus represent normalde velopme nt of LWR tec hnology.

The inc reasing leve ls of c onc ern for sa fetyamong the public and the utilities has contributedto an interest in safety features that are inherentto the design of the reactor rather than systemswhich rely on equipment and operators to func-tion p roperly. The em pha sis on inhe rent sa fetyis reflected to some extent in evolutionary designs

fo r LWRs, and to a muc h grea ter deg ree in inno-vative designs of alternative reactors. In this chap-ter, LWRs as well as several proposed alternativeswill be examined and their relative advantagesand disadvantages assessed.

83




SOME BASICS IN NUCLEAR POWERPLANT DESIGN

To assess the sa fety a nd reliab ility of c urrentreactors and compare them with alternative de-signs, it is important to understand the basic prin-c iples involved in g enerating p owe r with nuc le-ar technology. At the center of every nuclear

reactor is the core, which is composed of nuclearfuel. Only a few materials, such as uranium andplutonium, are suitable fuels. When a neutron

strikes an a tom of fue l, it c an be absorbed . Thiscould cause the nucleus of the heavy atom to be-c ome unstab le a nd split into two lighter a tomsknown as fission products. When this occurs, en-ergy in the form of heat is released along withtw o or three neu trons. The neutrons then strikeother atoms of fuel and cause additional fissions.With careful design, the fissioning can be madeto continue in a process known as a chain reac-tion.

A chain reaction can be sustained best in ura-nium fuels if the neutrons are slowed before theystrike the fissiona b le m a te ria ls. This is done bysurrounding the fuel with a material known asa moderator that absorbs some of the energy ofthe neutrons as they are released from the fissionproc ess. Seve ral different ma teria ls a re suitab le

as moderators, including ordinary water, heavywater, * and graphite.

The heat from the fission p roc ess is remo ved

from the core by a continuous stream of fluidc a lled the p rimary c oo lant. The reac tors exam -

ined in this c hap ter use wa ter or helium as thec oolant, although othe r fluids have b een c onsid-ered. The hea t in the co olant c an b e used direc tlyto p rod uc e e lec tric ity by d riving a turbine-gen-erato r, or it ca n b e transferred to a nothe r fluidmed ium a nd then to a turbine-generator. Bothmethods have been used effectively in U.S. nu-clear powerplants.

There a re ma ny p ossib le c om bina tions of fuel,coolant, and moderator that can be used in thede sign of nuc lear rea c tors. There are a dva ntag es

and disadvantages associated with the various

*A molecule of light water is made from one atom of oxygen andtwo atoms of the lightest isotope of hydrogen. By contrast, amolecule of heavy water is made with the isotope of hydrogen called

deuterium, which has twice the mass.

materials, and no single combination hasemerged as being clearly superior to the others.

Several designs have be en d evelope d for pro-duc ing electric ity c om mercially. The mo st c om -mon reactors are known as light water reactors,

which use o rdinary water as bo th c oolant a ndmoderator. LWR fuel is slightly enriched uranium,in which the percentage of f issionable materialhas been increa sed from its na turally oc c urringvalue of 0.7 percent to about 3 or 4 percent. Afterenrichment, the fuel is shaped into ceramicpellets of uranium dioxide and encased in long,thin fuel rods made of a zirconium alloy.

Another commercially feasible reactor is theheavy water reactor (HWR), which is moderatedby hea vy water and co oled b y ordinary water.The fue l in a n HWR is simila r in fo rm and c om-

position to LWR fuel, but it need not be enrichedto sustain a chain reaction. Another design is thehigh tem pe rature g as-co oled rea c tor (HTGR),which uses helium as a coolant and graphite asa mo dera tor. The HTGR c an use uranium a s afuel, but it usua lly is enric hed to a grea ter con-

c entration o f fissionab le ma teria l than found inLWR fue l. The fue l form is ve ry diffe rent fromLWR and HWR fuel, with the uranium shapedinto sma ll coa ted spheres, mixed with grap hiteto form fuel rods, and then inserted into hex-agonal graphite blocks.

In addition to selecting a fuel, moderator, andcoolant, reactor designers also must devise ameans to transfer the heat from the core to theturbines. In the United States, two different steamc yc les ha ve be en de velop ed for LWRs. The pres-surized wa ter rea c tor (PWR) shown in figure 19ma intains its p rima ry co olant und er p ressure sotha t it will not b oil. The hea t from the p rima rysystem is transferred to a secondary circuitthrough a steam generator, and the steam pro-duced there is used to drive a turbine.

The sec ond type of LWR tha t is in co mm ercialuse is the boiling wa ter rea c tor (BWR), shown in

figure 20. It eliminates the secondary coolant cir-c uit found in a PWR. In the BWR, the hea t fromthe core boils the coolant directly, and the steamprod uc ed in the c ore d rives the tu rbine. There



Ch. 4—Alternative Reactor Systems q 85

Figure 19.—Pressurized Water Reactor

Generator

Containment building

SOURCE: “Nuclear Power from Fission Reactors, ” U.S. Department of Energy, March 1982

is no need for a heat exchanger, such as a steamge nerator, or for two c oolant loop s. In ad dition,

since more energy is carried in steam than inwater, the BWR requires less circulation than the

PWR.

The two LWRs desc ribed above c an b e usedto illustrate another crucial part of reactor design.

Sinc e nuc lear rea c tors produc e highly rad ioac tivema teria ls as byp rod uc ts of the fission p rocess, itis essential that the design incorporates enoughsafety features to ensure the health and safety ofthe public. During normal operation, the radio-active materials are safely contained within thefuel rod s and p ose no threat to the p ub lic. The

concern is that during an accident the fuel maybecome overheated to the point that it melts andreleases the fission products that accumulate dur-

ing norma l ope ration.

Sa fety is designed into a nuc lear rea c tor onsevera l leve ls. First, every effo rt is ma de to pre-

vent minor events from developing into majorprob lems. This is ac c om p lished in pa rt by incor-porating inherent features into the design to en-sure stable and responsive operation. For exam-ple, the physics of the core dictates that most

reactors will internally slow down the fission proc-ess in response to high coolant temperatures, andthus da mp en the effec ts of p rob lems in remov-ing heat from the core. Both PWRs and BWRshave be en d esigne d to respo nd in this wa y.

Other features, known as engineered safety sys-tems, ope rate in p arallel with, or as a ba c kup to,the inherent physica l safe ty fea tures. They a re d e-signed to ensure that the chain reaction is inter-rupted promptly if there is a problem in the plantand to remove hea t from the c ore even under

extrem e c irc umsta nc es. This is nec essa ry bec ause

radioactive decay continues to produce heat longafter the rea c tor has be en shut do wn. If de c ayhea t is not remo ved , the c ore ca n overheat tothe p oint of me lting. In the event that the shut-

25-450 0 - 84 - 7 : QL 3




Figure 20.—Boiling Water Reactor

Containment vessel

SOURCE: “Nuclear Power from Fission Reactors,” U S. Department of Energy, March 1982.

nser

down or decay heat removal systems fail, addi-tional safety systems prevent the escape of fissionproducts to the atmosphere.

Rap id interruption of the nuclea r cha in reac -tion is accomplished by inserting control rodswhich contain neutron-absorbing boron into the

c ore. The control system is designed to shut d ow nthe reac tor automatic ally in the event tha t a b-normal c ond itions de velop in the c ore or prima rycoolant system. Even after the chain reaction isinterrupted, however, the coolant must continueto circulate to remove decay heat. If the coolant

pressure drops in a BWR or PWR–indicating that

som e o f the c oo lant has been lost from the p ri-mary system—the core is automatically floodedby an emergency core cooling system (ECCS). Ifthe sec ondary co oling system fa ils in a PWR, anauxiliary feedwater system is designed to takeover. Other backup cooling systems in these

p lants include high- a nd low -pressure injec tionpum ps and sp ray systems. These safe ty systemsare designed to operate automatically, with noreq uireme nt for ac tion b y the plant o pe rato rs.

They a re d ep end ent on human ac tion o nly inso-far as they must be designed, constructed, andma intained to function co rrec tly.

The fina l step in the design fo r the sa fety of anuclear powerplant is to incorporate features that

prevent the release of fission products in the

eve nt o f a fuel-me lting a c c ident. This is doneusing the concept of “defense in depth,” that is,providing successive barriers that radioactive ma-terials must breach before endangering the pub-lic . The ba rriers in LWRs are the fue l cladd ing,the heavy steel of the reactor pressure vessel, andthe thic k c onc rete o f the co ntainment building

that encloses the pressure vessel and other com-po nents in the c oo lant system .

These e xam ples nec essarily ove rsimp lify thec om p lex designs and interac tions of sa fety sys-tems. Many safety systems play a role in the

routine opera tion o f the p lant a s we ll. This sam-p ling serves as bac kground for the subseq uent

discussions of safety features of LWRs and of alter-native designs.



Ch. 4—Alternative Reactor Systems q87

THE SAFETY AND RELIABILITY OF LIGHT WATER REACTORS

Overview of U.S. Reactors

Of the 84 nuclear reactors with operating li-c enses in the United Sta tes tod ay, about two -thirds are pressurized w a te r rea c to rs. They are

offered by three companies—Babcock & WilcoxCo., Combustion Engineering, Inc., and Westing-ho use Elec tric C orp . The rem a ining rea c to rs (w iththe e xcep tion of one HTGR) are boiling w a ter

rea c to rs, sold b y Ge neral Elec tric Co . These fourc om pa nies a ll supply the nuc lear stea m supp lysystem (NSSS), or the nuc lea r-rela ted c om po-nents of the reac tor. The ba lanc e of the p lant co n-sists of suc h item s as the turbine-ge nera to r, theauxiliary feedwater system, the control room, andthe c ontainment b uilding. The b alanc e o f plantdesign typically is supplied by an architect-engi-neering (AE) firm, any one of which might team

up with a vendor to provide a reactor plant thatmeets the needs of a particular utility at a specificsite. So far, no c om plete ly sta ndardized plan tdesign has emerged, although some convergencehas occurred among the designs of each nuclearsteam system vend or. There is still a g rea t d ea lof difference among the designs of similar com-ponents (e.g., steam generators) and system con-

figurat ions. This is no t surp rising c onside ring thevarious combinations of vendors and AE firms

that ha ve be en involved in po werplant de sign.Furthermore, the utilities themselves may custom-

ize a reactor design to meet specific site require-

ments.

Even without the benefits of a standardized de-

sign, the LWRs that have operated in the UnitedSta tes for more than 20 years have ha d g oo d safe -ty and reliab ility rec ords. There ne ver ha s bee nan ac c ide nt involving a ma jor release o f rad io-ac tivity to the environme nt, and the op eratingpe rforma nce , while not spe c tac ular, has be enco mp arab le to tha t of c oa l-fired po werplants. Still,doubts linger about both the safety and reliabili-ty o f these LWRs. This sec tion exa mines the rea -sons for such concerns, including particular fea-

tures of these reactors that contribute to concern.

Safety Concerns

The o c c urrenc e o f seve ra l widely pub lic izedac c idents suc h as tho se a t Three M ile Island andBrowns Ferry nuclear plants have underscored

the p otential for a c ata strop hic ac c ide nt. Theseaccidents shook some of the confidence in ourunderstanding of nuclear reactors. For example,the sc ena rio tha t unfolded a t Three Mile Islandhad not been stressed prior to the accident: it in-volved the loss of coolant through a small leakin a pressure relief valve, whereas safety analysishad previously concentrated on large loss-of-cool-ant a c c idents. Most stud ies of these serious ac -c ide nts have faulted the plant o pe rato rs morethan the reactor hardware (1 O), which indicatestha t LWR designs a re not a s forg iving of huma nerror as they might be.

Sa fety c onc erns a lso a rise b ec ause nuc learpowerplants have encountered hardware mal-functions in virtually every system, including con-trol rods, steam generators, coolant pumps, andfuel rod s. The ma jority o f these ha rdwa re prob -lems have been resolved by retrofits, changesin me thod s of ope ra tion, and red esign. Som eproblems are expected as a new reactor matures,

but ma ny o f the LWR p rob lem s have persisted .Others continue to surface, some because of theintense scrutiny of p lants follow ing the Three MileIsland accident and others because of the aging

of the earlier reactors. Most of the difficultiesprobably have technically feasible solutions, butit is not always clear that they would be cost ef-fective to implement. Meanwhile, the discoveryof ne w p rob lem s and the slow resolution o f old

ones continues to erode confidence in the safe-ty of LWRs.

Confidence in LWRs might be enhanced ifthere was an objective standard for judging thesafety of these plants. As a step in this direction,the Nuclear Regulatory Commission (NRC) hasproposed a set of qualitative and quantitative safe-ty goals for nuclear powerplants on a 2-year trial

ba sis (4). These safet y goa ls w ill provide a me a ns



88 • Nuclear Powe r in an Age of Uncertainty

for answe ring the question , “How sa fe is sa feenough?”

There is a funda menta l problem with spec ify-

ing standards for safety: there is no technique forqua ntifying the safety o f a nuclear po werplant inan objective and unambiguous way. One attemptto d efine nuc lea r sa fety is p rob ab ilistic risk a ssess-

ment (PRA), which outlines sequences of eventsthat could lead to accidents and then assignsprobabilities to each basic event (12). PRA is be-c om ing a useful too l for suc h ta sks as c om pa r-ing certain design options in terms of their safe-ty impact. However, the technique is still in itsinfancy and the results vary widely from one prac-

titioner to the next. The va riations oc c ur bec a usethe users of PRA must put in their own assump-t ions ab out fac tors c ontributing to ac c ide nts andtheir probabilities of occurrence. More researchis required to establish reasonable and standardassumptions and to refine the process of assess-ing risk.

Another important component of safety anal-ysis is the c onseq uenc e of a n a c c ident. Thisdepends on the amount of radioactive materialthat can be released to the environment follow-ing a nuclear reactor accident, otherwise knownas the nuclear source term. Recent findings indi-c ate that the source terms now used in regula-tion and risk analysis may overestimate the mag-nitude of potential fission product releases (5).Only further analysis can tell whether reductionsin the source terms can be fully justified, and, if

so, the magnitude of the appropriate reductionfor ea c h fission prod uct a nd for eac h ac c ide nt

scenario. Modeling and analysis programs arenow being conducted by NRC and by the Elec-tric Pow er Resea rch Institute (EPRI), the Am erica nNuc lear Soc iety, and by the Industry Deg ra d edCore Rulemaking Program. These studies shouldeventually produce realistic estimates of fissionproduct releases, but the task is complex and like-

ly to b e lengthy.

Reliability Concerns

Reliability and safety concerns are closelyrelated, since the same factors that create con-cern about the safety of LWRs also raise ques-tions about their reliability. If a safety system

malfunctions or threatens to do so, the plant must

be shut down for a lengthy and often expensiveperiod of maintenance. On the other hand,c hronic reliability prob lem s a re likely to indica teor contribute to fundamental difficulties that

could reduce safety.

The reliab ility o f LWRs is easily q ua ntifiab le, in

contrast to the difficulties in defining safety. De-tai led data on reactor performance have beenc ollec ted sinc e the b eg inning o f the nuclear era,and they can be analyzed to determine trends.Two m ea sures of performa nc e a re c om mo nly

used —ava ilab ility and c ap ac ity fac tor. The a va il-ab ility is de fined as the p ercenta ge of a t ime pe -riod during w hic h the reac tor was ava ilab le for

operation (whether or not it was actually in serv-ic e). The c ap ac ity fac tor is the ratio o f the ac tualamount of electric generation to the total theo-retical output of the plant during the same timeperiod. Each of these quantities has some draw-backs as a measure of plant reliability: the capaci-ty factor is affected by the demand for electrici-ty and the plant availability is insensitive to thec ap ab ility of the plant to o pe rate at full po wer.Since nuc lea r powe rplants usua lly a re ba se-Ioaded, the capacity factor is generally a bettermeasure of reliability. Both capacity and availa-bility are shown in figure 21 as a function of timefor all years from 1971 through 1980 (’17). To pro-vide a b asis for c om p arison, relia bility rec ords area lso show n for c oa l-fired p lants larger tha n 400megawatts electrical (MWe). It can be seen that

the average availability for the two types of plantshas been nea rly ide ntic al a t a bo ut 69 percent. The

average capacity factor for nuclear plants overthe sam e time p eriod wa s 60 pe rc ent, whic h wa s3 percenta ge points bette r tha n for coa l. Thus,nuclear plants operate reliably enough comparedwith their closest counterparts, even though theaverage performance has not been as outstandingas anticipated by the original nuclear powerplant

designers.

it is instructive to reexamine performance datafor groups of reactors as well as the industry asa w hole. Ca pa c ity fac tors are shown for eac h

reactor type and vendor in table 13 (27). Whencomparing the data on a lifetime or cumulativeba sis, it c an be see n tha t the re a re o nly slight d if-ferences among reactor vendors or types. It also



Ch. 4—Alternative Reactor Systems • 89

Figure 21.—Comparison of Fossil Units (400 MWe and Above) to All Nuclear Units

Equivalent availability Capacity factor100 1 100 I

I I 1 I I I J‘W71 72 73 74 75 76 77 78 79 80 71 72 73 74 75 76 77 78 79 80

Year of operation Year of operation

Year

Equivalent availability 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 Avg.

Fossil 72.2 67.8 73.2 68.3 67.7 67.2 66.7 67.4 68.0 69.5 68.6

Nuclear 71.6 73.7 74.4 64.0 66.7 65.6 69.8 74.3 64.8 64.5 68.9b

Year

Capacity factor 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 Avg.

Fossil 60.0 59.6 62.4 57.7 57.3 69.4 57.4 55.3 56.3 59.5 56.6

Nuclear 58.9 63.1 64.1 53.6 59.4 59.0 65.2 68.3 52.9 59.5 59.8

SOURCE. National Electric Reliability Council, “Ten Year Review 1971-1980 Report on Equipment Availability “

Table 13.—Comparison of Lifetime CapacityFactors for U.S. Reactors

Lifetime capacityfactor

Reactor type:PWR . . . . . . . . . . . . . . . . . . . . . . . 61 .OO/o

BWR . . . . . . . . . . . . . . . . . . . . . . . 58.70/o

Reactor supplier:Westinghouse . . . . . . . . . . . . . . . 62.7%Combustion Engineering . . . . . . 59.70/0Babcock & Wilcox. . . . . . . . . . . . 57.0 ”/0General Electric Co. . . . . . . . . . . 58.7%All plants . . . . . . . . . . . . . . . . . . . 60.2%

SOURCE: A. Weitzberg, et al , “Reliability of Nuclear Power Plant Hardware —Past Performance and Future Trends,” NUS Corp., Jan.15, 1983.

should be noted that these averages can masksubstantial spreads in the performance of in-dividual plants. As discussed in chapter 5, thecumulative capacity factors of the worst plants

are as low as 40 percent while those of the bestare a s high as 80 perce nt.

The ha rdwa re p rob lem s disc ussed a bo ve ha vec ontributed to low ava ilabilities in som e p lants.

These a nd other hardwa re p rob lems have b eenresponsible for lengthy periods of downtime asdiscussed in detail in volume Il. It is concludedthere tha t mo st of the these p rob lems have b eenor soon will be resolved (27).

Despite signs of progress, LWRs still are not

op erating troub le-free. The steam ge nerato rs inseveral plants have de grad ed to the p oint that it

ha s b ee n ne c essary to replac e the m. This repa iris estimated to cost between $60 million and $80million in ad d ition to the c ost of p urc ha sing re-

placement power. Other plants may have to un-




de rtake expensive retrofits or mod ify op erationto mitigate concerns over pressurized thermalshoc k (26).

Another impediment to achieving high availa-bility is the stream of retrofits that has followedthe ac c ide nt a t Three Mile Island . The Three M ileIsland action plan contains about 180 require-me nts for change s in ope ra tional p lants; thesechanges, of course, could not be incorporatedinto the basic powerplant design, but had to beadded to existing systems. This typ e of ret rofit-

ting is seen as a problem by both the nuclear in-dustry as well as its critics since it introduces thepossibility of adverse safety consequences. In fact,in some cases, new requirements might reducera ther than enha nc e sa fety. This c ould ha pp enif unanticipated interactions arise or if there is aninadequate understanding of the system the re-quirement is intended to improve.

The revision in NRC requirem ents fo r seismicrestraints on piping is often cited as an exampleof retrofit p rob lems. The restraints in nuc learpowerplants are designed to preserve the integrity

of pipe by limiting vibrations even if an earth-

quake should occur. Many pIant operators anddesigners complain that these restraints are ex-pensive to install and that they hold the pipes toorigidly to allow for thermal expansion. Further-

more, some critics of the current seismic require-ments feel that piping actually may be moreprone to failure in an overconstrained system.

These c ritics assert tha t to d ay’ s requirem en ts for

seismic restraints result from an attempt to makeit easier to analyze conditions in plants ratherthan from an identifiable need (l).

On the balance, retrofits probably have im-proved the safety of op erating nuc lea r po wer-plants. in fact, one assessment of plants before

an d a fter the Three M ile Island retrofits c onc lud esthat the probabil i ty of an accident has beenred uc ed by a fac tor of 6 in PWRs and by a fac torof 3 in BWRs, with the core melt probability forPWRs now only slightly higher than for BWRs.These imp rovem ents are att ributed p rimarily to

higher reliability of feedwater systems and regu-latory and inspection procedures that reduce theprobability of human error (19).

Examples of Specific Concerns

Since 1978, NRC ha s bee n required by Con-gress to prepare a list of generic reactor problems.This list is revised annua lly to reflec t ne w info r-mation and progress toward resolution. Each timea new safety issue is identified, NRC assesses the

need for imme diate ac tion. In some c ases, ac -tion such as derating (reducing the approvedoperating power) certain reactors, is taken toassure p ublic he a lth and safe ty. in o ther c ases,

an initial review does not identify any immediatethreat to the public, and further research is con-duc ted . Many g eneric safe ty issues have be enresolved and rem oved from NRC’ s list o f signifi-c ant sa fety item s (26).

Tab le 14 summa rizes the 15 mo st imp orta nt un -resolved sa fety issues as determined by NRC in1982. A few of the items on that list will be ex-

amined here as examples of the types of concernsthat remain about LWRs and some of the factorspreventing their resolution.

One of the mo st w ide ly p ublic ized safety p rob -lems is the potential in PWRs for frac ture o f thereactor vessel from pressurized thermal shock.Reactors are designed to be flooded with relative-ly c old w ate r if a loss of c oolant a c c ide nt oc c urs.The sud den te mp era ture d ifferent ial ca uses sur-fac e strains, known as therma l shoc k, on the thickmetal wall of the reactor vessel and imposessevere differential stress through the vessel wall.

While plant designers have understood and ac-counted for this phenomenon for ‘years, theyhave only recently discovered that two other fac-

tors may make the effect more acute than antici-pated. One is that the emergency cooling systemis likely to be actuated following a small-breakac c ide nt (e.g., the one at Three Mile Island ) whenthe rea c to r vessel is still highly p ressurized . Insuch a situation, the stresses due to thermal shockwould be added to those due to internal pressure.

The sec ond fac tor is that the w eld and plate m a-terials in some older reactor vessels are becom-ing brittle from neutron exposure faster than had

be en e xpec ted . Such emb rittlement increases thevulnerability of the vessel to rupture followingpressurized thermal shock. While the possibility



Ch. 4—Alternative Reactor Systems • 91

Table 14.—Unresolved Safety Issues

issue/Descridion issue/DescrilMion

Water hammer: Since 1969 there have been over 150 reportedincidents involving water hammer in BWRs and PWRs. Theincidents have been attributed to rapid condensation ofsteam pockets, steam-driven slugs of water, pump startupwith partially empty lines, and rapid valve motion. Most ofthe damage has been relatively minor.

Steam generator tube integrity: PWR steam generators haveshown evidence of corrosion-induced wastage, cracking,reduction in tube diameter, and vibration-induced fatiguecracks. The primary concern is the capability of degradedtubes to maintain their integrity during normal operationand under accident conditions with adequate safetymargins.

Mark I containment long-term program: During a large-scaletesting program for an advanced BWR containment system,new suppression pool loads associated with a loss ofcoolant accident were identified which had not been ex-plicitly included in the original design of the Mark I con-tainment systems. In addition, experience at operatingplants has identified other loads that should be recon-sidered. The results of a short-term program indicate that,for the most probable loads, the Mark I containment systemwould maintain its integrity and functional capability.

Reactor vessel material toughness: Because the possibilityof pressure vessel failure is remote, no protection is pro-vided against reactor vessel failure in the design of nuclearfacilities. However, as plants accumulate service time,neutron irradiation reduces the material fracture toughnessand initial safety margins. Results from reactor vessel sur-veillance programs indicate that up to 20 operating PWRswill have materials with only marginal toughness after com-paratively short periods of operation.

Fracture toughness of steam generator and reactor coolantpump supports: Questions have been raised as to thepotential for Iamellar tearing and low fracture toughnessof steam generator and reactor coolant pump support ma-terials in the North Anna nuclear powerplants. Since similarmaterials and designs have been used on other plants, thisissue will be reassessed for all PWRs.

Systems interactions in nuclear powerplants: There is some

question regarding the interaction of various plant systems,both as to the supporting roles such systems play and asto the effect one system can have on other systems, par-ticularly with regard to the effect on the redundancy andindependence of safety systems.

Determination of safety relief valve pool dynamic loads andtemperature limits for BWR containment: Operation of BWRprimary system pressure relief valves can result inhydrodynamic loads on the suppression pool retainingstructures or structures located within the pool.

Seismic design criteria: While many conservative factors areincorporated into the seismic design process, certainaspects of it may not be adequately conservative for allplants. Additional analysis is needed to provide assurancethat the health and safety of the public is protected, andif possible, to reduce costly design conservatism.

Containment emergency sump performance: Following a lossof coolant accident in a PWR, water flowing from a breakin the primary system would collect on the floor of con-tainment. During the injection mode, water for core cool-ing and containment spray is drawn from a large supplytank. When the tank water is depleted, a recirculation mode

is established by drawing water from the containment flooror sump. This program addresses the safety issue of theadequacy of the sump and suppression pool in the recir-culation mode.

Station blackout: The loss of A.C. power from both off siteand onsite sources is referred to as a station blackout. Inthe event this occurs, the capability to cool the reactor corewould be dependent on the avail ail it y of systems which donot require A.C. power supplies and the ability to restoreA.C. power in a timely manner. There is a concern that astation blackout may be a relatively high probability y eventand that this event may result in severe core damage.

Shutdown decay heat removal requirements: Many im-provements to the steam generator auxiliary feedwatersystem were required after the accident at Three MileIsland. However, an alternative means of decay heatremoval in PWRs might substantially increase the plants’

capability to deal with a broader spectrum of transients andaccidents and thus reduce the overall risk to the public.Seismic qualification of equipment in operating plants: The

design criteria and methods for the seismic qualificationof equipment in nuclear plants have undergone significantchange. Consequently, the margins of safety provided inexisting equipment to resist seismically induced loads mayvary considerably and must be reassessed.

Safety implications of control systems: It is generally believedthat control system failures are not likely to result in theloss of safety functions which could lead to serious eventsor result in conditions that cannot be handled by safetysystems. However, indepth plant-by-plant studies have notbeen performed to support this belief. The purpose of thisprogram is to define generic criteria that may be used forplant-specific reviews.

Hydrogen control measures and effects of hydrogen burns

on safety equipment: Reactor accidents which result indegraded or melted cores can generate large quantities ofhydrogen and release it to the containment. Experiencegained from the accident at Three Mile Island indicates thatmore specific design provisions for handling large quan-tities of hydrogen releases maybe appropriate, particular-ly for smaller, low-pressure containment designs.

Pressurized thermal shock: Neutron irradiation of reactorpressure vessel weld and plate materials decreases frac-ture toughness. This makes it more likely that, under cer-tain conditions, a crack could grow to a size that mightthreaten vessel integrity,

SOURCE: US. Nuclear Regulatory Commission, “Unresolved Safety Issues Summary,” Aug. 20, 1982,




of a pressure ve ssel fa ilure is pec ulia r to o nly afew older reactors, it is of concern that such apotentially severe condition was not recognizedsooner. Measures to mitigate the problem of pres-surized thermal shock include reducing the neu-tron flux near the outer walls, increasing thetemperature of emergency cooling water, heating

the rea c tor vessel at ve ry high tem peratures toreduce brittleness, and derating the plant (1 5,27).

BWRs are not susceptible to pressurized ther-mal shoc k, but they have b een p lag ued b y aprob lem know n a s intergranular stress c orrosionc rac king. This p rob lem, which involves defec tsin the reactor coolant piping, is now listed by

NRC as resolved, but it continues to be the subject of extensive and costly research programs

throughout the industry. Most of the service pip-ing sensitive to such cracking has been designedout of the later BWRs, but reactors currentlyunder construction will have recirculation loop

p ip ing with som e susc ep tibility to this pheno m-enon (15,27).

Another problem on the list of unresolved safe-ty issues deals with the corrosion or fatigue crack-ing o f stea m g ene rato r tube s (15). This is of c on -cern because these tubes separate the primarycoolant from the secondary system, and there issome question whether degraded tubes will beab le to ma intain their integrity under ac c ide nt

conditions. NRC estimates that steam generatordegradation has accounted for about 23 percentof the non refueling o utag e time in nuc lea r rea c -tors. The c orrosion ha s be en a ttributed to a c om -bination of inappropriate water-chemistry treat-ment and poor quality materials in the steam gen-erato rs. The result has been wa stage , c rac king,reduction in tube diameter, separation of clad-ding from the tube sheet, and deterioration of the

metal plates that support the tubes inside the gen-era tors. The seve rity o f the p rob lem va ries withsteam generator design and water treatmentmethods. Much of the corrosion has been

brought under control, but plant operators con-tinue to inspect their steam generators regularlyand plug the degraded tubes when necessary.Op erato rs of severa l nuc lear units–Surry 1 a nd2 in Virginia and Turkey Point 3 and 4. in Florida—have already had to replace their steam gener-ators. Other units may face expensive and lengthy

ove rhauls in the future. The fa tigue c ra c king ap -pears to result from flow-induced vibrations, andresolution of this problem may require designmodifications.

Two g enera l sa fety issues d ea l w ith unc erta in-ties over the behavior of the complex systemsfound in nuclear powerplants. One concern is

that the interactions among the various systemsare not fully understood and could contribute toan accident under some circumstances. in par-ticular, it is possible that some system interactions

Photo credit: Atomic Industrial Forum

These four steam generators are awaiting installation ina large pressurized water reactor. In this type of reactor,water flows through the core to remove heat and thenthrough narrow tubes in the steam generator to transfer

it to a secondary coolant loop.



Ch. 4—Alternative Reactor Systems q 93

could eliminate the redundancy or independenceof safety systems. Another concern is that the fail-ure of a control system might aggravate the con-sequences of an accident or prevent the operatorfrom taking the p rop er ac tion. These c onc ernsrelate to the ne ed to unde rstand the entire reac -tor and its interrelated systems rather than to any

specific feature of its hardware. Resolving eitherof these concerns probably would require anal-ysis on a plant-specific basis, but NRC is attempt-ing to identify generic criteria. Both of these issues

contribute to a larger question: are LWR designsmore complex than necessary and could a sim-pler, but equally safe, reactor be designed?

Another safety concern is that a nuclear plantma y de velop a serious p rob lem and fa il to shut

down automatically in an incident known as an“ anticipa ted transient w ithout sc ram . ” In such

a situation, the emergency cooling system would

have to remove not only the decay heat, but alsothe heat generated by full-power operation. NRChas removed this item from its list of generic safetyissues, but many critics feel that it continues torepresent a va lid c onc ern. This issue wa s high-lighted by the d isc ove ry at the Sa lem plants in

New Jersey that faulty circuit breakers preventedthe scram systems from activating automatically(16).

The resolution of severa l of the unresolvedissues may require adding new equipment to LWRsin operation or under construction. One of theseitems is the provision for an alternative means of

remo ving de c ay hea t. NRC is examining whethe ran ad ditional system might sub sta ntially inc reasethe ca pa bility of the reac tor to de al with a b road errange of accidents or malfunctions. Another issueis whether the installation of a means to controlhydrogen is required to prevent the accumula-tion of dangerous levels of the gas in the contain-me nt ve ssel. These c onc erns a re an outg row th

of lessons lea rned from th e Three M ile Island a c -cident. Either measure discussed above would bevery costly to retrofit on existing plants but mightbe ea sily de signe d into a new plant.

These exam p les indic a te tha t the re are still un-resolved generic safety questions concerning theLWR, and that the resolution of these issues willinvolve a c omp lex trad eoff be tween c ost a nd

sa fety. While no single issue ha s bee n ide ntifiedas a fatal flaw, neither is there a clear indicationthat current LWR technology is fully mature andstable.

Lessons Learned FromOperating LWRs

The utilities op era ting LWRs ha ve ga inedknowledge of both the strengths and weaknessesof their plants. A recent survey of the utilitiesreflects the concerns about safety and reliabilitymentioned above and gives specific recommen-da tions ab out fea tures that m ight m itiga te thep rob lem s (13). The follow ing recom me nda tionswere made:

q

q

q

q

q

safety and control systems in new LWRs

should be simpler and easier to operate, test,and maintain. Utility personnel expressed

preferences for passive and fail-safe charac-teristics whenever possible;safety systems should be separated fromnonsafety systems and dedicated to singlefunctions. plant operators worry that over-lap ping func tions ma y lea d to a dve rse im-

pacts in a complex accident scenario;the response of existing plants to abnormaloccurrences should be slowed. Because ofthe low inventory of primary and secondaryc oo ling w a ter in current LWRs, the time b e-tween the start of a transient and the onsetof melting in the core can be short. If all safe-

ty equipment works as designed, it is likelythat transients wo uld be brought und er co n-trol quickly, regardless of the cause. If cer-tain important safety components fail, melt-ing could begin after 40 minutes in a PWR.The failure of all safety systems, which is notconsidered a plausible scenario, could causecore melting in less than 1 minute;

containment buildings should be larger toprovide adequate space for maintenance.Current structures do not allow enoughroom to e asily hand le e quipme nt that is be -ing disassembled for maintenance; and

the potential for retrofits on future plantsshould b e reduc ed as muc h as po ssib le byta king a fresh look at the LWR. Suc h a n ef-fort should integrate the retrofits into a new




design rather than piece them on top of anexisting one.

Both NRC’s list of unresolved safety issues andthe survey of utility executives provide concreteexamples of reasons for concern over the safetyand reliability of current plants. Existing LWRshave serious, although resolvable, problems with

important hardware components; the interrelatedsafety and control systems hinder a deep under-

standing of their behavior; complexity has beenincreased by the large numb er of safety-ma n-

dated retrofits; and current reactors are somewhatunforgiving of operator errors. Despite this less-tha n-pe rfec t record o f the LWR, many in the in-dustry are relucta nt to a bando n it. They a rguethat they have made appreciable progress along

the learning c urve that wo uld have to be repea tedwith an alternative reactor concept.

ADVANCED LIGHT WATER REACTOR DESIGN CONCEPTSImp rovem ents c ould be m ad e to the c urrent

generation of LWRs by redesigning the plants toaddress the safety and operability problems out-lined above. Furthermore, the entire design could

be integrated to better incorporate variouschanges made to LWRs in the past decade. If such

a redesign e ffo rt we re suc c essful, LWRs wo uldprobably continue to be the preferred option inthe future. LWRs have the advantage of being fa-miliar, proven designs with a complete infrastruc-ture to support manufacturing, construction, andoperation.

Advanced LWR designs are being developedby both Westinghouse Electric Corp. and General

Electric Co ., a nd they should be av a ilab le to U.S.utilities before any new reactors are ordered.

General Electric is designing a new BWR that isan e volution of the mo st ad vanc ed plants c ur-

rently unde r c onstruction. The ne we r rea c tor hasbeen modified to enhance natural circulation ofthe primary coolant and hence improve passivec oo ling . This increa ses the ab ility of the c oo lantsystem to remove decay heat in the event thatthe ma in c irc ulation system fa ils. The design fu r-ther provides for rapid depressurization of the

primary system so that both low- and high-pres-sure pumps can supply water to the reactor ves-sel. This sa fety fea ture p rovide s a d d ition a lassurance that emergency coolant would beavailable in the event that primary coolant is lost(6).

Westinghouse Elect ric Co rp . is deve lop ing an

a dv anc ed version of its PWR. The c om p any h asundertaken a joint program with MitsubishiHea vy Industries, the Westinghouse lic ensee inJapan. It is expected that the design development

will be c omp leted by 1984, and there a re p lansto initiate verification testing by 1986. Westing-house is reviewing this design effort with NRCso tha t the ad vanc ed PWR c ould b e read ily li-c ensed in the United Sta tes.

The We stingho use d esign foc uses on red uc ingrisks, improving daily operations, anal controllingcosts (8,11). It attempts to reduce economic riskto the owne r and hea lth risk to the p ublic by in-c orpo ra ting seve ral new fea tures. The c oo lantpiping has been reconfigured and the amount ofwater in the core has been increased to reducethe p ossibility tha t a p ipe b rea k could drain theprimary coolant enough to uncover the core. Pro-

tection against other accidents that could uncover

the core has been provided by safeguarding

against valves that could fail in an open position.Additional risk reduction efforts have been fo-cused on the response of plant operators and sys-

tems following an accident, with the goal of re-quiring less op erator ac tion and more a utoma ticresponses.

Improvements in normal plant operations areanother important feature of the Westinghousead vanc ed de sign. The reac tor has be en rede-signed to operate 18 to 24 months on a singleba tc h of fuel, rathe r than the c urrent 12 month

c yc le. The long er fuel c yc le is ma d e p ossible byenlarging the diameter of the reactor and reduc-

ing the p ow er density and neutron flux. This iscombined with a dif ferent way to moderate theneutrons at the b eg inning o f the c ycle so tha tmore plutonium is produced; the extra fuel is

burned a t the e nd of the c ycle whe n the fission-able uranium is depleted. Other efforts have been




ma de to reduce the am ount of t ime required torefuel and inspect the reactor when it is shutdo wn. The c om bined e ffect o f these final two fea -

tures is to increase the availability of the reactor,pe rhap s to a s muc h a s 90 pe rc ent.

Efforts have also been directed toward increas-

ing the reliability of the advanced PWR. For ex-am ple, the stea m ge nerators have b een rede-signed so that there will be fewer stresses on thetubes and less potential for contaminants to col-lect. Other improvements have been made to in-

crease the reliability of the nuclear fuel and itssupport structure.

Operational improvements have been made inthe a rea of oc c upa tiona l radiation expo sure. Thenew plants have been designed for easy accessto reduce worker exposure. In addition, radia-tion shielding has be en a dd ed to a rea s wherehigh radiation fields can be expected. Finally,

ease of ma intenance a nd repair have b een fac -tored into the redesign of the steam generators;this should reduce the large occupational expo-sures that have resulted from steam generatormaintenance in current PWRs.

An a rea that is c losely related to b oth safety a ndoperation is the effort to increase the designma rgins of the advanc ed PWR. The increased

amount of coolant in the core and the reducedcore power density make the reactor less sensitiveto upsets. Furthermore, the physics of the designdictates that the core will be even better at slow-

ing the fission process in response to a rapid risein temperature.

The c ap ital c ost of the ad vanc ed PWR pro-posed by Westinghouse probably would begrea ter tha n tha t o f c urrent PWRs. An e ffort hasbeen made to hold down the capital cost by sim-p lifying fluid systems, ma king mo re use of m ulti-

plexing, and completing at least three-quartersof the design before construction is initiated.Moreover, other features of the design should

compensate for the increased capital costs,resulting in lifetime costs that should be compar-

able to those of today’s LWRs. Significant fuel sav-ings a re expec ted , with red uc tions in bo th ura -nium a nd enric hment requ irem ents. This is dueto the reduction in specific power, the more ef-fec tive use o f plutonium c rea ted within the c ore,




An intermediate step in the redesign of thePWR has been taken by the Central ElectricityGenerating Board (CEGB) in Great Britain (22).CEGB did not red esign t he basic PWR; rathe r, itad de d safety fea tures to the Stand ardized NuclearUnit Pow er Plant System (SNUPPS), whic h w asde velop ed a nd m arketed in the United State s in

the 1970’s. Most of the changes increase the sep-aration and redundancy of safety systems, suchas add ing a steel c onta inment shell to the no r-

mal concrete containment structure, increasing

the number and c ap ac ity of pumps in the emer-gency-cooling system, and increasing the number

of independent diesel generators to provide elec-trica l power if the norma l supp ly is interrup ted .These c hange s are likely to red uc e the p rob ab ilityof a c ore meltdown, and the prob ab ility for therelease of fission products is much lower still.

These sa fety imp rovem ents will no t be inexpe n-sive; the new design is estimated to cost 25 per-c ent mo re tha n the original SNUPPS p lant.

HEAVY WATER REACTORS

If the new designs for LWRs are perceived asbeing less than adequate to ensure safe andreliable operation, it is possible that alternative

designs will become more attractive to utilities

and investors. The HWR is a p ot ent ial ca nd ida tebe c ause it is the o nly other type o f rea c tor thathas been deployed successfully on a commercialsc a le. The HWR has bee n d eve lope d mo st ex-tensively in Ca nad a, whe re the CANDU (Cana dadeuterium uranium) reactors produce all the nu-clear-generated electricity. In addition, HWRshave emerged as competitors with LWRs in sev-eral other nations, including India, Korea, and

Argentina.

The interest in this typ e of reac tor orig inally

de rived from the effec tive wa y in which hea vy

water moderates neutrons, with a resultant in-crease in fuel economy when compared to LWRs.There a re a lso va rious inherent safe ty fea tures ofthe HWR that make it an attractive alternative tothe LWR. in addition, the current generation ofHWRs has compiled an excellent operating rec-ord. In fact, the HWRs in operation worldwidehave the most impressive reliability record of anyc om me rc ial rea c tor type . As show n in tab le 15,HWRs operated at an average capacity factor of71 percent in 1982, which far exceeds the records

of either PWRs or BWRs. Moreover, in 1982, 5of the top 10 best p erforme rs interna tiona lly w ere

HWRs, even tho ugh this type of p lant a c c ountsfor only 5 percent of the total nuclear capacity.Both lifetime and annua l ca pa c ity fac tors for thewo rld ’s best p ow er rea c tors a re show n in tab le16.

Table 15.—World Comparison of Reactor Types(150 MWe and larger)

Average annualNumber of load factor

Type of reactor reactors (percent)Pressurized heavy waterreactor . . . . . . . . . . . . . . 13 71Gas-cooled reactora

(Magnox) . . . . . . . . . . . . 26 57Pressurized (light) waterreactor . . . . . . . . . . . . . . 106 59Boiling water reactor . . 57 60

aTh e natural. uranium gas-cooled reactors referenced here differ Significantlyfrom the HTGR discussed in this report.

SOURCE: “Nuclear Station Achievement 19S3,” Nuclear Engmeerirrg /nterna-fiona/, October 1983,

The design o f an HWR is some wha t simila r to

a PWR in tha t p rima ry c oo lant transfers hea t fromthe core to a secondary coolant system via asteam generator (3,6,21 ). In many other ways, thedesign of an HWR differs significantly from that

of a PWR. As imp lied b y its name , heavy wa teris used to mod erate the neutrons ge nerated inthe c ha in rea c tion. This is a m ore effec tive mo d -erator than ordinary water, and so the core canbe composed of less concentrated fissionable ma-terial than in a PWR. As a result, the HWR canop erate w ith unenric hed or natural uranium. Na-tions that have not developed enrichment tech-nology p erceive this as an ad vanta ge , but in the

United States uranium enrichment is readily avail-ab le. it is likely tha t U.S. utilities wou ld e lec t tooperate HWRs with uranium that is slightly en-riched. On such a fuel cycle, the HWR wouldrequire only 60 percent of the uranium used in



Ch. 4—Alternative Reactor Systems 97

Table 16.—World Power Reactor Performance(150 MWe and larger)

Nominalrating Capacity

Reactor( )gross factor

type Unit MWe (percent)

Annual a

PHWR

PHWRPHWRPHWRBWRPWRPWRBWRPWRPHWR

Cumulative b

PHWRPHWRPWRPHWRGCRPWRPHWR

BWRPWR

Pickering 2

Bruce 3Bruce 4Pickering 4MuehlbergTurkey Point 3Beznau 2GaronaGrafenrheinfeldBruce 1

Bruce 3Bruce 4Beznau 2Pickering 2Hunterston AStade 1Pickering 4

MuehlbergObrigheim

542

826826542336728364460

1299826

826 826 364 542 169 662 542

336 345

96

96 95 91919089888887

88 85 85 83 83 82 81

80 80 aErOm July 1!382 through June 1983.bFrom start of operation through June 1983

SOURCE “Nuclear Stat Ion Achievement 1983, ” Nuclear .Errgmeerlng /nterrra-tlorral, October 1983

an LWR to p rod uc e the sam e a mount o f elec -tricity.

The use o f hea vy wa ter instea d of o rd inarywater as a moderator and coolant provides a fuel-economy advantage, but it also suffers from adisadvantage. Heavy water is expensive, and the

initial inventory of heavy water represents about20 perce nt o f the c apita l co st o f HWRs. Tota l

capital costs of HWRs are probably comparableto those of LWRs, but fuel cycle costs over thelife of the plant should be lower.

Another feature of the HWR that distinguishesit from a PWR is the use of hundreds of smallp ressurized tub es instead of a single la rge p res-sure vessel. The pressure t ub es in an HWR en-close the fuel assemblies and heavy water coolantwhich flow s throug h the t ub es. They are p osi-tioned horizontally in a large unpressurized vessel

known as a calandria, as shown in figure 22. Thecalandria is filled with heavy water, which actsas a mod erator and is kept a t low temp eratureand pressure. The he avy wa ter in the c aland riasurrounds the coolant tubes but is isolated fromthe fuel.

A disadvantage of the pressure tube configura-tion is tha t the thin wa lls of the t ub es restric t thetemperature o f the co olant mo re than the heavysteel pressure vessel of an LWR. Hence, the over-all efficiency of energy conversion is somewhatless in an HWR. Effic ienc y is furthe r limite d inthe HWR by the heat that is deposited in the

heavy water moderator. Current HWRs achievean efficiency of only 29 percent, while LWRs

typically achieve a 33-percent efficiency.

The pressure tube c onfiguration ha s som e a d -vantages in that it separates the moderator fromthe c oo lant. The reac tivity co ntrol device s andsafe ty system s a re loc a ted outside t he prima rycoolant loop and cannot be affected by a loss ofc oolant ac cident. Moreover, the mod erator ac tsas a backup heat sink that could cool the fuel ifthe primary coolant system fails. This reduces thenecessity to develop elaborate systems to provide

emergency core cooling and decay heat removal,which have been a primary concern in LWRs.

The HWR d iffers from the LWR in its method

of refueling. In an LWR, the reactor must be shutdown, the cover of the pressure vessel removed,and the fuel rods exchanged in an operation thatlasts for several weeks. HWRs can utilize onlinerefueling because they use pressure tubes ratherthan a single pressure vessel. This means that thereac tor can c ont inue to op erate w hile d epletedfuel assemblies are removed and replaced withfresh fuel. This me thod of refueling c ontribute s

to the overall availability of the reactor since there

is no ne ed to shut it down. It a lso e nhanc es rapididentif ication and removal of fuel elements thatlea k rad ioa c tive m ate rials into the c ooling wa ter.

As discussed above, the HWR appears to havecertain safety and operational advantages withrespect to the LWR. However, there are otherfactors that make it unlikely to be considered asa viable alternative to the LWR in the UnitedSta tes. As w ith a ll a lterna tives to the LWR, theheavy water reactor suffers from lack of familiarity

and experienc e in the United Sta tes. There areno vend ors in the United Sta tes which o ffer

HWRs, and hence there is no established domes-tic infrastructure to build and service them. Fur-thermore, there are no utilities with HWR ex-

perienc e. These fa c to rs wo uld no t nec essarilyprohibit the introduction of an HWR into theUnited Sta tes, sinc e the ma nufac turing req uire-




Figure 22.—Heavy Water Reactor

Generator

Turbines


ments, design, and operational skills for the HWRand LWR are similar. However, any alternativedesign would have to overcome barriers beforebeing accepted. It would have to be clearly

superior to the LWR, but in some areas that haveproved to be most troublesome for LWRs for in-stance, operational complexity, the HWR mayactually add to the uncertainty.

In addition to lack of familiarity, the HWR of-

fers no c a pital c ost a d va nta ge s to th e LWR. Eventhough lifetime costs may be lower, it is unlikelythat utilities would be willing to assume large

Pump

inment building

capital debts without a clear demonstration of thead-vantages of an alternative reactor.

Another issue relates to uncertainties in thelic ensing proc ess. The HWR is a fully com me r-cial and lice nsab le reac tor in Ca nad a and otherna tions. In the United Sta tes, how eve r, it wouldbe nec essary to m od ify the lic ensing proc ed uresto match a new reactor. It also is likely that designmodifications would have to be made to the

HWR in order to accommodate to the U.S. reg-ulatory structure; there is a spectrum of opinionson how extensive these changes would be (24).Cost and availability uncertainties would be in-




Photo credit: Ontario Hydro

In a heavy water reactor, natural uranium is used as thefuel. The fuel bundles are loaded horizontally, each inits own tube containing pressurized water. The pressuretubes in a heavy water reactor replace the large steel

vessel in light water reactors

THE HIGH TEMPERATURE

The HTGR is c oo led b y helium an d mo de rate d

by grap hite, and , as show n in figu re 24, the en-tire core is housed in a prestressed concrete reac-to r vessel (PCRV) (2,6,25). The rea c to r uses en-

riched uranium along with thorium, which issimilar to nonfissionable uranium in that it can

troduc ed into a system tha t a lrea dy is less p re-d icta b le tha n desired . Initial c apita l c osts might

be inc rea sed by de sign mod ific ations to imp roveefficienc y o r me et string ent seismic requirem ents.The c om me rc ializat ion proc ess c ould b e e x-

tended if significant changes are required in thelicensing and design of this reactor. One possi-

ble schedule for development was developed byEPRI and is shown in figure 23.

In the United States, the HWR is not perceivedas offering enough advantages to abandon LWRtechnology. Nonetheless, the HWR, may becomea source of e lec tricity fo r U.S. ut ilities. It is possi-ble that Canadian reactors operating near the U.S.border might significantly increase the export ofpow er to U.S. grids if the HWRs in Cana da c on-tinue to operate safely and reliably.

Figure 23.—Schedules for Alternative Reactors

1980 1990 2000 2010 2020

Heavy water reactor

1,000 MW prototypeA

1,000 MW commercial

Commercialization uncertainty

High temperature gas-cooled reactor

330 MW demonstration(Fort St, Vrain)

900 MW prototype

1,500 MW commercial

Commercialization uncertainty

commercialization target

SOURCE: “Alternative Nuclear Technologies, ” Electric Power ResearchInstitute, October 1981.

GAS-COOLED REACTOR

be transformed into useful fuel when it is ir-

radiated. Because helium is used instead of wateras a c oolant, the HTGR c an op erate a t a highertemperature and a lower pressure than an LWR.This results in a h ighe r therma l efficienc y for elec -tricity generation than can be achieved with the




Figure 24.—High Temperature Gas-Cooled Reactor

SOURCE: GA Technologies and Office of Technology Assessment.

other alternative reactors discussed in thisc ha pter; it also ma kes the HTGR partic ularly

suited for the cogeneration of electricity and proc-ess steam.

The HTGR ha s seve ral inherent sa fety c ha rac -teristics tha t reduc e its relianc e o n enginee reddev ice s for sa fe rea c to r op era tion. First, the useof helium as a primary coolant offers some ad-vantages. Because helium is noncorrosive in theop erating te mp erature rang e o f the HTGR, itcauses little damage to components. Further-mo re, it is transpa rent to neutrons and rem ains

nonradioac tive a s it c arries hea t from the c ore.The use of g raphite as a mo derato r also ha s som e

advantages. It has a high heat capacity that greatly

reduces the rate of temperature rise following asevere accident, and hence there is less poten-tial for dama ge to t he c ore. As a result, HTGRsdo not require a containment heat removalsystem.

The d esign of the fuel and c ore structure for

the HTGR a lso has inherent sa fe ty fea tu res. Thec ore is c ha rac terized by a low pow er density inthe fuel, only about one-tenth that of a light waterreactor. As discussed above, the core has a largethermal ca pa citance d ue to the presence ofgraphite in the core and support structures. As

a result, temperatures would rise very slowly evenif the flow o f co olant w as interrup ted , The o p er-ators of an HTGR wo uld have a grea t de a l of time

to diagnose and correct an accident situation be-fore the core is damaged. Figure 25 compares the

10-hour margin to fuel fa ilure fo r an HTGR withconventional LWRs, in which the margins aremeasured in minutes.

Ano the r sa fety fea ture o f the HTGR is thePCRV, which contains the entire primary coolantsystem. The PCRV p rovides more shield ing from

the rad ioa c tive m ate rials in the c ore than a steelvessel since the thick concrete naturally attenu-



Ch. 4—Alternative Reactor Systems . 101

Figure 25.—Comparative Fuel Response Times

0.01 —

0.1 1.0

Hours after instantaneous loss-of-cooling

SOURCE: Gas-Cooled Reactor Associates.

ates radiation. in addition, catastrophic failure ofa PCRV is regarded to be much less likely thanfa ilure o f a steel vessel. The vertic a l and c ir-

cumferential steel tendons that are used to keepthe concrete and liner in a state of compressionare isolated from exposure to da ma ging neutronfluxes and extreme temperatures. Furthermore,they are independent of one another and can bereadily inspected; it is extremely unlikely thatma ny of t he t end ons wo uId fail simultaneo usly.

The e ntire p rima ry c oo lant loop , inc luding thestea m g ene ra tors, helium c irc ulators, and othe rauxiliary equipment, is included within cavitiesin the PCRV. The a dv an tag e o f suc h a c onfigura-tion is tha t p ipe b rea ks within the p rimary loo p

cannot result in a rapid loss of coolant. As a result,the only engineered safety system needed to pro-tect the core from overheating in the event themain core cooling fails is a forced circulation,dec ay hea t rem ova l system. In the HTGR, this isprovided by a core auxiliary cooling system,

which is dedicated to decay heat removal andinco rpo rate s three red unda nt c ooling loop s forhigh reliab ility. This is enha nc ed by a PCRV linercooling system that provides an additional heatsink for dec ay hea t.

Because of the high thermal efficiency of thisreactor and the safety features discussed above,the HTGR ha s long bee n c onside red as a possi-

ble a lternative to the LWR for com me rc ial po we rgenera tion. Work on the HTGR wa s initiated inthe United States soon after the LWR was devel-

op ed . The c onc ep t wa s suc c essfully dem on-strated in 1967 when a 40-MWe reactor wasplac ed in c omm ercial ope ration. This prototypeunit, Peach Bottom 1, was constructed by Phila-delphia Electric Co. and was the world’s first

nuclear station to produce steam at 1,000° F. Theplant operated at an average availabil ity of 88 per-

c ent b efore b eing d ec om missioned in 1974 foreconomic reasons.

Research and development (R&D) leading tocommercial-sized systems continued after thePea c h Bottom 1 dem onstration. A c oo pe rativeeffort of Pub lic Servic e C o. of C olorad o, theAtomic Energy Commission, and General AtomicCo. led to the construction of the 330-MWe FortSt. Vrain reac tor in C olorad o. This p lant intro-

duced several advanced features relating to thedesign of the fuel, steam generators, helium cir-c ulato rs, and PCRV. The p lant first g ene rate d

power in 1976 but experienced problems in itsea rly yea rs that c ontribute d to a disap po intingava ilab ility rec ord. How eve r, the ma jority of thesystems in the Fort St. Vrain reac tor have per-formed well. Furthermore, radiation exposureshave been the lowest in the industry, even thoughextensive modifications were made to the pri-ma ry system afte r the plan t sta rted op erating. TheFort St. Vrain experience also demonstrates thatthe HTGR is ma nag ea ble a nd predicta ble, evenunder extreme conditions. In the past 7 years,forced circulation cooling has been interrupted17 time s a t the Fort St. Vra in rea c to r. The op era -tors generally were able to reestablish forced cir-culation within 5 minutes, with the longest inter-ruption lasting 23 minutes. Even with so manyloss of flow incidents, there has been absolutelyno damage to the core or any of the components.

Fort St. Vra in also e xpe rienc ed som e unex-

pected technical difficulties-slow periodic fluc-tua tions in ce rta in co re e xit tem pe ra tures we re

ob served . The fluc tua tions we re a ssoc iated withsma ll mo vem ents of fuel in the c ore, ca used byd ifferent ial therma l expa nsion of fue l b loc ks. Theprob lem w as resolved a t Fort St. Vra in by m a in-taining the spa c ing be twee n fuel reg ions withc ore restra int de vices. The next g ene rat ion ofreactors will avoid such problems by redesigningthe fuel bloc k.

25-450 0 - 84 - 8 : QL 3




Photo credit: Gas-Cooled Reactor Associates (GCRA)

The high temperature gas-cooled reactor useshelium as a coolant instead of water. Heliumoffers some advantages since it is less corrosivethan water and does not become radioactive. Thehelium circulator for the Fort St. Vrain reactor

is shown above

Since 1977, the HTGR prog ram has foc used onthe development and demonstration of a com-me rcia l-sized p lant . The em pha sis is c urren tly on

designing a four-loop , 2240 mega wa tt thermal(MWth) reactor that can generate electricity at

high efficiency or be applied to the cogeneration

of electricity and process steam. This design in-

c orpo ra tes the lessons learned from the o pe ra -tion o f Fort St. Vra in, experienc e from the ea rlierdesign o f com me rcial HTGRs, and information

obtained from cooperative international pro-grams. Key design changes have been made to

simplify the plant, improve its licensability andreliability, correct specific component-designdeficiencies, and increase performance margins(7). The d esign work has bee n sponsored by the

U.S. Dep artme nt o f Ene rgy (DOE).

Another design effort is directed toward devel-

oping a much smaller gas-cooled reactor, knownas the mo dular HTGR (14). This c onc ep t c ap i-

talizes on small size and low power density toproduce a reac tor that ma y be ab le to dissipa te

decay heat with radiative and convective cool-

ing. In other words, it would not require emer-

ge nc y co oling o r de c ay he at remo val systems.Suc h a rea c tor would be inherently safe in thesense that no operator or mechanical actionwould b e required to p revent fuel from melting

after the reac tor is shut d ow n. The design fo r this

type of reactor is still in a preliminary stage, butits potential for walk-away safety may warrant fur-

ther investigation.

In ad dition to the d ome stic effort to d evelop

the HTGR, there a re also inte rnationa l programsto d esign and c onstruct gas-co oled rea c tors. TheFederal Republic of Germany has operated a 15-

MWe prototype plant since 1967 with great suc-cess, achieving an average availability of more

than 85 percent. A 300-MWe demonstrationplant is scheduled for startup in 1984, and work

has bee n initiated on the design of a 500-MWe

HTGR. Jap an also ha s be en involved in the d eve l-

opment of a very high temperature gas-cooled

reac tor for proc ess hea t ap plica tions. The Jap a-nese program is directed at designing a 50-MWth

reactor to begin construction in 1986.

INHERENTLY SAFE REACTOR CONCEPTSincentives for developing a more forgiving reac-

tor arise from several sources. As previously dis-

cussed, the design of current LWRs has devel-

op ed in a p a tc hwo rk fashion, and there are still

a num ber of unresolved sa fety issues. The a c c i-de nt a t Three Mile Island increased the incen tiveto develop a foolproof reactor when it became

clear that LWRs are susceptible to serious ac-




c idents a rising from huma n error. A more fo rgiv-ing reactor design became desirable in terms ofinvestment protection as well as public health and

safety.

The mo dular HTGR d isc ussed above is an ex-

ample of an effort to develop an inherently saferea c tor. Another examp le of a reac tor that a t-tempts to improve safety dramatically is based onLWR technology. [t is known as the Process In-herent Ultimately Safe (PIUS) reactor, and it is be-

ing developed by the Swedish nuclear firm ASEA-

ATOM (6,9,23). The design g oa l is to ensure sa fe-

ty, even if the reactor is subjected to human er-ror, eq uipm ent fa ilure, or natural disaster. In thePIUS reactor, this goal was translated into twoprimary design objectives: first, to ensure safe

shutdown a nd a deq uate dec ay heat removalunder any credible circumstances; and second,to use the construction and operating experience

of current LWRs as a basis for development. Thiseliminates some of the uncertainties associatedwith a new design.

The sa fety g oa l was param ount in the designof the PIUS reac to r c onc ep t. The nuc lear islandwas designed to ensure that the fuel would never

melt, even if equipment failure were to be com-po unded by op erato r error. To a c c om p lish this,two c ond itions ha ve to b e m et . First, it is nec es-

sary to ensure that the core always remains cov-ered with water. In LWRs, engineered safety sys-tems ensure that cooling water is available to thecore. However, confidence in these systems wasshaken by the ac c ident a t Three Mile Island whe nthe fuel was exposed long enoug h to m elt.

The b asic c onfigurat ion o f the PIUS reac tor is

similar in many ways to that of a conventionalPWR. It employs a primary loop of pressurizedwa ter that transfers hea t to a sec ond ary stea mloo p throug h a stea m g ene ra tor. The m ain dif-

ferences between the two reactors are the sizeof t he pressure vessels and the loc a tion o f thep rimary system c om ponents. In a PWR, the fue lis surround ed by wa ter and enc losed in a p res-sure vessel that is slightly larger than the core;

the primary system pump, steam generator, andpiping are located outside the vessel. In the PIUSreactor, the core is located at the bottom of a verylarge p oo l of wa ter. As show n in figu re 26, the

pool and core are enclosed in an imposing con-crete vessel (13 meters in diameter and 35 metershigh) that is reinforced with steel tendons. In thePIUS reactor, the other primary components aresubmerged in the same pool of water that con-ta ins the fuel, and a ll pe netrations to the PCRVare located at the top of the vessel. With such

a configuration, it is impossible for any type oflea k or eq uipm ent ma lfunction to d rain off thecool ing water.

During normal operation, primary coolant is

pumped through the core and primary loop. Atthe bottom of the plenum under the reactor core

there is a large open duct extending down intothe poo l. The po ol wa ter ordinarily is preventedfrom flowing into the core circuit by the dif-ference in density between the hot water withinthe c ore and the c ool po ol water. At the statichot-cold interface in the duct, a honeycomb grid

helps prevent turbulence and mixing between thehot and cold fluids. If for any reason the coretemperature should rise to the point at whichsteam is formed, the pressure balance at the in-terface would be upset, and the pool water wouldautoma tica lly flow upwa rd and flood the c ore.Natural thermal convection through the poolwould provide enough circulation to cool thec ore, and the p ool water would keep the co rec ove red for ab out a we ek. This system relies com -pletely on thermohydraulic principles and is total-ly independent of electrical, mechanical, orhuma n intervention. The princ ip al unc ertainty in

this reactor concept is in the stability of the pres-sure balance between the hot primary circuitwater and the cold berated water.

The PIUS reac tor is designe d to automa tica llyshut d ow n as soo n as the p oo l wa ter be gins toflow through the c ore. This is gua rantee d by

maintaining a high concentration of boron in thepool water; boron is a strong neutron absorber

and a utom at ica lly interrup ts a c ha in rea c tion. Thisfeature of the PIUS reactor is attractive becausethe possibility of a transient without a subsequentreactor scram is virtually eliminated.

If the PIUS reactor proves to be reliable, it mightresolve some of the troublesome problems withLWRs. With inherent me c ha nisms for au toma tic

shutdown, natural convective cooling, and a




Figure 26.—Process-inherent Ultimately Safe Reactor

Upper cinterfaceregion

Pool ofberatedwater

Prestressedconcretereactorvessel

Lower openinterface region

SOURCE: ASEA-ATOM and Office of Technology Assessment

Generator

large heat removal capacity, there do not appearto be any credible mechanisms for uncoveringor overheating the core of the PIUS reactor. How-ever, the changes from the standard LWR design

create new problems and uncertainties. The d e-

sign of the PIUS is different enough from currentLWRs that further development will be requiredfor co mp one nts, ma terials, and proc ed ures. Thedesign and construction of the PCRV will pose

a p rob lem , since it is larger tha n a ny o ther similar

generator

Recirculationpump

R e a c t o rcore

vessel. The steam genera tor in the PIUS reac toralso will require additional development since itwill be o f a different co nfiguration tha n in co n-ventional PWRs. More importantly, maintenancema y be ve ry diffic ult since the c om pone nts will

be subm erged in a p ool of be rate d w ate r. It istherefore essential that all primary system com-

ponents perform reliably and with little main-tenanc e. The subm erged c omp onents and pip-ing pose another problem—since the pool water



Ch. 4–Alternative Reactor Systems • 105

will be ab ou t 380° F less tha n the p rimary systemcoolant, it is necessary to insulate the primaryc oo lant loop from the bo rate d p ool wa ter. Theinsulation for such an application has not yet

been fully developed or tested. Another poten-tial difficulty relates to fuel handling. Nuclear fuelassemblies are removed and exchanged routinely

in current LWRs, but a similar operation in thePIUS rea c tor is c om p lica ted by having to wo rk

at a distanc e of 80 feet.

Another serious concern relates to the thermalhyd raulic response o f the rea c to r. There is c on -siderable uncertainty about the flow patterns inthe lower interface region between the poolwa ter and the p rima ry co olant. If the bo unda ryis not sta ble, norma l ope rat ions c ould b e inter-rupted unnecessarily by the inflow of boratedpool water through the core, shutting the reac-tor down. Computer simulations have been per-

formed by ASEA-ATOM to dete rmine the c ha r-a c teristics of the interfac e reg ion. The TennesseeValley Authority has supplemented this with a

sma ll-sc a le test to ob serve flow p a tte rns. Theuncertainties associated with the liquid-interfaceregion can only be resolved with larger and more

definitive experiments, such as th 3MWm testp lanne d by A SEA-ATOM.

In ma ny respec ts, the PIUS design b uilds on

dem onstrated LWR tec hnolog y. The fuel for thePIUS rea c to r is essent ially the same as fo r LWRs.The PIUS c ore is designed to use burnab le p oi-sons to maintain a constant reactivity throughoutthe fue l cyc le. These ha ve b ee n used extensive-ly in BWRs, and the experienc e is d irec tly ap p li-c ab le to t he PIUS reac to r. The wa ter c hemistryand waste handling systems for the PIUS are alsovery similar to today’s LWRs. Finally, most of thematerials that would be used in the PIUS are iden-tical to those in conventional LWRs. In fact, thetemperature, pressure, and flow conditions in

the PIUS reac to r wou ld b e less severe t ha n inLWRs.

Overall, the PIUS rea c to r rep resents a fa irlydramatic departure from conventional LWRs.

One consequence of this reconfiguration is thatthe economics become far less certain. Becauseof the requirement for a large PCRV, the nuclearisland of the PIUS is likely to be significantly moreexpensive than that of an LWR. Furthermore, the

remaining technical uncertainties relating to com-ponent and materials development could be cost-ly to resolve. The orig ina to rs of the PIUS design

suggest that other plant costs might be reducedby easing or eliminating the safety qualificationsfor the balance of plant systems because reactorsafety would not be dependent on them. I n cur-rent LWRs, safety qualification of secondary andauxilia ry systems co ntributes signific antly to theoverall cost of the plant. If nuclear regulatorsagree to such a reorientation, it is conceivabletha t the ove rall co st o f the PIUS rea c tor wo uldbe comparable to today’s PWRs or BWRs. It

should be noted , howe ver, that ma ny nuclear-grade systems would still be required to remove,process, and return radioactive gases and liquidsfrom the pressure vessel.

Tec hnica l an d ec ono mic unc ertainties are sig-nific ant fa c tors in the d ec ision to de velop a nynew reac tor conc ept. In spite of these unkno wns,

further development of the PIUS reactor might

be warranted due to its po tential safety a dva n-tage s. These ad vantage s might restore pub lic c on-f idence in the safety of nuclear power, but theymust be tested further before any final judgments

can be made.

Regardless of the merits of this particular de-sign, the PIUS concept illustrates that innovativerevisions of the stan da rd LWR d esign c a n e me rgeif designers are not constrained in their thinking.Even if the PIUS c onc ep t itself turns out to ha ve

som e insurmounta b le prob lems, explorato ry re-search might continue concerning the basic con-

cept of inherent safety.

THE SMALL REACTORConsiderable sentiment is often expressed in norm. When the c urrent g eneration of reac tors

favor of reactors that are smaller than the 1,000- was being designed in the late 1960’s, it seemedto 1,300-MWe LWRs that now represent the natural to continue scaling up the size because




larger nuclear units were c hea pe r to b uild pe rkilowatt of capacity. Moreover, utilities were

growing rapidly and needed large increments ofnew p ow er. The situat ion is very different now.Many seem to feel that a carefully designed smallreactor might be easier to understand, more man-ageable to construct, and safer to operate. Al-

though many of these claims seem intuitively con-vincing, they are difficult, if not impossible, tosub sta ntia te . OTA sponsored a sea rch fo r evi-dence that small plants have any advantages overlarge plants in terms of safety, cost, or operabili-ty (20). This sea rch reve a led no firm sta tistica l da tain supp ort of the sma ll rea c tor, although it sum-marized some of the arguments that make it anattractive concept (see vol. II).

Utilities may find small pIants especially appeal-ing tod ay be c ause the y allow mo re flexib ility inp lanning for the tota l load of the utility. In ad d i-

t ion, the consequences of an outage would havea sma ller imp a c t on the o verall grid. Furthermo re,reducing the size of plants would limit the finan-cial exposure of the utility to loss and increaseoverall system reliability. Initially, small plants ap-pear to suffer a disadvantage in unit constructionc osts since they c annot rea lize t he full bene fitsof economies-of-scale. However, more of theplant c ould be fabrica ted in the fac tory rather

tha n c onstructed in the field, and this c ould resultin large cost savings if the market is large enoughto justify investment in new production facilities.Moreover, the construction times for small plants

would probably be much less than for their largercounterparts. Overall, it is not clear that smallplants would necessarily be more expensive than

today’s large ones.

The op erab ility o f d ifferent sized plan ts ma ybec omp ared on the b asis of a vailab ility. As shownin figure 27A & B, the a va ilab ility of sma ll p lants

generally exceeds that of currently operatinglarger plants, although only by about 5 percent.This trend surfac es both w hen ava ilab ilities areplotted as a function of the number of years afterstart of operation (which compares plants of thesame ag e) and as a function of ca lenda r year(which compares plants operating in the same en-vironm ent). These d ifferenc es c ould b e d ue toeither the number or duration of outages atsmaller plants, which indicates that small plants

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

Years after start of commercial operation

B.—As a Function of Calendar Year

676869 70 71 72 73 74 7576 77 78 79 8081

Calendar year

may be easier to control or maintain. However,this comparison is not conclusive since most ofthe sma ller rea c tors were d esigned and built inthe 1960’s and have not b een a ffec ted by a smany design changes as the newer, larger plants.

It is diffic ult to c om p are the sa fety o f sma ll a ndlarge reactors, but one indication is the occur-rence of events that could be precursors to severe

ac c ide nts. It ap pe ars that the freq uenc y of such



Ch. 4—Alternative Reactor Systems . 1 0 7

events is independent of reactor size. However,the initiating events that do occur at small plantsmay be easier to manage than similar events atlarge p lants. This result is based on too sma ll a

sample to be conclusive, but it may warrant fur-ther study. Another safety comparison can bemade on the basis of the consequences of an ac-

c id ent. The w orst-ca se a c c id ent in a sma ll plantwould be less damaging because the fission prod-uc t invento ry is muc h less in a sma ller p lan t. This

effect, however, might be offset by the larger

number of small units needed to comprise the

same generating capacity.

Sma ll rea c tors a re unlikely to b e a ble to c om -pete commercially with their larger counterpartsunless R&D that specifically exploits the poten-t ial for modular, shop fabrication of componentsis sponsored . This wo uld a llow sma ll p lan ts totake full advantage of the increased productivityand quality of work in a factory setting.

THE STANDARDIZED REACTOR

The c onc ep t of a stand ardized rea c tor has be en

widely discussed for years in the industry but hasyet to be c ome a rea lity (18). The a dva ntag es aremany: more mutual learning from experience

am ong rea c tor ope rato rs, grea ter opp ortunity forindepth understanding of one reactor type, andsharing of resources for training operators or

de veloping proce dures. Sinc e muc h of the c on-cern over current reactors centers on their man-agement rather than on their design, the oppor-tunity to concentrate on learning the correct ap-plication of one well-understood design is

appealing.

Utilities and vend ors wo uld b e espec ia lly en-thusiastic about standardized designs if that con-cept were coupled with one-step or streamlined

lic ensing. The simp lifica tion o f the lice nsing p roc -ess might bring concomitant benefits in reducedcapital costs. If the plants were smaller than those

of the current generation, larger numbers of smallplants would be built to meet a given demand,and this would facilitate standardization.

A ma jor ba rrier to design ing a sta ndard p lantis the d iffic ulty in m arketing ide ntica l rea c tors,given the c urrent ind ustry struc ture and reg ula-

tory c lima te in the United Sta tes. There a re ma ny

opportunities for changes in today’s plants, suchas to match a particular site, to meet the needsof a spe c ific utility, and to a c c omm od ate NRCregu lations. In a dd ition , the e xisting institutiona lstructure does not lend itself easily to industry-wide sta nd ardiza tion. There are c urrently fivereactor suppliers and more than a dozen AE firms.While e ac h rea c tor vendo r is moving to wa rd a

single standardized design, balance of plantdesigns by the AEs c on tinue to va ry. It is unlike-ly that a single dominant plant design will ariseout of all combinations of vendors and AEs,which implies that there may not be industrywide

standardization. However, it is possible that a fewprominent combinations of the more successfulreac to r sup p liers and AEs will join forces to p ro-

duc e a mo re mana gea ble numbe r of stand-ardized designs.

CONCLUSIONS

No single reactor concept emerges as clearly shutdow n, dec ay hea t remova l, and fission prod-superior to the others since the preferred design uc t c ontainment are p rovided by simp le, pa ssive

varies with the selection criterion. If safety is of systems which do not d ep end on op erator ac -paramount concern, the reac tors that incorporate t ion or control by mechanica l or electrica l means.ma ny inhe rent sa fety fea tures, suc h a s PIUS or The full-sc a le HTGR is a lso a tt rac tive in te rms of

the mo dular HTGR, are ve ry a ttrac tive. In suc h safety since it provides more time than any of thereac tors, the c rit ica l safety funct ions of reac tor water-coo led c oncep ts for the operator to res-




po nd b efore the c ore ove rhea ts. The rema iningreac tors ap pea r to be roughly comp arable rega rd-ing safe ty fe a tures. The HWR ha s the low est in-

ventory of radioactive materials, and the indepen-

dent mo derato r loo p serves as a passive, alter-native decay heat removal system. In addition,the HWR has c om piled a superb record in Can-

ada. Advanced LWRs incorporate the benefits ac-crued from many years of extensive operationalexperience. Finally, small and/or standardizedreactors may have operational advantagesresulting from a better understanding of and con-trol over their designs.

If the reactors were to be ranked on the basisof reliable operation and easy maintenance, a dif-ferent order results. The adva nc ed LWR is veryattrac tive be c ause these c riteria have hea vily in-fluenc ed its design. Sma ll rea c tors a lso appe arhigh on the list because their size and shop-

fabricated components may facilitate operation,maintenance, and replacements. HWRs rate high

because they have performed well to date, andthey do not require an annual refueling shut-

dow n. The fe w HTGRs tha t are in op era tion ha vehad mixed performance records, but the newestdesign addresses some of the problems that con-

tribute d to p oor reliab ility. One fac tor enhanc -ing overall performance is the ease of mainte-na nc e in an HTGR resulting from inherently lowrad iation levels. There a re ma ny unc ertaintiesassociated with the PIUS concept. It is likely to

pose maintenance problems. It is also possible

tha t the beha vior of the PIUS will be e rra tic innormal transients, thus increasing the difficulty

of operation. In other ways, however, the PIUSc ould be simp ler to ope rate .

Any attempt to rate these reactor concepts onthe basis of economics is very difficult. Experiencewith LWRs indic a tes tha t the pric e o f fac ilities of

the same design can vary by more than a factorof 2, so estimates of costs of less developed reac-

tors a re highly suspec t. Only a few spec ulativec om me nts c an b e m a d e. Sma ll rea c tors suffer ac ap ital-c ost p ena lty due to lost ec onom ies-of-

sc a le, b ut it is p ossible that this c ould b e red uc edby fabricating more components in factories andkeep ing c onstruc tion times short. HWRs are ex-pected to have comparable capital costs, but theirlifetime costs may be lower than those of LWRs

since the HWRs have lower fuel costs. Standard-ization of any of the reactors discussed wouldreduce costs, if the reactors could be licensed and

constructed more quickly. The HTGR appears to

be comparable in cost to LWRs, but there aregreater and different uncertainties associated with

it. It is premature to estimate the cost of a PIUS-

type reactor for several reasons. First, it is still inthe conceptual design phase, so types andamounts of materials cannot be determinedprecisely nor can construction practices andschedules be accurately anticipated. In addition,the PIUS designers are relying on low costs in the

balance of plant to compensate for the highercosts of the nuclear island. It is not clear whetherthe balance of plant systems can be decoupledfrom their safety functions; the regulatory agen-cies obviously will have a ma jor imp ac t on thisdec ision, and henc e the c ost of a PIUS-type p lant.

A final criterion applied to these reactors mightbe the certainty of our knowledge of them. Howp red icta ble will their pe rforma nc e b e? The rank-ing here is almost the reverse of that for safety.Advanced LWRs are clearly superior in terms offamiliarity because they have evolved from plantstha t ha ve o pe rated in the U.S. for mo re tha n 20years. HWRs have also compiled a Icing record,but d esign m od ific ations might have to be ma debefore the reactor could be licensed in the UnitedStates. There is much less experience with HTGRsin the United Sta tes, with o nly a single fa c ility in

op eration. The PIUS c onc ep t lag s far b ehind the

othe r rea c tors in terms of c ertainty since it hasnever been tested on a large scale.

This survey has exam ined ma ny reac tor c on-

cepts and found that none were unambiguouslysuperior in terms of greater safety, increasedreliability, and acceptable cost. Most representa c om promise a mong these fac tors. A few c ouldnot be adequately compared because so manyuncertainties surround the design at this stage.The p resen t lull in nuc lea r orders p rovide s a n op -po rtune time to reduc e the unce rtainties and ex-pand our knowledge of the less well-tested con-

cepts. A demonstration of advanced LWRs maysoo n oc c ur in Japan , and the results should bevaluable input to future decisions on the LWR

concept. If continued, the Department of Ener-gy’ s deve lopm ent prog ram o n HTGRs will co n-



Ch. 4—Alternative Reactor Systems Ž 109

t inue to provide information and experience thatc ould m ake the HTGR a v ia ble a lternative to theLWR. It may also be valuable to examine theoperation of Canada’s HWRs to determine if anyof their experienc e c an be ap plied to U.S. rea c -tors. If c onsiderab le sentiment c ontinues to b eexp ressed in favor of sma ll reac tors, som e initial

design work may be appropriate. Finally, ap reliminary investiga tion o f the PIUS reac torwo uld tea c h us still more ab out a c onc ep t thatis very prom ising. Work on this or a nothe r “ fresh

look” design would require government supportsinc e the existing rea c tor d esigners do not see

a b ig e noug h ma rket to supp ort new resea rc hprograms.

Until the results of future investigations are in,nothing on the horizon a pp ea rs drama tic ally b et-ter than the evolutionary designs of the LWR.There is a large inertia tha t resists a ny mov e a wayfrom the current reactor types, in which so much

time ha s be en spe nt and from w hic h so m uc hexperienc e ha s be en a c c rued . Howe ver, if to-day’s light water reactors continue to be plaguedby op erationa l diffic ulties or inc idents that ra isesafety concerns, more interest can be expectedin alternative reactors.

CHAPTER 4 REFERENCES

1.

2.

3.

4.

5.

6,

7,

8.

9.

10.

Ab el, P., Workshop Disc ussion on Tec hno log ica lChanges, Dec. 8, 1982, U.S. Congress, Office ofTec hno log y A ssessme nt, Jan. 3, 1983.Agnew, H. M., “Gas-Cooled Nuclear Power Reac-

tors, ” Scientific American, vol. 244, No. 6, June1981.Atomic Energy of Ca nada Limited, “ CANDU-The

Facts,” November 1982.Burnham, D., “Safety Goals Set by NuclearPanel,” New York Times, Jan. 11, 1983.

Civiak, R. L., “ Poten tial for Red uc tion in the Pre-dicted Release of Radioactive Material Following

a Seve re Nuclea r Acc ident, ” Cong ressional Re-search Service, Mar. 21, 1983.Fraas, A. P., “ Survey a nd Assessment of the Tec h-nological Options Available to the Nuclear indus-

try in the 1980 to 2000 Period,” Institute for Energy

Analysis, January 1983.

Gas-Cooled Reactor Associates and GeneralAtomic Company, “ High Tem pe rature Gas-Cooled Reactor Steam CycIe/Cogenerat ion Design

and Tec hnology Developme nt Plan for Nuc learSteam Supply System, Volume II Summary,”GCRA82-011, JUIY 1982.

George, R. A. and Paulson, C, K., “A Nuclear Plant

Design for the 1990’s—Mee ting Tom orrow’ sNeed s,” p resented to the Ame rica n Powe r Con-ference, 1983.Hannerz, K., “ Towa rds Intrinsica lly Safe LightWater Reactors,” Institute fo r Energy Analysis,June 14, 1982.Kemeny, J. G., Chairman, “Report of the Presi-dent’s Com mission on the Ac c ident a t Three Mileisland,” Washin~ton, D. C.. October 1979.

11,

12.

13,

Legg ett, W. D., “A dva nce s in Nuclea r Powe r,”presented at the Second Joint Nuclear Engineer-

ing Conference o f the Am erica n Soc iety of Me-chanical Engineers and the American Nuclear So-

ciety, July 26, 1982.Lewis, H., et al., “ Risk Assessment Rev iew GroupReport to the U.S. Nuclear Regulatory Commis-sion,” NUREG/ CR-0400, September 1978.

Martel , L., et al., “ Summa ry of Disc ussions WithUtilities and Resulting Conclusions,” Electric Pow-

er Research Institute, June 1982.14. McDonald , C. F. and Sonn , D. L., “A New Sma ll

15

16

HTGR Power Plant Concept With Inherently SafeFeatures–An Engineering and Economic Chal-lenge,” presented at the American Power Confer-

ence, Apr. 18-20, 1983.MHB Tec hnica l Assoc iates, “ Issues Affecting the

Viability and Acceptability of Nuclear Power

Usage in the United States,” Dec. 28, 1982.“Mounting Reports of Failures in Electrical Break-ers With Undervolta ge Trip,” Nuc/eonics Week,vol. 24, No. 14, Apr. 7, 1983.

17. National Electric Reliability Council GeneratingAva ilab ility Data System, “ Ten Yea r Revie w1971-1980 Report on Equipment Availability.”

18. Office of Tec hno log y Assessme nt, U.S. Co ng ress,Nuclear Powerplant Standardization: Light Water Reactors, OTA-E-134, April 1981.

19. Phung, D. L., “Light Water Reactor Safety Sincethe Three M ile Island Ac c ident,” Institute for En-ergy Analysis, july 1983.

20. Prelewicz, D. A., et al., “Nuclear Powerp!ant

Size,” ENSA, Inc ., Nov. 19, 1982.21. Rob ertson, J. A. L., “ The CANDU Rea c tor System:



170 q Nuclear Power in a n Age of Uncertainty

22

23

24

An Appropriate Technology,” Science, vol. 199,Feb. 10, 1978.Smith, D., “A Reactor Designed for Sizewell,”New Sc ientist, vol . 98, No. 1329, Oct. 28, 1982.Tren, l., “Safety Considerations for Light WaterReactor Nuclear Power Plants: A Swedish Perspec-

tive, ” I n s t i t u t e f o r E n e r g y A n a l y s i s ,ORAU/ lEA-83-7, May 1983.

U.S. Department of Energy, “Heavy Water Reac-tors, Preliminary Safety and Environmental infor-mat ion Document, Vol. 11,” NonproliferationAlternative Systems Assessment Program, January

1980.

25. U.S. Depa rtment of Energ y, “ High Tem pe rature

Gas-Cooled Reactors, Preliminary Safety and En-

vironmental Information Document, Vol. IV, ”Nonproliferation Alternative Systems AssessmentProgram, January 1980.

26. U.S. Nuclear Regulatory Commission, “Unre-

solved Safety issues Summary,” NUREG-0606,VOI. 4, No, 3, Aug. 20, 1982.

27. Weitzberg, A., et al., “Reliability of Nuclear PowerPlant Hardware—Past Performance and FutureTrend s,” NUS Co rpo ration for the Offic e o f Tec h-nology Assessment, NUS-4315, Jan. 15, 1983.



Chapter 5

Management of the Nuclear Enterprise


The control room at a typical nuclear plant



Contents

Page

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113

Variations in Quality of Construction and Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114Construction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. .. . $ . . . . . . . . . . . 114

Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116

Management Challenges in Construction and Operation . . . . . . . . . . . . . . . . . . . . . . . . . 118Technological Factors That Influence Construction and Operation . . . . . . . . . . . . . . . 118

External Factors That Influence Operation and Construction . . . . . . . . . . . . . . . . . . . . 121Internal Factors That Influence Construction and Operation . . . . . . . . . . . . . . . . . . . . . 124

improving the Quality of Nuclear Powerplant Management . . . . . . . . . . . . . . . . . . . . . . 127Technical Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127Institutional Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128

Assessment of Efforts to Improve Quality. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132

New Institutional Arrangements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . , 134A Larger Role for Vendors in Construction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135Service Companies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136Certification of Utilities. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137Privately Owned Regional Nuclear Power Company. . . . . . . . . . . . . . . . . . . . . . . . . . . 138Government-Owned Regional Nuclear Power Authority. . . . . . . . . . . . . . . . . . . . . . . . 138

Chapter 5 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139

Tables

Table No. Page

17. U.S Rea c tor Typ es, Supplie rs, Arch itec t/ Eng inee rs, and Construc tors . . . . . . . . . . . . 11418. Construction Records of Selected U.S Light Water Reactors . . . . . . . . . . . . . . . . . . . 11519. Perfo rma nc e o f Selec ted U.S. LWRs. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117

20. Early Operating Experience of U.S. Commercial Light Water Reactors . . . . . . . . . . . 124

21. Classifica tions for INPO Eva lua tions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132

Figures

Figure No. Page

28. Comparison of Commodity Requirements for Coal and Nuclear Powerplants . . . . . 11929. Comparison of Manpower Requirements for Coal and Nuclear Powerplants . . . . . . 12230. Historical Labor Requirements in the Nuclear Power Industry . . . . . . . . . . . . . . . . . 12331. NRC Regulatory Guidelines Issued From 1970 to 1980 . . . . . . . . . . . . . . . . . . . . . . . 124



Chapter 5

Management of the Nuclear Enterprise

INTRODUCTION

In the previous chapter, alternative reactor

types were reviewed in terms of safety, operabili-

ty, and economics. While l ight water reactors(LWRs) lack some of the inherent safety features

that characterize the alternatives, this compari-

son yielded no compelling reason to abandon

LWR technology in favor of other, reactor types.The excellent pe rformanc e records of som e o f

the p ressurized wa ter rea c tors (PWRs) a nd boil-ing water reactors (BWRs) indicate that LWRs canbe very reliable when properly managed. Largeand c om plex nuclear units c an also b e b uilt with-in budget and on schedule, as proven by somerec ent exam p les. These c ases indica te tha t it ispossible to construct and operate nuclear power-

p lants efficiently and reliab ly.

Unfortunately, not all utilities perform to thesame high standards. Numerous examples of con-struc tion m alprac tic e a nd op erating violationshave surfaced in recent years, and many of theseproblems are serious enough to have safety andfinancial implications. Utilities evidently need todepend on more than government safety require-me nts and the c onservatism of nuc lear designs

to c om pensa te for errors. As the a c c ident a t ThreeMile Island so vividly illustrated, LWRs are notentirely forgiving machines; they are susceptible

to certain combinations of human error and me-c hanica l failure. Althoug h LWRs a re built to ac -c omm od ate to some problems in c onstruc tion,maintenance, and operation, there is a limit to

the extent of malfunctions and operational errorthat c an b e tolerated. The c onstruc tion a nd o pe r-a tion of nuc lear po we rp lants are highly sop his-ticated processes. Because nuclear technology isvery complex and has the potential for accidentswith major financial and safety implications, man-ag eme nt of the nuc lea r enterprise m ust be of a nintensity that is seldom required in other utility

operations, or indeed, in most other commercial

endeavors. Many utilities readily grasped theunique characteristics of nuclear technology anddevoted their best management resources to itsdevelopment. Others, unfortunately, seem to

have misjudged the level of effort required tomanage nuclear power operations successfully.

This is no t surprising, c onside ring the va riab ilityin the nuc lear utility industry. Forty-three utilities

op erate 84* nuclea r po werplants, and 15 ad di-tional utilities are in the process of constructingtheir first nuclear units (40). Among these vari-ous organizations c an be found a wide variety

of management structures and philosophies, ex-perience, commitment, and skill. While utilitiesare not the only organizations that seem to haveunderestimated the difficulties involved with nu-clear powerplants, they must assume the ultimateresponsibility for the safety and financial successof their plants.

The d iversity of t he utility ind ustry ha s no t

created major difficulties in managing nonnuclear

generating plants. Many different organizationalstyles and structures have been used successful-ly to construct and operate fossil fuel stations anddistribution systems. With the advent of nucleartechnology, however, several new questions canbe raised:

q

q

q

Is the tec hnolog y so sensitive to its ma na ge -ment that it is not adequately safe or reliablewhen poorly managed?If so, ca n the qua lity of mana ge ment b e im-proved to a uniformly acceptable level?Alternatively, can the technology be modi-fied so that it is less sensitive to its manage-ment?

Management and quality issues will be ad-

dressed in this chapter by illustrating the sensitivi-ty of nuclear power operations with a few recentexamples, a look at factors that contribute to suchprob lem s, and a review o f current e fforts to en-sure uniformly high levels of performance.

“Includes all plants with operating licenses, even though some(Three Mile Island Units 1 and 2, Dresden 1, and Diablo Canyon1 ) are not currently in operation.

113




VARIATIONS IN QUALITY OF CONSTRUCTION AND OPERATION

The following discussion will address variationsin quality of construction and operation on threelevels:

1. An overview of the nuc lear ind ustry will bepresented to demonstrate that various proj-ects differ significantly from one another.This will provide a qualitative basis for assess-ing the sensitivity of the technology to itsmanagement.

2. Som e o f the mo re suc c essful pIan ts will beexamined to ide ntify the c ond itions unde rwhich it is possible for nuc lear po we rplantsto be constructed and operated to the high-est standa rds of q ua lity.

3. Some less successful plants will be examinedto identify the factors that contribute to poor

management and to understand the cost andsafety implications.

The e xamp les tha t have be en selec ted for dis-cussion are not intended to fully span the rangeof good and bad pract ices; they are, however,useful in illustrating the differences in the waysin which nuclear power has been implementedin recent years.

Construction

The c onstruc tion of nuc lear po we rp lan ts in theUnited Sta tes is fa r from b eing a sta nd a rd izedprocess. As shown in table 17, a utility mustchoose among several reactor types and vendorsand among an even larger selection of architectengineers (AEs) and constructors. Wide differ-ences in design and construction practices canresult from these va rious c om b inations. A utilitycan superimpose additional changes on the basicd esign to c ustom ize its p la nt a c c ording to its spe -cial needs or to accommodate to specific site re-q uirem ents. Suc h fa c tors pa rtially e xplain the va r-iations in construction time and quality discussedbelow.

There a re no simple m ea sures of q ua lity in c on-struction, and no attempt will be made to developcomprehensive measures. But efficiency in con-struc tion c an be pa rtially indexed by c onstruc-

Table 17.—U.S. Reactor Types, Suppliers,Architect/Engineers, and Constructors

Reactor types:Pressurized water reactors (PWR)

Boiling water reactors (BWR)High temperature gas-cooled reactors (HTGR)Reactor suppliers:

Babcock & Wilcox Co. (PWR)Combustion Engineering, Inc. (PWR)General Atomic Co. (HTGR)General Electric Co. (BWR)Westinghouse Electric Corp. (PWR)

Architect engineers and/or constructors:American Electric Power Service Corp. (AE, C)Baldwin (C)Bechtel Power Corp. (AE, C)Brown & Root, Inc. (C)Burns & Roe, Inc. (AE, C)Daniel Construction Co. (C)Ebasco Services, Inc. (AE, C)Fluor Power Services (AE, C)

General Atomic Co. (C)Gibbs & Hill, Inc. (AE, C)Gilbert Associates, Inc. (AE)Kaiser Engineers (C)J.A. Jones Construction Co. (C)Sargent & Lundy Engineers (AE)Stone & Webster Engineering Co. (AE, C)United Engineers & Constructors, Inc., (AE, C)Wedco (a subsidiary of Westinghouse Electric Corp.) (C)

SOURCE: “World List of Nuclear Power Plants,” Nuclear News, February 1983.

tion time a nd c ost, whic h d iffer widely amo ngthe utilities shown in table 18. Only plants begin-ning c omme rcial operat ion a fter the ac cident a tThree M ile Island we re includ ed , so a ll of t hese

units we re a ffected to some de gree b y the reg u-lato ry c hang es that have oc c urred sinc e 1979.

These d ata should be interp rete d with som ec a re. Seve ral of the longe r c onstruction times ma yreflect inordinate licensing delays or a utility’sdecision to delay construction in response to slowgrowth in the demand for power. In addition,som e of the projec tions for very short c onstruc-tion time s ma y be o verly op timistic. It is a lso d if-ficult to make direct comparisons of constructionc osts sinc e they a re b ased on d ifferent a c c ount-ing schemes. Furthermore, both estimates and ac-

tua l expend itures a re rep orted by the utilities in“current dollars. ” Annual expenditures are thensumme d without ac c ounting fo r the time va lueof money, with the total construction costs ex-



Ch. 5—Management of the Nuclear Enterprise q 115

Table 18.—Construction Records of Selected U.S. Light Water Reactors

Construction Year oftime a commercial operation Cost c (actual or

Plant (years) (actual or expected) expected, $/kWe)

Shortest construction times: St. Lucie 2 . . . . . . . . . . . . . . . . 6 1983 1,800Hatch 2 . . . . . . . . . . . . . . . . . . 7 1979 607Arkansas Nuclear One 2 . . . . 7 1980 308

Perry 1 . . . . . . . . . . . . . . . . . . . 8 1985 2,200Palo Verde 1 . . . . . . . . . . . . . . 8 1984 1,900Byron 1 . . . . . . . . . . . . . . . . . . 8 1984 1,500Callaway 1 . . . . . . . . . . . . . . . 8 1984 2,500

Longest construction t/roes: Diablo Canyon 1. . . . . . . . . . 16 1984 1,700Diablo Canyon 2 . . . . . . . . . . . 14 1984 1,700Salem 2 . . . . . . . . . . . . . . . . . . 13 1981 704Zimmer 1 . . . . . . . . . . . . . . . . . 13 1985 2,400Midland 1 . . . . . . . . . . . . . . . . 13 1985 2,700Seq uo ya h 2 . . . . . . . . . . . . . . . 12 1982 740Watts Bar 1 . . . . . . . . . . . . . . . 12 1984 1,500

aeased on construction permtt to commercial operation.b“World List Of Nuclear Power Plants,” Nuclear News, February 1983, and other recent updates.‘Komanoff Energy Associates, 1983, expressed In mixed current dollars.

pressed in terms of “ mixed c urrent d ollars.” Thisaccounting system tends to further distort actualcosts.

It is interesting to note that the best and worstconstruction schedules from table 18 differ by anaverage of about 6 years. In fact, the plants withthe longest construction times took twice as longto c om plete as those w ith the shortest sc hed ules.Dramatic differences also can be observed in thecosts, even when the construction schedules aresimilar. For example, the Callaway 1 unit is pro-

jected to cost $2,500 per kilowatts electrical

(kWe) after 8 years of construction, while theByron 1 plant is projected to cost only 60 per-cent of that with the same construction schedule.

A recent study by the Electric Power ResearchInstitute (EPRI) a tte mp ts to identify the reasonsfor the variations noted here (3). In a statisticalanalysis of all nuclear powerplants, it was foundthat 50 to 70 percent of the variation in Ieadtimecould be accounted for by regulatory differences,deliberate delays, and variations in physical plantcharacteristics, EPRI ascribed the remaining var-iation to management practices and uncontrol-

lab le eve nts. To m ore fully und ersta nd the im-portance of uti l i ty management in the construc-tion phase, it is valuable to examine a few specificexamples.

Two of the mo re nota ble nuc lea r po werplantconstruction projects are Florida Power & LightCo .’s St. Luc ie 2 unit a t Hutc hinson, Fla. a nd thePalo Verde 1 plant at Wintersburg, Ariz. ownedby Arizona Pub lic Service Co . As shown in tab le18, both units are projected to be completed withrelatively short construction schedules. Neitherutility has encountered significant regulatory dif-ficulty nor much opposition from interveners (6).Both units had to accommodate to the wave ofbackfit and redesign requirements of the NuclearRegulatory Commission (NRC) that followed theac c ident a t Three Mile Island , and yet no signif-

icant delays have been experienced at either site.These exam p les indica te tha t nuc lear pow er-plants can be constructed expeditiously, even inthe most difficult regulatory environment.

In contrast to these examples, other plants havehad a long history of problems. Quality controlin nuc lear powerp lants, as in othe r c om merc ialendeavors, is important in assuring consistencyand reliability. In industries such as nuclear powerand aerospace, where the consequences of fail-ure ca n be seve re, qua lity is gua rantee d b y su-perimposing a formalized, independent audit

structure on top of c onventional quality controlmeasures in d esign, p roc ureme nt, ma nufac turing,and construction (30). Deficiencies in the qualitycontrol procedures at nuclear reactors are cause




\ . ..

.

Photo credit: Atomic industrial Forum

Florida Power & Light used experience gained inbuilding St. Lucie 1 to construct St. Lucie 2 in the

record time of 68 months

for serious concern because they may make itimpo ssible to verify tha t the plants a re sa fe.

Deficiencies in the quality assurance programat the two Diablo Canyon nuclear powerplantswe re unc ove red in 1981 (18). These d eficienc ieshad gone undete cted by NRC and the p lantowner, Pacific Gas & Electric Co. (PG&E), foryea rs. They d id no t surfac e until a fter NRC ha d

granted PG&E a preliminary lic ense to op erateone o f the reac tors at low po wer. Since the prob-lems have surfaced, a number of errors in seismicdesign have been identified, and it is not yet cer-tain that the plants will be able to withstand adesign basis earthquake. Diablo Canyon’s licensehas been suspended and will not be reinstateduntil NRC can be convinced that the safety equip-ment provides adequate protection to the public

and that the quality assurance weaknesses havebe en c orrec ted . (The Diablo C anyon plants arediscussed further in ch. 8.)

Other ma nag eme nt c ontrol failures have re-sulted in lengthy c onstruc tion delays. A rec entexample is the 97-percent-complete Zimmer 1plant owned by Cincinnati Gas & Electric Co., inwhich alleged deficiencies in construction prac-

tices ha ve led to a n NRC stop-work order. The

NRC has uncovered a number of problems at theplant resulting from wha t it ca lls a “ wide sp rea dbreakdown of Cincinnati Gas and Electric’s man-ag ement . . . “ (20). The Zimme r plant ha s bee ntroubled for years by poor construction practicesand an inadequate quality assurance program. *

NRC cited deficiencies in 70 percent of the struc-tural steel welds, inadequate documentation andqualification of welders and quality assurancepersonnel, unauthorized alteration of records,and inadequate documentation of quality for ma-terials in safety-related components (4).

The exam p les d isc ussed above rep resent theextremes of good and bad construction experi-enc es. They indic ate that nuclear po werplantsc an be c onstructe d with varying e mp hasis onquality, and that such differences in management

approach result in noticeable differences in the

plants.

Operation

As with construction, it is difficult to identifyspecific measures of safety or quality in plantop era tions. There is, how ever, an intuitive c or-relation between safe and reliable plant opera-tion. Two pa ra llel argume nts for this c onne c tionc an be m ad e. First, a safe plant is one which is

constructed, maintained, and operated to highstandards of excellence. Such a plant is also likelyto be a reliable performer. Conversely, a reliable

plant that operates with few forced outages is lesslikely to ta x its sa fety-relate d eq uipme nt b y fre-quent c yc ling . Som e c aution must be used ineq ua ting sa fety with reliab ility; it is possible tha ta plant could be operated outside of its most con-servative safety margins in the interest of increas-ing its capacity factor. But in general, good plantavailabilities (or capacity factors for base-loadedplants such as nuclear units) are reasonably goodindica tors of w ell-run p lants. The a verag e c umu-lative c apac ity fa c tor of a ll U.S. rea c tors is c ur-rently 59 pe rc ent. This me a ns tha t a ll of the rea c -tors in the United States have operated an aver-

age of 59 percent of their design potential

*On Jan. 20, 1984, Cincinnati Gas & Electric announced that Zim-mer would be converted to a coal-fired facility.




throughout their lifetimes. As discussed in chapter4, this is c omp arab le to the c ap ac ity fac tors of

base-loaded generating coal units.

The a verage d ata c onc ea l the more interestingvariations in individual plants. A number of LWRshave o perated for years at muc h lowe r ca pa cityfactors, while some plants have consistently ex-c eed ed the a verage by w ide ma rgins. Tab le 19,which lists lifetime c apa c ity fac tors for the be stand wo rst p lants in the United Sta tes, illustrate sthe wide range of performance that can be foundamong reactors of comparable design, Note that

the best plants have lifetime load factors that are20 points greater than the 59-percent average ca-

pa c ity fac tor disc ussed ab ove, w hile the p oorestpe rformers ha ve c apac ity fac tors 10 points less

than the industrywide a verage. The m ana ge mentof maintenance and operations is one of severalfac tors tha t cont ributes to the se d ifferenc es.

The d ata in tab le 19 sugg est an imp ortant p oint:no single external characteristic can be identified

that unambiguously distinguishes between goodand poo r pe rformers. The lists of best a nd wo rstplants each contain both PWRs and BWRs, smalland large rea c tors, new a nd o ld units, and u til-

ities with previous nuclear experience as well asthose with only a single plant. Although there aremore PWRs than BWRs in both lists, this merelyreflec ts the fac t that the re a re nea rly twic e a smany PWRs as BWRs in operation. It should be

noted that although size does not appear to bea dominant characteristic of either good or poor

performers, there is a tendency for the best per-forme rs to be sma ller than the ir less suc c essfulcounterparts. While 20 percent of all mature reac-tors a re large r tha n 1,000 meg awa tts elec tric a l(MWe), 4 of the 10 plants with the worst capaci-ty fac tors a re large r than 1,000 MWe; only 1 p lantof this size is in the list of the best performers.

Three o f the be st p lants show n in ta ble 19 havebeen in operation for more than a decade–PointBeach 2, Connecticut Yankee, and Vermont Yan-kee. It is c lea r from the p erforma nc e rec ords of

these units that a nuclear powerplant can be avery reliable source of electricity over manyyears. Other less fortunate plants have experi-enced considerable operating difficulties, as in-dicated by the worst performers listed in table19. Four of these plants have operated at less than

50-percent capacity factor throughout their l ife-time.

Table 19.—Performance of Selected U.S. LWRs

Lifetimecapacity

Plant factor

Best capacity factorsb:Point Beach 2 . . . . . . . . 79Connecticut Yankee . . . 76Kewaunee . . . . . . . . . . . 76Prairie Island 2 . . . . . . . 76Calvert Cliffs 2. . . . . . . . 75St. Lucie 1 . . . . . . . . . . . 74Prairie Island 1 . . . . . . . 71Monticello . . . . . . . . . . . 71Vermont Yankee . . . . . . 70Calvert Cliffs 1 . . . . . . . . 70

Worst capacity factors b : Beaver Valley 1 . . . . . . . 34Palisades . . . . . . . . . . . . 39Davis Besse 1 . . . . . . . . 40Brunswick 2 . . . . . . . . . . 41Salem 1 . . . . . . . . . . . . . . 46Indian Point 3 . . . . . . . . 46Brunswick 1 . . . . . . . . . . 48Rancho Seco . . . . . . . . . 50

Duane Arnold. . . . . . . . . 51

Designcapacity

(MWe)

497582535530845830530545514845

852 805 906 821

1090965821918538

Number of reactorsType of Years of in operation byreactor oper ati on a same utility

PWRPWRPWRPWRPWRPWRPWRBWRBWRPWR

PWRPWRPWRBWRPWRPWRBWRPWR

BWR

111598669

1210

8

711576668

8

2 1

1

3 2 3 3 3 1

2

1

2 1

3 2 2

3 1

1

aBy the end of January 1983.blncludes only plants greater than 100 MWe in operation 3 years or 10n9er.

SOURCE “Licensed Operating Reactors, Status Summary Report, data as of 01-03 -83,” U S Nuclear RegulatoryCommlsston, February 1983

25-450 0 - 84 - 9 : QL 3




Operating difficulties also can be inferred fromNRC’s system of fines and enforcement actions.NRC recently proposed several fines for allegedsafety violations which it claims resulted frombrea kdowns in mana ge me nt c ontrols. The largestof these penalties is a proposal for an $850,000fine to be collected from Public Service Electric

& Ga s Co . for p rob lems a t its Sa lem 1 nuc lea rpowerplant in New Jersey (21). On two occasionsin February 1983, the circuit breakers in the reac-tor’s automatic shutdown system failed to operateas designed to shut the reac tor down sa fely. Inboth cases, the plant operators initiated a manualshutdown and avoided damage to the plant.While the problem can be partially attributed toa design flaw in the shutdown equipment, itmight have been avoided if the a utoma tic shut-dow n equipment had bee n properly maintained

(5). NRC b ased its fine o n evide nc e o f lax man-agement, deficiencies in the training of staff, andinattention to certain safety procedures.

NRC a lso ha s proposed stiff fines a t o the r util-ities for difficulties related to managementcontrols. Carolina Power & Light Co. has paid

$600,000 because it failed to develop certain pro-

cedures and conduct tests at its Brunswick pIantsin North Carolina, and because its quality-assur-anc e staff failed to d etec t the p rob lem. Boston

Edison Co. was fined $550,000 for managementprob lems a t its Pilg rim 1 p lant. These a nd otherexamples demonstrate that there are serious man-agement difficulties in some operational plantsand that poor management can have importantsafety and economic implications.

MANAGEMENT CHALLENGES INCONSTRUCTION AND OPERATION

The nuc lea r utilities have identified a numb erof o bstac les to m aintaining qua lity in the c on-struction and operation of nuclear powerplants;othe r orga niza tions suc h a s NUS Co rp. a nd EPRIalso have investigated the difficulties involved innuclear operations (1 7,38). For discussion pur-

poses, the factors that adversely affect nuclearpower operations can be categorized accordingto the ir sources. Som e prob lems arise from the

nature of the tec hnology itself; others c an be at-tributed to the external conditions that influenceall utilities. While these factors affect the manage-ment of all nuclear powerplants, there appearsto be a great deal of variation in the ability of util-ities to c op e w ith them. A third source of p rob -lems arises from the utility management itself.

Technological Factors That InfluenceConstruction and Operation

As presently utilized, nuclear technology ismuch m ore complex than other conventional

generating sources; this creates difficulties in con-struction and operation beyond those experi-enced in fossil units. Most of the unique charac-teristics of nuclear powerplants arise from the fis-


These workers at the Connecticut Yankee nuclearplant are making underwater adjustments to equipmentin the reactor vessel. While cumbersome, submersionof the equipment is a protection against radiation

sion p rocess and effo rts to susta in, c ontrol, andmonitor it during normal operation. Nuclear reac-

tors have other unique features to contain radio-a c tive ma terial p roduc ed as a result of the fissionproc ess and to p rotec t the work forc e a nd thepublic. Finally, and most importantly, nuclear




po werplants are eq uipp ed with many levels ofsafety equipment that prevent the release of ra-dioactive material in the eventofanaccident(19).

Ove ra ll, nuc lear p ow erplants are mo re sop histi-c ated tha n fo ssil-fuel sta tions. This ha s ob viousimplications for the construction of nuclear units.As shown in figure 28, nuclear powerplants re-

quire greater quantities of most of the major con-struction materials than coal plants (34). Certaincomponents are also more numerous in nuclearplants than in fossil units. For example, a largenuc lea r plant m ay have 40,000 valves, while a

fossil plant may have only 4,000 (1 7). Anothercase in which requirements for nuclear reactorsexceed those of coal plants is in the area of pipingsupports. Bec ause nuc lear p lants must b e a b leto withstand an earthquake, they are equippedwith complex systems of piping supports andhang ers de signed to a bsorb shoc k without d am -aging the pipes. A typical nuclear unit might con-

tain tens of thousands of elaborate pipe hangers,supports, and restraints that must be designedand installed according to highly specific criteria.In contrast, a comparable coal plant might con-ta in o nly ab out 5,000 pipe sup po rts who se d esignand installation is not subject to the same restric-tive standards found in nuclear powerplants (35).

In view of this, it is not surprising that the con-struction of a nuclear reactor is considerablymore labo r intensive than that of a c oa l plant. Asshow n in figure 28 a ne w nuc lear unit m igh t re-quire 64 percent more workhours than a similarlysized coal unit.

The d esign effort for nuc lear reac tors also b e-comes increasingly difficult as the plants becomemore co mplex. A particularly challenging aspectof nuclear powerplant design is anticipating po-tential interaction among systems or unexpectedfailure modes within a single system. As discussedin cha p ter 4, it is of vital c onc ern to e nsure tha t

Figure 28.—Comparison of Commodity Requirements for Coal and Nuclear Powerplants

Concrete

Piping

Reinforcing steel Coal

and structural steelNuclear

Coal

Nuclear

Coal

Nuclear

Coal

Wire and cable

Nuclear

CoalLabor

Nuclear

SOURCE. United Engineers and Constructors, Energy Economic Data Base, September 1981 (Based on 1,139 MWe PWR and 1,240 MWe high. sulfur coal plant)




the smooth operation of safety systems is not im-paired by malfunctions in unrelated and less crit-ica l area s. The risk of suc h a dverse inte rac tionhas increased with the steady growth in the num-ber and complexity of nuclear plant components.In fac t, it is not a lwa ys ea sy to p red ict the ove rallimpact of changes that superficially seem to con-

tribute to safety. For example, NRC requestedone operating utility to install additional pumpsto reduce the risk of a Three Mile Island-type ac-cident. An extensive analysis of the system usingprobabilistic risk assessment indicated that adding

the extra pumps would not necessarily reducethe risk. It w as disc ove red tha t the loc a tion, notthe numb er, of pum ps wa s the key to enha nc -ing safety (8).

Finally, the operation of a nuclear plant be-comes more difficult as the plant increases incomplexity. Many of the routine actions that mustbe taken to control the reactor during normal op-

eration or to shut it do wn d uring a n ac c ide nt areha nd led a utom atic ally. There are, howe ver, un-usual combinations of events that could produceproblems with these automatic safety systems andwhich cannot be precisely predicted. For this rea-son, nuclear reactor operators are trained to re-spond to unusua l situa tions in the p lan t. This isnot extraordinarily difficult in very simple, small

rea c tors, suc h a s resea rc h o r test rea c tors, whic hcan be designed with a great deal of inherent safe-ty and operated with less sophisticated control

systems. In today’s central power stations, how-

ever, there are many complex systems that havethe potential to interact, making it difficult forop erators to respo nd c orrec tly and rap idly to a b-normal situations. Furthermore, if the controlroom design is poor, operators may not receivepe rtinent da ta in a timely and understand ab lema nner. This wa s part of the prob lem a t ThreeMile Island, where important indicators were onthe back of a control panel and unimportantalarms ad ded to the c onfusion of the ac c ide nt

sequence (15).

Nuclear units also differ from fossil plants in

their size. The lat est g ene rat ion nuc lea rpowerplants are very large, on the order of 1,000to 1,300 MWe. It was expected that nuclear unitswould be cost-sensitive to size changes andwould be most economical in very large units.

Coal plants, on the other hand, are much less sen-sitive to scaling factors, and most are less than

ha lf the size o f the newe st nuc lea r plant s (7). Thusnuclear construction projects not only involvemore sophisticated systems, but also larger ones,with a wo rk forc e o f 2,000 to 4,000 pe r unit. Thiscan significantly increase the difficulties in coor-

dinating and monitoring the activities of the vari-ous parties involved in nuclear construction.

In addition to being complex, nuclear technol-ogy is very exacting. As discussed above, the safe-

ty of nuclear powerplants is ensured by sophisti-cated control systems that must respond rapidlyand reliably to prevent an accident or mitigateits c onseq uenc es. These c ontrol and sa fety sys-tems must be constructed, maintained, and oper-ated according to the highest standards of excel-lenc e. This is so important tha t NRC ha s devel-oped detailed procedures for monitoring and ver-ifying quality. During the construction process,

NRC requires extensive inspection and documen-tation of all safety-related materials, equipment,and installation (13). In response to such require-ments, construction practices have become in-c rea singly spe c ific and inspe c tion proc ed ureshave become more formalized. An undesirableconsequence of this is that it is extremely difficultto construct nuclear powerplants in accordancewith very rigid and explicit standards. One ex-

amp le o f the c om p lica tions tha t c an result is inthe installation of pipe supports and restraints.It is not uncommon for field engineers to haveto work to tolerances of one-sixteenth to one-thirty second of an inch, which can be more re-strictive tha n the fab rica tion toleranc es used inmanufacturing the equipment to be installed (35).This results in greate r labor req uireme nts tha n infossil plants and can increase the level of skill re-quired. In addition, various levels of checks, au-dits, and signoffs are required for most construc-tion work in nuclear powerplants, adding to thelabor requirements necessary to complete instal-lation. One NRC publication has estimated thatthese checks add 40 to 50 percent to the basicengineering and labor costs (30). Figure 29 com-

pares labor requirements, including quaIity con-trol and engineering, for typica l coa l and nuclearplants. Note that for all the items listed here,nuclear reactors require at least half again asmuc h labo r as c oa l plants.




Photo credit: OTA Staff

To keep track of the many components of a nuclearplant under construction, each of which maybe subjectto modification, each pipe and pipe support is

labeled with a number that corresponds toa number on a blueprint

Another consequence of strict quality control

is that a large amount of paperwork is generated.

According to one recent estimate, approximate-

ly 8 million pages of documents have been pro-

duced to support the quality assurance program

for a nuclear unit that began operation in 1983

(32). In the midst of such massive requirementsfor pa pe rwo rk, it c an be d iffic ult, if not impo ssi-ble, to ma intain a p ositive a ttitude towa rd qua li-ty for its intrinsic value. This becomes even moredifficult in an environment where rework is re-quired freq uently, sinc e this adds to the pap er-work burde n and de c rea ses the mo rale of theworkforce.

The exac ting na ture o f nuclear tec hnologymanifests itself somewhat differently during op-

eration. It o ften is nec essa ry to ma intain extrem e-ly tight control of sensitive systems to keep theplant running smo othly. For examp le, the wa ter

chemistry system in LWRs must be carefully mon-itored a nd ad justed to p revent c orrosion a nd re-move radioactive materials from the coolingwater. Failure to maintain these systems within

narrow limits c an lea d to severe d am ag e in such

major components as steam generators or con-densers and this can ultimately curtail plant

operations (36). As discussed in chapter 4, cor-rosion has been a serious problem in many oper-ating PWRs and has led to replacement of steamgenerators in four nuclear units.

External Factors That InfluenceOperation and Construction

Certain other factors appear to be less relatedto the technology than to the environment inwhich commercial nuclear plants must operate.For example, the nuclear industry has experi-

enc ed problems with shortages of trained per-sonnel. The c om me rcial nuclea r po we r ind ustryrequires engineers, construction crews, and oper-ating teams to be qualified in very specialized andhighly technical areas. As shown in figure 30, thedemand for technical personnel with nucleartraining grew rapid ly d uring the 1970’s (2). At thebeginning of the 1970’s, the nuclear work forcewas very small, but many reactors had been or-dered and were entering the construction phase.

Lab or req uirem ents grew stea dily a nd pe aked inthe 4-year period 1973 to 1977, when the numberof p eop le emp loyed in the nuclea r industry in-creased at the rate of 13 percent a year. In theea rly yea rs of the c om me rc ia l nuclea r ind ustry,the greatest shortages were found among reac-tor d esigners. This c ontribut ed to the prac tice ofinitiating c onstruction w ith incom plete designs.While it was recognized that 60 percent or moreof the design should be completed before con-struction was initiated, some utilities began withhalf that or less. As plants have progressed fromthe design phase, through construction, and intoop era tions, the em pha sis on p ersonne l ha s a lsoshifted. Reactor designers are no longer in shortsupp ly, but there is a need for more pe op le q ual-ified in plant operations, training, and certain en-gineering disciplines (1 2).

A sec ond external problem is inadequate com-munication among utilities. Only a few utilities

had any experience with nuclear power beforethe 1970’s. A structured method for transferring

learning might have accelerated the overall pro-gress by providing wa rnings abo ut com mo n er-rors and transmitting effec tive p rob lem-solvingap proac hes. Such c om munication netw orks didnot exist in any formal manner until the accidenta t Three Mile Island stimulated an industrywide

effort to imp rove the transfer of nuclea r ope ra-tions informa tion. Even to day the re is little struc-ture in sharing information regarding reactor con-struction, with the primary mechanism being thetransfer of trained personnel from one utility toanother.




Figure 29.—Comparison of Manpower Requirements for Coal and Nuclear Powerplants

Reinforcing steel Coal

and structural steel

Concrete

Piping

Wire and cable

Welding

Hangers andrestraints

Nuclear

Coal

Nuclear

Coal

Nuclear

Coal

Nuclear

Coal

Nuclear

Coal

Nuclear

( + 480/o)

SOURCE. J D. Stevenson and F. A. Thomas, Selected Review of Foreign Licensing Practices for Nuclear Power Plants, April 1982

An additional consideration is that nuclear reac-tor owners have had to deal with a rapidly chang-ing regulatory environment throughout the pastdecade. Frequent revision of quality and safetyregulations and backfit requirements have greatly

affected construction and operation patterns. Asshown in figure 31, NRC issued and revised reg-ulatory guidelines at an average rate of three permonth in the mid-1970's (33).

Plants that were under active construction dur-

ing this time had to continually adjust to thechanging regulatory environment. While nosingle NRC requirement overtaxed the utilitieswith plants under construction, the scope and

number of new regulations were difficult to han-

dle. As a consequence, the utilities had to divertscarce engineering forces from design and reviewactivities to deal with NRC (37).

The utilities had to d ea l with more than a stea dy

increase in regulatory requirements: a series ofregulatory “shocks” was superimposed on thec umulative e ffect o f “ normal” regulation. A studyby EPRI identifies three ma jor events that we refollow ed by a flurry of NRC a c tivity: the C a lvertCliffs decision in 1971 to require Environmental

Impact Statements for nuclear plants, the fire atthe Browns Ferry nuclear plant in 1975, and the

ac c ident at Three Mile Island in 1979 (3). Th eaftermath of these incidents has created an at-

mosphere of regulatory unpredictabil ity that has



Ch. 5—Management of the Nuclear Enterprise q123

Figure 30.—Historical Labor Requirements in the Nuclear Power Industry

150

100

50

o I I I I I I I I I I I I I 1 I1970 1972 1974 1976 1978 1980 1982

Nuclear facility engineering,design, operation, andmaintenance

Miscellaneous (includingconsulting, lab testing,radiography, etc.)

Fuel cycle

Manufacturing and

production

Year

SOURCE ‘ Occupational Employment in Nuclear-Related Activities, 1981 .“ Oak Ridge Associated Universities for the Department of Energy, April 1982

particularly affected plants in the construction

phase. In some cases, major portions of nuclear

construction projects have had to be reworkedto c om p ly with c hanging req uirem ents. For ex-

ample, after the fire at the Browns Ferry plant,NRC issued new requirements to fireproof alltrays c a rrying e lec tric a l ca bles. While this wa s not

an unreasonable request, it did disrupt many con-

struction schedules.

In ma ny c ases, cha nge s in NRC regulations ob -viously enhance plant safety. In other cases, it

is not clear that safety is increased by adding ormodifying systems and components. As discussedin chapter 6, the adverse impacts of certain reg-

ulations include eq uipm ent we arout due to e x-cessive surveillance testing and restrictions on ac-

c essibility to vita l equipm ent as a result of fire o rsecurity barriers (37).

Piping system design provides another exam-

ple of po ssible a dverse effec ts of reg ulation. Thecurrent trend in NRC guidelines is to require morerig idly sup ported syste ms. This is no t ne c essari-ly because flexible systems are less safe, but ana-

ytical techniques cannot unequivocally provethem sa fe. Rigid systems are easier to a na lyze,

but can present serious operational difficulties

during routine changes in temperature (23).

Finally, rapid technological changes have fur-

ther complicated nuclear powerplant construc-tion and operation. Nuclear reactors were scaledup from the earliest demonstration plants of sev-

eral hundred megawatts to full-scale 1,000-MWeplants within a decade, By 1968, most orderswere p lac ed for units grea ter than 1,000 MWe.

As shown in table 20, there were only three LWRswith a generating capacity greater than 100 MWe

in op erat ion in the United Sta tes when the firstof these orde rs wa s p lac ed . Thus the designs forthe larger plants were not built on the construc-tion and operating experience of gradually scaled

units. By the time the first 1,000-MWe units beganop eration in 1974, an ad d itional 70 large plantshad b een orde red . There wa s little opp ortunityfor orderly, deliberate design modification andtransfer of knowledge in this rapid scale-up.

The fa c tors disc ussed ab ove have c ontribu ted

to the complicated task of maintaining rigid stand-ards of excellence in nuclear powerplants. As aresult, the construction and operation of nuclearreactors has demanded the full resources, both

technical and financial, of the utilities. Many util-

ities have fa iled to fully me et t hese c ha lleng es.Others, however, have managed to cope with allof these complications—plants have been con-

structed with few major setbacks and operated



124 qNuclear Power in an Age of Uncer ta in ty

Figure 31.— NRC Regulatory Guidelines Issued From

40(

300

100

0

1970 to-1980

.

. ,

.

19701975Year ending

we ll. This sug gests tha t some of

among util it ies can be attributedin factors internal to the utility.

1980

the va riabilityto differences

Internal Factors That InfluenceConstruction and Operation

dif-Factors related to uti l ity management areficult to assess since they are less visible than ex-ternal factors; moreover, they are not easily quan-

tified. Nonetheless, they can influence the finan-cial success of a nuclear project or plant safety.As discussed above, there area number of char-acteristics that distinguish the management of nu-


Cable trays increased in weight and complexitybecause of fire-proofing and separation of functionrequirements following the fire in the electrical system

at Browns Ferry in 1976

Table 20.—Early Operating Experience of U.S.

Commercial Light Water Reactors

Date ofcommercial

Unit Size (MWe) Operation TypeDresden 1 . . . . . . . . . . 207 8/60 BWRYankee Rowe . . . . . . . 175 6/61 PWRBig Rock Point . . . . . 63 12/62 BWRHumboldt Bay 3 . . . . 63 8/63 BW RConnecticut Yankee . 582 1/68 PWRSan Onofre 1 . . . . . . . 436 1/68 PW RLa Crosse . . . . . . . . . . 50 11/69 BW RNine Mile Point 1 . . . 610 12/69 BW ROyster Creek 1 . . . . . . 620 12/69 BW RSOURCE “Update — Nuclear Power Program Information and Data, October-

December 1982, ” U S. Department of Energy, February 1983

clear technology from that of other conventional

po we r technolog ies. The c om plexity of the rea c -tor and the demands for precision and documen-tation provide significant challenges to utilitymanagers.

Even more important are the difficulties in deal-ing w ith a c ha nging environm ent. Suc c essful utili-ty manag ers have ha d to ma intain a great dea lof flexib ility to kee p up with the rap id grow th in

the size and design of nuclear plants and changesin regulatory structure. In fact, some utilities havereorganized several times in an attempt to con-trol their nuclea r projec ts b ette r. The mo st c om -mon changes have been away from tradit ionalline management and towards matrix or projectmanagement (3). While this has been successful



Ch. 5—Management of the Nuclear Enterprise q12 5

in some cases, it is not always sufficient to im-

prove the qual i ty of ut i l i ty management. Other

factors are also very important, as discussed

below.

Managing nuclear power projects requires a

commitment to safety and quality that is lessessential in othe r elec tric utility op erations. This

implies far more than a concern for schedules and

budgets, which pervades all commercial endeav-

ors. Because there is some possibility that an ac-

cident could occur in a nuclear reactor, every ef-

fort must be made to protect the investment of

the utility and the safety of the public. It is im-

portant that nuclear managers adhere to the spirit

as well as the letter of safety and quality-assurance

reguIations.

The Palo Verde plants are good examples of

commitment to quality (6). When Arizona Pub-

lic Servic e a nnounc ed its nuc lea r p rog ram in

1972, it thoroughly studied all aspects of design-ing and constructing nuclear powerplants. Manyadvanced safety features were incorporated intothe Palo Verde design from the beginning of theproject. One unexpected consequence of this at-

tention to safety is that Arizona Public Service an-ticipa ted ma ny of the Three Mile Island bac kfitand redesign requirements. As a result, regula-

to ry cha ng es in resp onse to Three M ile Island ha dless imp ac t on the Pa lo Verde p rojec ts tha n on

many other plants which had not originallyplanned to inco rpo rate the e xtra safety fea tures.

Sinc erity of c omm itment c an be ob served inseveral ways. Highly committed senior managerscan impress their commitment on project mana-gers, who in turn can communicate it to de-

signe rs, ma nufa c turers, a nd c onstruc tors, Thestreng th o f utility co mm itment is a lso ind ic a tedby th e leve l of q ua lity req uired in the u tility’s co n-tractual and procedural arrangements with sup-pliers of material, equipment, or personnel, For

example, if a contract primarily emphasizes theschedule for physical installation, the messagefrom project management is production. On theother hand, if the contract also emphasizes

owner-acceptance and adequate documentat ion,the me ssa ge is q ua lity a s well as p rod uc tion. Thelatter case provides the proper incentives for high-

qua lity wo rk (1 3).

Corporate co mmitment a lso c an b e indica tedby the wa y in which a utility respo nds to c hang esor prob lems. The m ore suc c essful utilities ha vea history of responding rapidly and with adequatefinancial resources to resolve problems and adaptto new situa tions. Other utilities with less ea ger-ness to confront their problems directly have ex-

perienced construction delays and operationaldifficulties (3).

An important factor in the management of any

powerplant is the distribution of responsibilitya nd autho rity. This is pa rtic ularly vita l in the c on -struction of nuclear plants because of the com-plexity of the tec hnology a nd the need to c o-ordinate the activities of vendors, architects, engi-neers, construction managers, consultants, quali-


As a plant nears completion, responsibility is graduallytransferred from the construction managers tothe operating division. These tags give an idea of

the detailed level at which explicitresponsibility is assigned




ty inspectors, test engineers, operators, and crafts-men. In this environment, it is vital that a utilityesta blish c lear lines of a uthority a nd spec ific re-sponsibilities to ensure tha t its ob jec tives will bemet .

When authority and responsibility are diffusedthroughout an organization rather than focusedin a few key positions, widespread problems canresult. This oc c urred a t the Washington Pub licPower Supply System (WPPSS) in the late 1970’s,when extensive construction delays and regula-tory difficulties plagued the WNP-2 plant in Rich-Iand, Wash. (9). Responsibility for design andconstruction was distributed between the ownerand the construction company in an obscuremanner, with neither the owner nor the construc-tion manager claiming authority in decisionmak-ing nor accepting responsibility for mistakes. Ad-ditional problems arose whe n a uthority wa s fur-ther delega ted to the c ontrac tors without suffi-c ient p rovisions for mo nitoring feed bac k. Whilemost contractors assumed the responsibility for

maintaining quality, others were less conscien-tious. In some cases unqualified people wereused for construction work, and documentationwa s incomp lete and inconc lusive. The p rojec tmanagement effectively lost control of the sub-c ontracto rs and wa s in no p osition to de tec t andcorrect these problems.

When this situation came to the attention ofNRC in 1980, a stop-work order was issued for

WNP-2. WPPSS tackled its problems directly by

completely restructuring its project management.It established clear and direct lines of responsibili-ty for design and construction by creating the new

position of Project Director and by specifying therole of the c onstruc tion m ana ge r rep orting tohim. New review and surveillance procedures

were init iated by the construction manager andoversee n b y WPPSS. Co nstruc tion fina lly w as re-sumed when NRC was satisfied that the majordeficiencies had been resolved, and WNP-2 isnea rly c om p lete. The four othe r WPPSS p lants

und er co nstruc tion we re less fortuna te. Two unitsin the early stages of construction were moth-ba lled in 1981, and WPPSS ha s since d efaultedon repaying the outstanding debt on those plants.

Subsequently, construction was halted on 2 otherplants that were more than 60 percent complete.

These four plants we re troub led by the inab ility

to assure continued financing and the decreas-ing nee d for powe r in the Northwest. The m an-agement restructuring came too late to graceful-

ly reverse the effects of early planning decisions(l).

The e xamp le d isc ussed above is only one of

many in which utilities learned that it is in theirbest interests to monitor and control the activitiesof the ir c onstructo rs and a rc hitec t/ eng ineers. Itwa s c om mo n in the 1970’s for utilities, espec ial-ly those w ith little experienc e, to relinq uish mostof the responsibility for design, cost, and schedule

to their contractors. As problems developed, theutilities gradually became more involved withtheir constructors; this resulted in shifting respon-sibility to the p lant ow ners. The same oversightcould be applied to architect/engineers, and thereare recent signs that this is happening in the largerand more experienced nuclear utilities.

Once a util ity has developed a workable orga-nizational structure, it is further challenged to

coordinate and motivate the many diversegroups of p eo ple involved in nuc lea r construc-tion. At the p ea k of co nstruc tion o n a large nu-clear unit, as many as 6,000 craftsmen, engineers,and supp ort personnel ma y be wo rking to ge ther.In such situations, scheduling can become alogistic nightmare, and a sense of teamwork andhaving com mo n go a ls c an vanish. These p rob -lems are exacerbated by changing regulatory re-

quirements that can result in construction rework

and delays and by the lack of continuity that re-sults from long construction times (1 7).

Coordination can be equally challenging withina utility’s management structure, especially if theutility is und ergo ing o rga niza tiona l c ha ng es. Thisoccurred at Commonwealth Edison Co. in Chi-cago, where a matrix-management structure wasreplaced with formal project management. In thenew organization, each nuclear construction proj-ec t wa s given its ow n sta ff, inc luding eng inee r-ing, construction, testing, and startup personnel.An independent staff of quality and safety

engineers was maintained in a central off ice toprovide oversight and ensure uniformity amongthe p rojec t tea ms. There a re som e o verlapp ingand conflicting functions in the new system, andstrong lead ership from the ma nage ment within



Ch. 5—Managemnt of the Nuclear Enterprise Ž127

Commonwealth Edison has been required to in-still a sense of teamwork among the variousgroup s (25),

Another problem that some utilities face is thatnuclear projects may make excessive demandson their resources. Some utilities have not beenable to hire qualified management personnel, and

they have not had sufficient time to gain the man-agement and technical experience independent-ly. As previously discussed, this often resulted inc onstruction b eing started with inco mp lete de -signs. Furthermore, limited resources have madeit difficult for some utilities to provide for ade-

qua te training in qua lity inspe c tion a nd rea c toroperations while simultaneously constructing anuclear plant.

A final consideration is experience in construc-

tion and op eration. It is mo re diffic ult for a util-ity with no experience to cope with nuclear

power’s unique characteristics than it is for a util-ity with seve ral nuc lear plan ts. In fac t, a rec entEPRI study concludes that nuclear experience isone of the most significant variables influencing

c onstruc tion time s (3). Tha t stud y further co n-c lude s that a utility ca n c omp ensate for lac k ofexperience by relying on an architect/engineeror constructor that has previously dealt with nu-Iear projects.

Lack of experience was a major source ofHouson Lighting& Power Co.’s problems in con-

structing its South Texas projects. It selected theAE firm, Brown & Root, Inc., even though thatfirm was inexperienced with large-scale nuclearplants. After a numb er of qua lity problems c am eto light, NRC issued a stop-work order in 1980

for all safety-related construction. Houston Light-ing & Pow er is a ttem p ting to resolve these d if-ficulties by replacing Brown & Root with the moreexperienced firms of Bechtel Corp. and EbascoServices, Inc.; they are also acquiring in-housec apab ility by hiring w ell-trained eng inee rs (28).This latte r ap p roac h ha s b ee n used suc c essfullyby othe r utilities. When Arizona Pub lic Servic e

Co. initiated construction of its first nuclear pow-erplants, it expanded its staff with engineers andmanagers experienced in nuclear power.

IMPROVING THE QUALITY OF NUCLEARPOWERPLANT MANAGEMENT

The p rob lems d isc ussed ab ove ind ic a te tha tgreat dedicat ion is needed to manage nuclearpow erp lants. This tec hno logy presents ma ny

challenges to successful management, and notall utilities have demonstrated that they have theskill, resources, and commitment to meet thechallenge.

Several different a pp roa c hes c an be taken toalleviate management problems. One possibilityfor reactors that will be sold in the future is toredesign them to be simpler and safer, and hence

less susceptible to management control failures.This app roac h sug gests rea c tors tha t are more“ forgiving” of huma n errors tha n c urrent LWRs,

as discussed in chapter 4. A parallel effort might

attempt to raise the quality of managementthrough institutional controls.

Technical Approach

The te c hnic al imp rovem ents that c ould b emade to the current generation of nuclear reac-tors range from minor evolutionary modificationsto c lean-sheet designs. As d isc ussed in de ta il incha pter 4, a recent EPRI survey indicates thatmany utilities would like to see at least minor

changes in new LWRs to enhance conservatism,reliability, operability, and maintainability (1 7).

It was proposed that the next generation of LWRdesigns focus on simplicity, reduced sensitivity




to anticipated transients, and reduced system in-teractions. Modifications of this sort could relievesome of the pressures on management by requir-ing less p rec ision d uring both c onstruc tion a ndoperation.

Ma nag eme nt bene fits also c ould be d erivedfrom standardizing the LWR, with or without

mo d ific a tions. This wo uld a llow the utilities totransfer lea rning from one unit to a nothe r, andall plants could gain from the experience of anyplant owner. Another potential benefit of stand-ardization is that the regulatory climate is likelyto be less active once an industry-wide design has

be en selected and ap proved . As a result, stand -ardization should reduce the frequency of NRC-inspired design changes (22).

The m od ific a tions d isc ussed above fit within

a pa ttern of e volutiona ry de velopme nt of LWRtec hnology. They do not involve drama tic c hang es

to components or to the basic reactor design, butfoc us on rec on figuring the c urren t system . Thesechanges would be welcomed by the nuclear util-ities, and they could made nuclear reactors some-wha t easier to ma nage . It is unlikely, how eve r,that they wo uld signific antly red uc e the overalllevel of management intensity required to han-dle nuclear projects. It is possible that this couldbe accomplished by a more innovative and dras-tic alteration to the present tec hnology.

More extreme technical alternatives include re-designing the LWR to op timize it fo r sa fety or re-plac ing the LWR with a n a lternat ive d esign. Thesereactors might prove to be less sensitive to man-agem ent c ontrol failures than c urrent LWRs. For

exam ple, the Fort St. Vra in high tem p erature g a s-c oo led rea c tor (HTGR) has expe rienc ed seve ra lincidents that have had no significant conse-

quences, but which might have been serious inan LWR. In one incident, forced circulation cool-ing was interrupted for more than 20 minutes.Bec ause the HTGR has a high hea t c ap ac ity andthe fue l has a high me lting po int, there w as no

damage to the core or components.

Even drastic c hang es in te c hnology, howe ver,

c anno t eliminate a ll of the p rob lems that fa c emanagers of nuclear projects. In particular, modi-f ication of the technology cannot replace high-Ievel c omm itment to qua lity and safety, which

must accompany the construction and operationof at least the nuclear island of any reactor. Norwould a new d esign eliminate the d ema nds formanagement skills during the construction proc-ess—effective distribution of responsibilitieswould continue to be important, as would the

need to coordinate and motivate the construc-

tion work force. In short, inherently safe reactorsmight m arkedly red uc e the p rob lems tha t arisefrom technological considerations, but new de-signs cannot alone ensure high-quality construc-tion and safe and reliable operations.

Institutional Approach

Institutional approaches focus on internal andexternal fac tors that a ffect p erformanc e rather

than on technical considerations; in this sense,they complement the activit ies taken to reducethe complexity and sensitivity of nuclear reactors.

Institutional measures can be divided into twotypes of activities:

q those that create a favorable environmentfor successful utility management of con-

struc tion and op eration. Suc h a c tivitieswould focus on external problems, includingefforts to enhance communications, increasethe sup p ly o f trained pe rsonne l, and stab ilizethe regulatory environment; and

q those that monitor utility operations to detectma nag eme nt failures and elic it be tter per-formance from the less successful utilities.

Suc h effo rts wo uld foc us on p rob lems tha tare specific to individual utilities.

Two princ ipa l organ iza tions a re now involvedin institutional controls that monitor operationsand improve c om munica tions. The NRC ha s longbeen involved in programs designed to regulatethe nuc lea r industry and to p rotec t the p ublic .In the p ast, its initiat ives we re foc used p rimarily

on design and licensing issues for reactors in theconstruction phase. As the nuclear industry con-tinues to m ature a nd more p lants enter op era-tion, it is expe c ted tha t the e mp hasis grad ua lly

will shift to operating plants.The Institute o f Nuclea r Pow er Opera tions

(IN PO) is a more recent participant in this area,and its influence is growing rapidly. INPO is spon-




Photo credit: Pennsylvania Power & Light

Site-specific control room simulators, used increasingly by utilities to train nuclear plant operators, are duplicates ofactual control rooms. Possible nuclear operating events are simulated on control instruments by computer. The simulator

shown here trains operators of PP&L’s Susquehanna plant and cost $6 million.

sored by the nuclear utility industry, and everyutility with a nuclear plant in operation or under

construction is a member. In addition, utilityorganization in 13 other nations participate inIN PO. It was formed in 1979 in response to the

ac c ident a t Three Mile Island .

Creating the Right Environment

An area that has received considerable atten-tion is the improvement of communication

among utilities. The ut ilities ha ve c om bined

force s to form the Nuclea r Sa fety A nalysis Ce nter(NSAC) to analyze technical safety problems and

solutions. NSAC has a lrea dy add ressed a numb erof issues, inc luding t he a c c ident a t Three M ileIsland, NRC’s unresolved safety items, and de-

graded cores.

INPO has been active in collecting, evaluating,

and redistributing utility reports of operatingexperienc e. This is partic ularly imp ortant in viewof the massive amount of information that is gen-erated by operating nuclear powerplants. It is achallenging task to distinguish the vitally impor-tant from the m ore mund ane repo rts. INPO d e-velop ed the Signific an t Events Eva luation a nd In-

forma tion Netw ork (SEE-IN) to hand le informa -

tion on an industrywide basis. In 1982, more tha n

5,000 event reports were screened, and approxi-

mately 100 significant reports relating directly to

plant reliability and safety were identified (41).The mo st frequently c ited prob lems involvevalves and electrical and instrument controls,closely followed by the reactor coolant system,

steam generators, diesel generators, and piping.Afte r ad d itiona l rev iew, INPO d istilled this infor-




mation into a few important recommendations.Among these recommendations are measures topreclude equipment damage, reduce prolongedoutages, and minimize radiation exposure.

In addition to identifying generic problems inthe industry, the SEE-IN program also checks to

see if its recommendations are implemented.Thus fa r it ha s b ee n ve ry suc c essful in enc ou rag -ing utilities to m ake voluntary cha nges to c om -

ply with its rec om me ndat ions. Nearly ha lf of a lloperating plants implemented every “immediateatte ntion” rec omm end ation within a year, andmany other plants are making progress in thisdirec tion (41).

NRC is planning a similar program in which itwill system atica lly ana lyze the informa tion tha tit collects through various reporting mechanisms.NRC plans to computerize its data base and searchmethods so that it can better detect generic prob-lems (13).

A related NRC activity is focused on construc-tion rather than operation. A long-term effort toreview quality-assurance problems and to pro-pose changes that could improve quality in de-sign, construction, testing, and operation has

bee n initiate d . This review will sta rt with an e xam -ination of nuc lear plants at Diab lo Canyon, SouthTexa s, Midlan d , Ma rble Hill, and Zimme r to ide n-tify specific problems and causes. At various timesin the past, NRC ha s issued stop -work orders a teach of these plants due to concerns about the

quality of construction. NRC will also examineprojects with good records to identify the positivemea sures that c ould b e a pp lied ge neric ally (13).

The d a ta ana lysis effo rts by both NRC a ndINPO should enhance the formal transfer oflearning a mo ng the utilities. Other INPO activitiesattempt to improve communication less formal-Iy by providing a forum for the exchange of ideas.

Managers involved in nuclear plant constructionand ope rat ion are encouraged to meet a t work-shops and conferences to share their experiencesin detecting and solving problems. Another way

in which INPO e nco urag es the e xc hang e o f in-formation is through its electronic communica-tion system known as “Nuc lea r Note pa d .” This

system provides timely transfer of news on im-

portant items by linking all operating nuclearplants in a sing le c om puter netw ork (1 1).

INPO also is active in developing guidelines inthe areas of training accreditation, emergency re-spo nse, and radia tion p rote c tion (39). These a c tiv-ities are coordinated with NRC needs and re-

q uirem ents. There is c urrently a grea t d ea l of va ri-ab ility in the wa ys in w hich utilities hand le p rob -lems in these areas. As greater uniformity devel-ops, the utilities should be better able to learnfrom one ano ther and raise the level of p erform-ance on an aggregate basis.

Another important activity within NRC is the

effort to moderate regulatory activity by control-ling requirem ents for change s during c onstruc-

tion and operation. In 1981, NRC established theCommittee for Review of Generic Requirementsto assess backfitting proposals and to try to reducethe burden of shifting requirements. As discussedin cha pte r 6, this is a d iffic ult ta sk since safe ty isnot easily quantifiable, and many technical uncer-tainties remain. However, this effort has the po-

tential for enhancing the ability of utilities to con-struct new plants in a timely and efficient manner.

Detecting and ImprovingPoor Performance

Improvements have been made in certain as-pe c ts of the c om me rc ial nuc lea r industry in re-cent years. The accident at Three Mile Island

c onvinced ma ny utilit ies of the impo rtanc e o f at-

tention to quality, and some have made volun-tary changes to improve management. For exam-ple, a number of utilities have modified their

organizational structure in an attempt to find onebe tter suited to b uilding a nd o pe rating nuc lea rpowerplants (3).

The utilities tha t a re sensitive to qua lity c on-

cerns and responsive to NRC and INPO initiativesare probably not a source of great concern;rather, the concern is centered on those utilitiesthat do not seem to be responding to the samemotivation. In fact, the most successful utilities

claim that they a re b eing “ held hostage ” by theleast capable and least committed utilities (23).They fea r that a nother ma jor nuclear ac c ide nt inany c omm ercial rea c tor would have disastrous



Ch. 5—Management of the Nuclear Enterprise • 131

consequences for all nuclear plant owners interms of public acceptance. It is, therefore, in theinterest of the best performers to ensure that poorpractices are detected and eliminated, whereverthe y oc c ur. This c onc ern ha s Ied to INPO’ s ex-tensive evaluation and assistance programs to at-ta in exce llenc e. NRC a lso eva lua tes utility pe r-

forma nce , but w ith a different pe rspe c tive. Its in-tent is to ensure that construction and operationof all nuclear units meet minimum standards, asdefined by NRC.

INPO’s evaluation programs appear to be well-

structured and in logica l relation to o ne ano ther.The first INPO efforts were devoted to establish-ing a comprehensive system for eva luating op -erating nuclear reactors. In an operating planteva luation, spe c ial tea ms of up to 15 peo ple a resent to each nuclear unit to assess the perform-ance of the utility in many different areas, in-

cluding those shown in table 20. A final reportis prepared to summarize the findings, make rec-ommendations for improvements, and identifygoo d p rac tice s. This report is reviewe d w ith thehighest levels of utility management, who devel-op a plan of action for implementing INPO rec-ommendations (26).

INPO has completed two rounds of operatingplant e valuations and has init iated a program toevaluate construction projects. Construction eval-uation proced ures have b een d evelope d a ndhave been applied to 18 near-term operatinglic ensee p lants. The first p ha se o f the c onstruc -tion assessment program was completed in 1982whe n 22 utilities with nuc lear p lants und er c on-

struc tion performe d self-eva luations. The sec ondphase of evaluations is being conducted by eitherINPO o r ind ep end ent o rga nizat ions under co n-tract to the utilities and monitored by IN PO. NRChas been following INPO’s evaluation effortsclosely and may restructure some of its own quali-ty initiatives around the industry program (39).

In addition to evaluating nuclear plants, INPOis assessing the corporate support of nuclear util-ities. Corporate evaluation criteria have been de-

veloped by a task force of senior utility execu-tives. These c riteria have been field-tested andare in use in INPO evaluation programs (26).

INPO evaluation reports have proven to be val-ua b le in severa l wa ys. First, the y fo rm the basisfor INPO’ s “ go od prac tice ” rep orts, whic h sum-marize effective approaches used throughout the

nuc lear ind ustry. These reports a re p artic ularlyuseful to utility managers who want to identifyproblem areas and adopt approaches that have

been used successfully in other plants.

The sec ond ma jor c ontribution o f INPO e va lua-tions is to highlight problem areas in individualutil it ies and make recommendations to improveperformance. In the event there is some reluc-

tanc e o n the p art of a utility to c omp ly with INPOrecommendations, a number of actions can betaken to enc ourage c oop eration. These a c tionsare designed to raise the performance of all util-ities by applying peer pressure, from other leadersin the industry. These p ressures c ou ld b e a p p liedthrough utility chief executive officers, boards of

directors, and insurers. Such actions have not yetbeen required.

The NRC has its own series of plant inspections.Starting in 1978, resident inspectors were locatedat each nuclear plant to monitor day-to-day oper-a tions. These insp ec tors p rovide mu c h of the in-formation that is used to develop the SystematicAssessments of Licensee Performance (SALP),which are prepared by NRC’s five regional of-fice s. The SALP’s ana lyze p erforma nc e in eac hof 10 categories, which are similar to the INPOc ateg ories show n in tab le 21. The go a l of this

evaluation is to identify areas in which manage-ment excels, areas which call for minor improve-ments, and those in which major weaknesses areevident. NRC uses this assessment to direct its in-spe c tion e fforts and to sugg est c hang es to theplan t ow ners.

A more comprehensive NRC evaluation effortinvolves the Performa nc e Assessme nt Tea m(PAT). This team op era tes from NRC hea dqua r-ters, and its inspe c tions p rovide a c hec k on theNRC regional offices and the INPO evaluations.Although the PAT eva luations overlap the SALP’ s,

they are b roa der in sc op e, w ith assessments ofmanagement controls and broad recommenda-tions for cha nge (24).




Table 21.—Classifications for INPO Evacuations

Organization and administration: Station organization and ad-ministration; management objectives; management assess-ment; personnel planning and qualification; industrial safe-ty; document control; station nuclear-safety review group;quality programs

Operations.K)perations organization and administration; con-duct of operations; plant-status controls; operator knowl-

edge and performance; operations procedures and docu-mentation; operations facilities and equipmentMaintenance: Maintenance organization and administration;

plant material condition; work-control system; conduct ofmaintenance; preventative maintenance; maintenance pro-cedures and documentation; maintenance history; main-tenance facilities and equipment; materials management

Technical support: Technical-support organization and admin-istration; surveillance-testing program; operations-experience review program; plant modifications; reactorengineering; plant-performance monitoring; technical-support procedures and documentation

Training and qua//f/cation: Training organization and admin-istration; licensed and nonlicensed operator training andqualification; shift-technical-advisor training and qualifica-tion; maintenance-personnel training and qualification;training for technical staff; training for managers andengineers; general employee training; training facilities andequipment

Radiological protection: Radiological-protection organizationand administration; radiological-protection personnel train-ing and qualification; general employee training in radio-logical protection; external radiation exposure; internalradiation exposure; radiological-protection instrumentationand equipment; solid radioactive waste; personnel dosim-etry; radioactive-contamination control

Chemistry:Chemistry organization and administration; chem-istry-personnel training and qualification; chemistry con-trol; laboratory activities; chemical and laboratory safety;radioactive effluents

Emergency preparedness: Emergency-preparednessorganization and administration; emergency plan; emer-gency-response training; emergency facilities, equipmentand resources; emergency assessment and notification;emergency-personnel protection; personnel protection;emergency public information.

SOURCE: Institute of Nuclear Power Operations

In 1982, NRC decided to limit the number ofPAT inspec tions in rec ognition o f the similarityto INPO’ s prog ram s. The PAT p rog ram orig inal-ly was scheduled to evaluate each reactor on a

3-to 4-yea r cyc le. They w ere limited to only twoto three inspections per year after 1982.

Another phase of the NRC inspection programfoc uses on ne a r-te rm op era ting lic ensee s (NTOL).To increase its confidence in the quality-assur-

ance programs at plants that will soon begin op-

eration, NRC now requires a self-evaluation of

quality-assurance programs for design and con-struc tion a t these p lan ts (13). This inc ludes areview of management involvement, audits, sig-nific ant p rob lem s, and c orrec tive ac tions. Theself-evaluations are supplemented by NRC re-gional office reviews. These assessments examinethe inspe c tion a nd enforcem ent history of theproject to determine whether additional inspec-tions a re nee de d . In ad d ition, NRC enc ourag esinde pend en t d esign review s a t a ll NTOL utilities.

The p urpo se o f the NRC inspec tion a c tivitiesis to identify severe or recurrent deficiencies.

There is less effo rt ma de t o a na lyze t he struc tureand commitment of uti l i ty management than toidentify prob lems tha t m ight a rise from the failureof m an ag em ent c ontrols. Thus, NRC e va luationsserve the purpose of indirectly monitoring the

source s of p rob lem s by d irec tly monitoring theirmanifestations. In contrast, the INPO evaluationsfocus directly on weaknesses in management sys-tem s and c ontrols.

In the event that any of the NRC inspectionsuncover major problems, NRC has recourse toa series of progressively severe penalties. Enforce-

ment actions include: formal notification of a vio-lation; imposition of a fine if the licensee com-

mits a major violation, willfully commits a viola-tion, or know ingly fa ils to rep ort a viola tion; andfinally, the modification, suspension, or revoca-tion of a license. In the most extreme cases, NRCcan refer the case to the U.S. Department of jus-tice for investigation of criminal violations.

ASSESSMENT OF EFFORTS TO IMPROVE QUALITY

The mana ge ment of co mme rc ial nuclea rof U.S. plants is quite g oo d , the reliab ility of the

powerplants has proven to be a more difficult task plants has been less than hoped for, and severalthan originally imagined by the early proponents accidents have occurred which have reduced the

of nuclear technology. While the safety record confidence of the public in the industry. Further-



Ch. 5—Management of the Nuclear Enterprise Ž 133

more, construction projects have been plagued

with cost and schedule overruns and questions

about qual i ty.

Nuclear power is not so intractable that it can-

not be managed in an exemplary fashion; this has

been demonstrated by the records of the most

capable uti l i t ies. However, ut i l i t ies with onlyaverage skills and commitment have been much

less successful in managing nuclear projects. Bet-

ter app roac hes are needed to improve the op era-

tion of today’s plants and to establish public con-

fidence that the uti l i ty industry could manage

new reactors if they are needed in the future.

Both technical and institutional changes could

help improve the management of nuclear pow erop erations. Tec hnica l mo d ific a tions wo uld beuseful insofa r as they d ec rea se the c om p lexityand sensitivity of nuc lea r plant s. Som e o f thesechanges are relatively simple to make and havebe en incorpo rated in the d esign of ne w LWRs.It is likely that other more drastic design changescould further decrease the sensitivity of nuclearreactors to their management by making theminherently safer and less dependent on engi-neered systems.

While a technical solution to all managementproblems would be welcome, it is not likely tobe forthcoming. Even if an ultrasafe reactor couldbe developed, the demands on its operators toensure reliable performance would still be greaterthan in a fossil plant. Furthermore, even drastic

changes in reactor design would not significantlydecrease the sophistication or complexity of thenuclear island, even though they might allow ared uc tion in the safe ty req uirem ents for the re-ma ining of p lant system s. Overall, nuc lea r co n-struction managers still would be taxed to coor-d inate m assive projec ts am id e xac ting req uire-ments for high levels of quality and extensivedocumentat ion.

Since te c hnological cha nges c anno t by them-selves eliminate all the difficulties involved in con-structing and operating nuclear units, it is impor-

tant that they be supplemented with institutionalmeasures to improve the management of thenuc lear en terprise. For examp le, NRC c ou ld re-duce pressures on utility managers by exercisingas much care as possible in expanding regulatory

requirements; INPO could further improve the

situation by enhancing communication amongthe u tilities. The m ore d iffic ult and importa ntchanges, however, relate to the internal manage-ment of nuclear utilities. Utility managers mustbecome aware of the unique demands of nucleartechnology, and they must develop the commit-

ment and skills to meet them. INPO could be in-strumental in stimulating this awareness and inproviding g uidanc e to the u tilities. INPO rec og -nizes this point and is striving to develop suchutility management awareness. It is likely that theutilities will tend to be mo re rec ep tive to INPOthan to an outside organization since INPO is a

creation of the nuclear industry.

It is eq ua lly imp ortant tha t the nuc lear utilities

be evaluated objectively to assure that they areperforming well. NRC and INPO have recognizedthe need for such evaluations, and both organiza-

tions currently are engaged in assessment activ-ities. The INPO a ssessme nts, wh ich now c overmany areas, continue to evolve, and so far ap-pear to have been handled with sensitivity andinsight. The INPO eva luations a tte mp t to assessthe performance of the utility management to

identify the root causes of the problems and rec-om me nd c orrec tive ac tions. The NRC insp ec tionprogram is more fragmented and somewhat unfo-c used . The rela tionships amo ng the va rious in-spection activities are not always clarified,although these activities should complement oneanothe r. Furthermore, most o f NRC’ s inspec tion

ac tivities c onc entrate on the c onseq uenc es ofquality problems rather than on the sources. Itshould be noted that NRC does try to identifymanagement control failures once a problem sur-fac es, but tha t this is not a part of its sta nda rd

evaluation procedure.

INPO a nd NRC c omm unic ate with one a notherc onc erning their respe c tive e valuation and in-spe c tion a c tivities, and they a re a ttemp ting tocoordinate their programs. In establishing their

respective roles, it should be noted that the INPO

evaluation teams may be better able to commu-

nicate with uti l i ty managers and discover thesource of p rob lems. But this doe s not imp ly tha tNRC should turn over its inspec tion a c tivities toIN PO; the p ublic must ha ve c onfide nce that theutil it ies are being evaluated objectively and ac-

25-450 0 - 84 - 10 : QL 3




curately, and only a government organization canguarantee such objectivity. NRC currently accom-plishes this by carefully monitoring INPO activi-ties and by performing independent assessmentson a limited basis. How eve r, NRC a lso performsa variety of other detailed evaluations that are notwell-coordinated with INPO activities. Som e d u-

plication of effort maybe appropriate, since NRCmust rema in objec tive and informed in its ove r-

sight role. However, it should be possible to bet-ter co ordinate the a ctivities of the two organi-zations to make better use of limited resourcesand relieve the utilities of redundant inspections.

Enforcement activities also can be very impor-ta nt in ra ising the level of the poo rest utility per-formers. Both NRC a nd INPO ha ve a numb er ofmeasures at their disposal to encourage utilitiesto make changes or penalize them if they don’tcooperate. INPO operates through peer pressure,and it is not clear that it would actually invokeits strongest measures. INPO has not yet foundit necessary to exercise all its options.

NRC operates on a different level with a seriesof enforceme nt ac tions that c an b e ta ken if itdete c ts an unw illingness of the nuc lear industryto regula te itself w ith sufficient stringenc y. NRChas proven w illing to exercise the op tion of fin-ing utilities whe n it de tec ts breac hes of sec urity

or sa fety reg ulations.

It is difficult at this time to assess the effec-tiveness of the efforts of the nuclear industry and

its regulators to improve plant performance.There is not ye t a ny c lear evidenc e tha t the util-ities have been able to translate NRC and INPOprograms into better reliability and fewer safety-related inc idents. INPO is still in the process ofestablishing its guidelines and evaluation proce-dures, and NRC is just starting to assume a moreactive role in evaluating management controls.

However, the next few years should provide theevidence needed to evaluate these initiatives. Itis not yet clear that significant improvements willoccur in management of construction since thereare few forma l me c hanisms for transferring lea rn-ing or developing more successful approaches.It is possible that operational reliability and safe-

ty will improve noticeably if the NRC and INPOinitiatives are suc c essful. Imp rovem ents in p lantreliability should be reflected in increased capaci-

ty factors and availabilities and in decreasedforced outa ge rates. Industry efforts to improvecomponent reliability and enhance maintenanceand operation should start showing results soon.

Improvements in safety are more difficult tomeasure, but one indication of plant safety is thefreq uenc y and severity of eve nts that c ould leadto a c c idents. These a re known a s p rec ursorevents, and NRC requires that they be reportedroutinely. If there is a significant decline in the

number or severity of precursor events in thecoming years, it is likely that private and Govern-ment efforts are achieving some measure of suc-cess in increasing safety. Conversely, if incidentssuch as the loss of the emergency shutdown sys-tem a t the Salem nuc lea r plant co ntinue to oc -cur, it wil l be difficult to place much confidencein the effectiveness of the efforts to improve

safety.

It ma y be very diffic ult to achieve signific ant

gains in performance in an industry with so manydifferent actors and such diverse interests and

ta lents. The industry’s sup port for IN P<) is a ma - jor step in g ene rat ing a uniform level of excel-lence. However, only time will tell if INPO canrem ain bo th strong and ob jec tive and if all util-ities will commit themselves to high standards inconstruction and operation.

NEW INSTITUTIONAL ARRANGEMENTS

In the previous discussions, we saw that many be a significant step in this process of improve-utilities have built and operated nuclear power- ment. NRC also is encouraging quality in nuclearplants safe ly and reliab ly, and othe rs a re now power management as a way to achieve safety.working to improve the quality in construction However, all of these efforts from within and out-and op eration. The c rea tion of INPO a pp ea rs to side the utility industry may not be sufficient to



Ch. 5—Management of the Nuclear Enterprise . 135

provide assurances to the public and investors

that adequate levels of economy and safety arebeing achieved.

in the introduction to this report the many ac-tors and institutions involved in nuclear energywe re d esc ribe d. One of the keys to b rea king the

present impasse among these institutions is a cleardemonstration that all utilities with nuclear reac-tors are operating them safely and reliably. If theefforts to improve utility management describedthus far are insufficient to satisfy all these actors,it is unlikely that new plants will be ordered. Inthis case, a future for conventional nuclear powerma y req uire c hang ing the existing rela tionshipsamo ng the va rious ac tors or crea ting new insti-tutions.

It should be noted that the potential advantages

of these new entities are only speculative at thistime. Some industry problems such as the overallshortage of qualified personnel would not behelped by simply rearranging people and institu-tions. How eve r, if current e fforts to improve u til-ity management have little impact, these alterna-tives might b e w orth further co nside ra tion. The

various changes are discussed briefly below andthe implications of the changes are explored inc hap ter 9. Som e a re only inc rem enta l adjust-ments to the current structure of the nuclear in-dustry, while other are major reorganizations re-quiring legislation. However, they share the com-mon g oa l of imp roving overall ma nag eme nt of

both construction and operation of nuclear pow-erplants.

A Larger Role for Vendors inConstruction

Many of the current problems in plant con-

struction can be traced to the overlapping andconflicting authority of the utility, the nuclearsteam supply system (NSSS) vendor, the AE, andthe c onstructo r. To o verco me this prob lem , onec ontrac tor (prob a bly the NSSS vend or) co uld a s-sume greater responsibility for overall design,

and, in some cases, construction management.This c hang e a lrea dy is oc c urring to som e exte nt.

For example, Wedco, a subsidiary of Westing-house, has acted as the constructor for nuclearplants in New York State, and Westinghouse itselfis constructing a plant in the Philippines.

The trend tow ard grea ter vendor respo nsibili-

ty for construction management may be helpedindirectly by the current financial problems in thenuclear industry. Cost uncertainties make it un-likely that utilities will order new nuclear plantsunless they can be assured of a fixed price. If in-flation were more moderate and licensing uncer-

tainties reduced (perhaps through the use ofstandardized and preapproved designs), vendorsmight offer some type of fixed price as they didwith the turnkey contracts of the early 1960’s.However, it is unlikely that vendors would grantthis type of contract unless they were assured ofgreater control over construction. It has been sug-gested that a single person within the NSSS com-pany be given point-source responsibility for safe-ty, quality control, and construction managementof the nuclear island. Westinghouse assumesthese responsibilities in its international projects,(where licensing is less complex) and has hadgood experience with the approach. Because ithas greater control, the vendor is able to offerfixed-p rice c ontrac ts to its c ustom ers abroa d .

A greater role for vendors in construction man-agement offers a number of potential advantages

in ad d ition to tho se just describe d . The NSSScompanies have a long history of experience innuclear energy, highly trained personnel, and thefinanc ia l inc entive to build the p lant we ll. Themajor potential disadvantage of this approach istha t vend ors c urrently have little expe rience in

construction management. If the vendors do notbuild up their c onstruction c apab ilities, this ap -proach may not be an improvement over usinga q ua lified AE and c onstruc tor. Othe r prob lem scould arise after construction, when utilities withlittle knowledge of their plants must assume re-sponsibility for maintenance and operation.

Another approach to integrating responsibilitiesfor construction management is used in Belgiumfor all large construction projects, includingnuc lea r powe rp lants. There, the c onstruc tion

company assumes financial liability for thenuclear plant for a decade after it is completed.The c onstruc tion c om pa ny is a ble to a ssume thisrisk because it can purchase insurance after anindependent assurance company has certified thequality of its work.




Service Companies

Currently, nuclear consultants and service com-panies provide a broad range of services to util-ities, including: interactions with NRC; systems

de sign and engineering intergration; ope rationaland maintenance tasks; fuel procurement; and

qua lity a ssuranc e. These firms c an he lp streng th-en the capabilities of the weaker utilities in bothconstruction and operation of nuclear plants. Forexample, many utilities are now calling on con-sulting firms to conduct independent audits ofconstruction quality and make recommendationsfor imp rove me nts. Teled yne, Inc ., has aud ited the

two Midland units in Michigan and the two Dia-blo Canyon plants in California, C. F. Braun eval-uated the LaSalle generating station in Illinois, andthe Quadrex Corp. was called in to examine thetw o Sou th Texa s p lants (18).

While services such as these can be useful, theycurrently are provided only at certain times for

one o r more spec ific ta sks. To resolve safe ty a ndmanagement problems among the weaker uti l i-t ies, it ma y be c ome de sirab le for servic e c om pa -nies to p lay a muc h large r role. This might a lsobe attractive to a disaffected public l iving neara troub led nuc lea r p ow erplant. These roles c ouldrang e from ha nd ling a ll qua lity assuranc e o r allengineering work to actual management of con-struction or plant operations.

Currently, service companies belonging to util-ity holding companies such as Southern Co., Mid-

d le South Utilities, Inc ., Ame ric an Elec tric Pow erCo., Inc., and General Public Utilities Corp. actsomewhat like the service organizations dis-cussed above. For example, a centralized nuclear

eng inee ring g roup provide s servic es to a ll of M id -dle South’s member utilities. In the 1950’s, a con-sortium o f New Eng land utilities forme d YankeeAtomic Electric Co., which built and operatesYa nkee Row e in Ma ssac huset ts. Three o the r c or-porations, owned by many of the same utilities,were subsequently formed to build and operateVermont Yankee, Maine Yankee, and Connecti-c ut Yankee . The service d ivision o f Yankee

Atomic provide s a broad rang e o f servic es to a llof these plants (except Connecticut Yankee) andothers in New England. A more recent entrantis Fue l Sup p ly Servic e, a sub sidia ry of the suc -

c essful Florida Pow er & Light C o. This orga niza -tion has been hired by Public Service Co. of NewHam pshire to speed up the c onstruc tion of the

troub led Sea broo k projec ts (27).

While all of the entities just described are

owned by utilities, it is possible to envision others

that wo uld b e inde pe nde nt. A numbe r of busi-nesses might be interested in offering their serv-ices to utilities as nuclear operating companies.Duke Power Co., a successful nuclear utility, hasexpressed interest in op erating p lants for othe rutilities. Some present nuclear service ‘companiesalso might be interested, if it were clear that theywe re b eing g iven ma nage me nt responsibility. Thefundamental shift in the present relationship be-tween utilities and service companies would haveto be clarified for both parties. Finally, high-tech-nology companies, especially those already in-volved in the nuclear business, might want to pro-

vide operating services.Service co mp anies are co mmo nly used at G ov-

ernment-ow ned fac ilities. The suc c essful use o fc ontrac tors at armam ent p lants, whose o pe ra-tions involve careful attention to safety and quali-ty c ontrol, sugg ests tha t the c om p lexities of nu-clear powerplants co uld be handled b y an inde-pendent service company. It has been estimatedthat the Departments of Defense and Energy andthe Nationa l Aerona utics a nd Sp ac e Ad ministra-tion have contracts for Government-owned, con-tractor operated facilities amounting to $5 billionto $10 billion per yea r (29). An a na lysis of thesefacilities indicates that operations are most suc-cessful when either the owner or the contractorhas the dominant managerial and technical role.In addition, financial liability has not been a prob-lem in these contracts: all liability rests with thefacility owner, and the threat of replacement pro-

vides the incentive for quality operations by theservic e c om p an y. Suc h arran ge me nts a lso mightwo rk well in servic e c ontrac ts be twe en a utilityand a nuclear powerplant operat ing company.

The nuc lear servic e c om p an y alterna tive p ro-vides a way to pool nuclear expertise and make

it available to many utilities at once. During con-struction, the servic e c omp any c ould p lay thevital role of system integrator. In addition, im-

plementation of this alternative would be great-




Iy simplified by the fact that it does not require

a major change in the other institutions, such asthe utilities, NRC, or the vendors.

How eve r, the p rop osa l also has som e d isad -vantages. First, it seems unlikely that utilitieswould be willing to give up responsibility for safe-ty and quality while retaining financial liability.Sec ond ly, unless the roles of the ac tors we remade very clear, the arrangement could simplyadd to the confusion that already exists in nucle-ar powerplant construction and cause continu-ing disag reem ents. In a dd ition, d ep end ing onwhere the o wning utility and servic e c omp anywere head qua rtered, the arrange ment c ould

cause problems in dealing with State public Serv-ice Commissions. Finally, without some mecha-

nism that required the weaker utilities to hire serv-ice companies, the existence of these entitiesmight have litt le impact on the overall quality of

nuclear power management.

Certification of Utilities

An independent review and certification of util-

ities as capable of constructing or operating nu-Iear powerplants could provide the “stick”needed to make the service company alternativewo rk. NRC c urrently has the autho rity to revokethe o pe rating lic ense o f any utility the a ge nc yfeels is not c ap ab le of safely op erating nuclea rplants. However, since the utility industry rec-ognizes that the agency is very reluctant to takesuch drastic action, this authority may not be suf-ficiently convincing to assure high-quality opera-

tion of a ll nuc lea r plants. * Ce rtifica tion might pro-

vide a more politically feasible alternative. Itmight be more acceptable to the utilities becausethe certification review could come from an inde-pendent panel of experts, rather than from NRC.If certain management characteristics were foundto affect safety negatively, utilities with thosecharacteristics could be decertified until thosecharacteristics were changed.

Certification might involve periodic review ofutility management by an independent panel, in-

c luding representa tives of NRC and INFO. The

*The recent refusal by the Atomic Safety Licensing Board to grant

an operating license to Commonwealth Edison’s Byron plant mayindicate a change in NRC’s reluctance on this issue.

panels could be similar in makeup and activitiesto tho se used in a c c red itation of c olleg es anduniversities. Like accreditation panels, the review-ers could issue warnings and allow the utilitiestime to imp rove their ma nag eme nt prior to de-nying c ertific a tion. Bec ause o f the unique d iffi-culties of nuclear plant construction, the require-

ments for certification of utilities proposing tobuild new plants could be made particularly strin-gent. Based on the review panel’s findings, NRCcould either grant or deny the construction cer-t i f icate.

Once a plant is completed, another review bythe panel could determine the company’s abil i-ty to manage it. Depending on the results of thereview, the co mp any might b e required to hire

an outside service company to take over or sup-plem ent o p erations. Therea fter, the utility and / orthe service co mpa ny could b e reviewed period-ically to make sure that changes in personnel hadnot diminished their management capabilities.Utilities presently operating nuclear plants alsowo uld be subjec t to the c ertif ic ation review . Onemodel of such a review- and certification-processis the accreditation procedure for utility trainingprograms currently being developed and imple-mented by IN PO,

The p rima ry ad vanta ge of a c ertif ic ation proc -ess is that it could force the weaker utilities toimprove their nuclear management capabilit ies,

obtain independent and external expertise, or re-fra in from ente ring the nuc lear p ow er business.

Such a step wo uld b e very co nvinc ing to thepub lic a nd skep tics of nuc lea r pow er. The p ri-ma ry ba rrier to imp lem enta tion is tha t the utilityindustry would be reluctant to accept it. Nuclearutilities alrea dy fee l overburdened by NRC a ndINPO inspe c tions, and the c ertific ation p ane l’sreview co uld a dd yet anothe r layer of “regula-tion. ” Another disadvantage of certification is thatits success depends on the existence of an entity(e.g., a service company) which has the exper-tise the utility lacks. Unless such entities are avail-

able and have the appropr iate management char-acteristics, the certification procedure will not im-prove the construction and operation of nuclearpowerplants.



138 qNuclear Power in an Age of Uncer ta in ty

Privately Owned Regional NuclearPower Company

Since the 1920’s, elec tric utilities ha s bec om e

increasingly coordinate d through horizonta l in-teg ra tion a nd p ow er poo ling. This trend ha s c ap -tured economies-of-scale, fostered the sharing of

expertise, and eased the task of planning (31). Theregional nuclear power company (RN PC) dis-cussed here is one approach to increased integra-tion that does not involve restructuring the wholeindustry. It is seen as a logical extension of thecurrent trend toward multiple utility ownership

and single utility ma nage me nt of m any existingnuclear plants.

The RNPC w ould b e c reated exp ressly to fi-nance construction and/or operate nuclear pow-erplants. It c ould b e o wned by a c onsortium o futilities, vendors, and AEs, and might place anorder for several plants at once, based on a stand-

ardized, preapproved design. All RNPC proposalscurrently under discussion call for a confined sit-ing p olicy to take ad vantag e of the b enefits of

clustered reactors. While some analysts feel theRNPC should b e a pplied o nly to ne w c onstruc-tion, others think that existing plants could be

transferred to RNPC authority. Federal legislationproba bly would b e required to transfer owner-ship of existing plants because of the tax andfinanc ial c om plic at ions (10)0

One possible advantage of an RNPC from a fi-nanc ia l perspe c tive is tha t its elec tric ity output

might not b e subjec t to some of the diffic ultiesposed by Sta te ra te reg ulation d isc ussed inchapter 3. Presumably, the RNPC would sellpow er to the utility grid a t who lesale rate s, andthe utilities in turn would distribute the powerto their customers. Interstate wholesale rates areregulated by the Federal Energy Regulatory Com-mission (FERC). With appropriate legislation, thepow er generated by a n RNPC c ould be deregu-late d totally or granted spe c ial trea tment in rate-making. Congress could exempt the RNPC fromFERC p ric e reg ulation, and the elec tric ity pur-c hased by loc al utilit ies c ould b e e xempte d from

State rate regulation when sold to customers. ThePub lic Utility Reg ulato ry Polic ies Ac t (PURPA)provides the precedent of a special pricing mech-anism, legislated by Congress, for a particular

c lass of e lectricity (in tha t c a se, p ow er from sma llprod uc ers). The law ha s be en up held a s c onstitu-tional over ob ject ions from Sta te g ove rnme nt.

While the initia l attrac tion of the RNPC m od elma y b e the se financial be nefits, such c omp aniesalso could be expected to improve nuclear powermanagement by their larger staffs, allowing agrea ter co nc entrat ion of expe rtise. The p rop osedchange would leave nonnuclear utility operationsuntouched, and would avoid the complicat ionsof mixed financial liability and authority in theservice company scheme. In addition, the greaterexpertise o f the larger co mp any c ould ma ke itless reliant on vendors and AEs during construc-tion.

The size o f the RNPC c ou ld p rove a s muc h adisadvantage as an advantage. A bureaucrat icop eration c ould d ec rea se the sense o f individua lresponsibility for the reactors, which in turn could

lead to a decrease in safety and reliability. Addi-tionally, while shared utility ownership of theRNPC could help share the financial risks and

burdens, it might be difficult to obtain financingfor a com pa ny whose only assets were nuc learpowerplants. in the past 2 years, the utilities own-ing the Yankee nuclear corporations in New Eng-land have had to b ac k financ ial offerings withtheir full utility assets.

Government-Owned RegionalNuclear Power Authority

This op tion is basica lly the same as tha t justde sc ribe d, excep t that the Go vernment wo uldeither ow n the entity or p rovide fina nc ial assist-anc e to it. The Fed eral Government has previous-ly assume d this role in the c reation o f the Ten-

nessee Valley Autho rity and the Bonne ville Pow erAdministration (B PA) to tap hydropower. OntarioHydro in Cana da and TVA, which ha ve succ esful-Iy built and operated nuclear plants, are thec losest models to such an authority. However,both of these entities own nonnuclear as well asnuc lea r p ow er. The RNPA envisioned here w ould

be involved only in nuclear projects,Several factors would justify the creation of one

or more Government-owned RNPAs. First, it maybe the only way to maintain nuclear power as




a national energy option. Given utilities’ currentreluctance to order new nuclear plants, the Fed-eral Government might use the RNPA as a vehi-cle to demonstrate that newly designed, stand-ardized plants could be built and licensed eco-nomically. Second, because of nuclear power’sunique characteristics, the Federal Government

has always had a major role in the developmentand regulation of this technology, and Gov-ernment ownership might be a logical extensionof that role. Finally, the advantages of large-scaleoperations cited for the privately owned regionalutilities would apply to RNPAs as well.

The primary barrier to creation of a Govern-ment-owned RNPA is that it involves greater Gov-

CHAPTER 5

1. Abel, P., “Meeting With Washington Public PowerSup ply System Personn el, Ma r. 15, 1983,” OTAmemo, Mar. 24, 1983.

2. Bak er, J., and Olsen, K., “ Oc c upa tional Emp loy-ment in Nuclear-Related Activities, 1981 ,“ OakRidg e Assoc iate d Unive rsities for the U.S. Depart-ment of Energy, ORAU-197, April 1982.

3. Bauma n, D. S., et al., “ An Ana lysis of Pow er PlantConstruc tion Lea d Time s, Volume 1: Analysis and

Results,” Applied Decision Analysis, Inc. for Elec-

tric Pow er Resea rch Institute, EPRI EA- 2880,February 1983.

4. “Cincinnati G & E Bows to Stop-Work Order,”Nucleonics Week, vol. 23, No. 46, Nov. 18, 1982.

5. “Commissioners Okay Salem-1 Restart After Set-ting Some Prerequisites,” Nucleonics Week, vol.

24, No. 17, Apr. 28, 1983.6. “ Doing It Righ t in the Nuc lea r Business,” The

Energy Daily, vol. 11, No. 14, Jan. 20, 1983.7. Friedlander, G. D., “ Coal or Nuclear? Eco nomics

Govern,” Electrical World, vol. 195, No. 10, Oc-

tober 1981.8. Garrick, B. J., “Advances in Using Probabilistic

9

Risk Assessme nt a s a Risk Ma nage me nt Too l,”presented at the Nuclear Power Reactor Safety

Course, Massac husetts Institute of Tec hno log y,July 18, 1983.Glassco ck, R., “ Respond ing to Quality Conc erns,”

presented at the Atomic Industrial Forum Work-shop on Achieving Excellence in the Construction

ernment intervention in the private sector. How-ever, this problem could be alleviated if the RNPA

were primarily a financing entity, assisting privateutilities with construction of new nuclear plants.Another model would be RNPA ownership of theplants, with operations handled by private serv-ice companies. A Government-owned RNPA

could also face the problem of public oppositionencountered by the RN PC. In addition, there isno guarantee that management by the FederalGovernment would be superior to private-utilitymanagement. Finally, if existing plants were tobe included in the RNPA scheme, the transfer ofthese plants to Government ownership could bedifficult.

10 .

11.

12.

13.

14.

15.

16.

17.

and Operation of Nuclear Power Plants, San Fran-cisco, Jan. 17, 1983.

Go lub, B., and Hyma n, L., “The Financ ial Dif-ficulties and Consequences of DeregulationThroug h Divestiture,” Public Utilities Fortnightly,Feb. 17, 1983.

Institute of Nuc lea r Power Op erations, “Forem ostSafety and Reliability Issues,” December 1981.Institute of Nuclear Power Operations, “1982Survey of Nuclear-Related Occupational Employ-

ment in U.S. Electric Utilities, ” December 1982,

INPO Report 82-031.Jorda n, E. L., “ NRC’s Design and Construc tionQuality Assurance Initiatives,” presented at the

Atomic Industrial Forum Workshop on AchievingExce llenc e in the Construction a nd Operation ofNuclear Power Plants, San Francisco, Jan. 17,1983.

Kelley, D., “Electric Utility Industry; NuclearPlants–Another Look,” Merrill Lynch, May 1982.Kemeny, J. G., “Report of the President’s Com-mission on the Ac cid ent at Three Mile Island ,” Oc -

tober 1979, Washington, D.C.Komanoff, C., “ The Bottom Line–Nuc lear Ca -pacity Factors: Another Bad Year,” Critical Mass,April 1983.Martel, L., Minnek, L., and Levy, S., “Summaryof Discussions With Utilities and Resulting Conc lu-

sions, ” Electric Power Research Institute, June1982.




18.

19.

20,

21 ,

22 ,

2 3 ,

24.

25 .

26,

27 .

28 .

29 ,

30 .

31

MHB Tec hnic a l Assoc iate s, “Issues Affec ting theViability and Acceptability of Nuclear PowerUsage in the United States,” Dec. 2, 1982.Nero, A., A Guideb ook to Nuc lear Reac tors,

University of California Press, Berkeley, 1979.“NRC Pulls the Plug on Zimmer Nuclear Power

Plant,” The Energy Dai/y, vol. 10, No. 215, Nov.16, 1982.

“ Nuc lea r Safe ty Fines,” Washington Post, No,159, May 13, 1983.Office of Tec hno log y A ssessme nt, U.S. Congress,Nuclear Pow er pl ant Standardization–Light Water

Rea c tors, OTA- E-1 34, Jun e 1981.Offic e of Tec hnolog y Assessme nt Workshop ,Washington, D. C., Dec. 9-10, 1982.Partlow, j., U.S. Nuclear Regulatory Commission,personal communication, Feb. 3, 1983.Peterson, E., Commonwealth Edison, Co., per-

sonal communication, August 1983.Procter, M., “ Trip Rep ort-Visit to INPO o n Jan.

26, 1983,” OTA m em o, Feb . 15, 1983.“ PS New Ham pshire Hires Florid a P&L Sub sid iary

to Help Bring Seabrook on Line,” E/ectric L)ti/ity,Aug. 1, 1983.“ Revamp ed South Texas Projec t Show ing StrongSigns of Recovery,” N u c l e o n i c s Week, vol. 24,No. 12, Mar. 24, 1983.

Ring, Leon, “Use of Nuclear Service Companies,”An Ac ce pta ble Future Nuclea r Energy System,

Firebaugh and Ohanian, (eds.) Oa k Ridg e, Term.:Institute for Energy Analysis, 1980.

Stevenson, J. D., and Thoma s, F. A., Se/ ec tedReview of Foreign Licensing Practices for Nuclear

Powe r P/ ants, NUREG / C R- 2664.

Theod ore Barry & Assoc iate s, A Study ofAg gregate

Alternatives in the U.S. Electric Utility Industry,

DOE-RG-1O295-1, U.S. Department of Energy,1982.

32.

33.

34.

35<

36.

37.

38.

3 9

40 <

41

Uhrig, R., Lette r to the Offic e o f Tec hno log y As-sessment, Aug. 1, 1983.United Engineers & Constructors, Inc., Case Study

Reports for the U.S. Department of Energy, 1981.

United Engineers & Constructors, Inc., “Phase IVFinal Report and Fourth Update of the EnergyEconomic Data Base (EEDB) Program,” prepared

for the U.S. Department of Energy, [JE&C-ANL-

810930, COO-641 1-1, September 1981.United Engineers and Constructors, Inc., “Regu-latory Reform Case Studies (Piping) -Draft,” UE&C

820212, February 1982.

U.S. Department of Energy, “How to ProlongSteam Generator Life in Nuclear Plants,” Update

—Nuclear Power Program Information and Data,

Feb rua ry 1983, DOE/ NE-0048 11.U.S. Nuclear Regulatory Commission, “A Survey

of Senior NRC Management to Obtain Viewpointson the Safety Impact of Regulatory Activities From

Representative Utilities Operating anc { Construc-ting Nuc lea r Power Plants,” NUREG-0839, August1981.

Weitzberg, A., et al., “Reactor Quality—FactorsAffecting Managerial and Institutional Compe-

tenc e of Utility Nuc lear Op erations, ” NUS Corp.,Jan. 15, 1983.Wilkinson, E., “ lNPO/ NRC Interactions, ” pre-sented at the American Nuclear Soc iety Topic alMeeting on Government and Self-Re,gulation ofNuclear Power Plants.“World List of Nuclear Power Plants,” Nuc/ear

News, vol. 26, No. 2, February 1983.Zebroski, E., “Analysis of Operating IEvents andPriorities,” presented at the America n NuclearSoc iety Top ica l Meeting o n Government and Self-

Regulation of Nuclear Power Plants,



Chapter 6

The Regulation of Nuclear Power

Photo credit: Florida Power & Light

One of aisles in the document vault at St. Lucie 2



Contents

Page

Federal Regulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ................143The Role of Eme rge nc y Planning. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ......150

State and Local Regulation . . . . .Need for Power . . . . . . . . . . . .Environmental Policy . . . . . . . .State Siting Activities. . . . . . . . .

Issues Surrounding Nuclear Plant

Backfitting . . . . . . . . . . . . . . . . . . .

Specific Concerns . . . . . . . . . . .

150. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 1

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 1

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153

Regulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .153

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 154

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155

Proposals for Change and Evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157

Hearings and Other NRC Proc ed ures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ......160Hearings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ...160Proposals for Change in the Hearing Process and Evaluation . . . . . . ...............161

The Two -Step Lic ensing Proc ess . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ...............168

Other NRC Responsibilities.

Safety Goals . . . . . . . . . . . . .NRC Safety Goal Proposal

17 1

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 172

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 172

Licensing for Alternative Reactor Types . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .......174

Chapter 6 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..................175

Table

Table No.22. State

Figures

Figure No.32. NRC

Page

Siting Laws . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .................152

Page

Responsibilities in Nuclear Powerplant Licensing . . . . . . . . . ................14633. Utility Responsibilities in Nuclear Powerplant Licensing and Construction . . . . .. ...147



Chapter 6

The Regulation of Nuclear Power

Nuclear power is one of the most intensivelyregulated industries in the United States, and thescope and practice of regulation are among themost volatile issues surrounding the future of nu-

clear power. Strong—and usually conflicting—opinions abound among the participants in thenuclear debate on whether the current regula-tory system is adequate to ensure safe and reliablepowerplants or is excessive, and whether it is en-forced adequately or is interpreted too narrowly.

Every aspect of the nuclear industry—from theestablishment of standards for exposure to radia-tion to the siting, design, and operation of nuclearpowerplants and the transportation, use, and dis-posal of nuclear materials–is regulated at theFederal, State, or local level. In general, the Fed-

eral Government retains exclusive legislative andregulatory jurisdiction over the radiological healthand safety and national security aspects of theconstruction and operation of nuclear reactors,while State and local governments share the reg-ulation of the siting and environmental impactsof nuclear powerplants and retain their traditionalresponsibility to determine questions of need forpower, reliability, user rates, and other relatedState concerns.

This chapter describes the existing regulatoryprocess at the Federal, State, and local levels;

reviews the various criticisms of that process rais-ed by the different parties in the nuclear debate;and discusses proposals for substantive and pro-cedural changes in nuclear power regulation. The

chapter focuses on the health and safety and en-vironmental regulation of nuclear powerplants;financial and rate regulation are discussed inchapter 3.

It should be emphasized that this chapter pri-marily reports on the existing regulatory processand on proposals for changes in that process.Arguments for and against the existing system andproposed changes are presented as they appearin the literature or as OTA determined them inthe course of this study. Such criticisms of the reg-ulatory system can reflect the biases and vested

interests of the commentators. In light of this, itis important to examine the arguments criticallyfrom a safety and efficiency perspective. WhereOTA found sufficient documentation to supporta particular argument, the basis for the conclu-sions is identified. In instances where OTA couldnot make such a determination, the argumentsare presented without conclusions to illustratethe scope of the controversy and the wide diver-gence among the parties’ perceptions of the cur-rent role of regulation and of the need forchanges in the regulatory system.

FEDERAL REGULATONThe primary forms of regulation under the

Atomic Energy Act (see box D) are: 1 ) the issu-ance of licenses for the construction and opera-tion of reactors and 2) inspection and enforce-ment to ensure that nuclear plants are built andoperated in conformance with the terms of a li-cense. This section describes the licensing proc-ess that was put in place during the 1970’s whenthe last group of plants received construction per-mits. This is precisely the licensing process that

has been the target of so much criticism by thenuclear industry, utilities, nuclear critics, and reg-ulators. In addition, this section discusses the wayin which this licensing process might operate inthe current climate. Although the basic regula-tions have not changed substantially since the1970’s, the way those regulations are applied toconstruction permits or operating licenses mightbe very different if an application were filedtoday.

143



144 q Nuclear Power in an Ag e of Uncertainty



Ch. 6—The Reg ulat ion of Nuclear Power q 145

In the 1970’s, a utility would undergo an ini-tial planning phase before it would apply to theNuclear Regulatory Commission (NRC) for a con-struction permit, It would select a site in accord-ance with NRC (and State and local) policies andguidelines; choose an architect/engineering (AE)firm; solicit bids for the nuclear steam supply sys-

tem (NSSS) and the balance of the plant; awardcontracts; and assemble data to be submitted toNRC with the construction permit (CP) applica-tion. During this planning phase, the utility alsowould ensure compliance with State and locallaws and regulations, which could require a vari-ety of permits for approval of the facility.

The utility then would file an application fora CP, as indicated in figures 32 and 33. The ap-plication would include: 1 ) a Preliminary SafetyAnalysis Report (PSAR) that presents in generalterms the plant design and safety features and

data relevant to safety considerations at the pro-posed site; 2) a comprehensive EnvironmentalReport (ER) to provide a basis for the NRC evalua-tion of the environmental impacts of the pro-posed facility; and 3) information for use by theAttorney General and the NRC staff in determin-ing whether the proposed license would createor maintain a situation inconsistent with the an-titrust laws.

NRC regulations require the antitrust informa-tion be submitted at least 9 months but not morethan 36 months prior to the other portions of the

CP application. A hearing might be held at thecompletion of the antitrust review, but it wouldnot be mandatory unless requested by the Attor-ney General or an interested party, The NRC alsomust make a finding on antitrust matters in eachcase where the issue is raised before the Com-mission.

Upon receipt of a CP application, the NRC staffwould review it to determine if it is completeenough to allow a detailed staff review, and re-quest additional information if necessary. The ap-plication would be formally “docketed” when

it met the minimum acceptance criteria.In the past, the PSAR included very incomplete

design information (only 10 to 20 percent in some

cases). Most parties in the nuclear debate agree

that many of the construction problems evidentin today’s plants could have been prevented ifmore complete designs had been available dur-ing CP review. I n recognition of this argument,NRC officials have indicated that they now wouldrequire an essentially complete design with a CPapplication, a move that has widespread support.

In the next step of the process, the NRC Of-fice of Nuclear Reactor Regulation would com-pare the details of the permit application with theNRC’s Standard Review Plan (SRP) and usuallywould submit two rounds of questions to the ap-plicant. These questions often would result inchanges in the plant design. The staff then wouldprepare a Safety Evaluation Report (SER) docu-menting the review and listing “open issues, ”which are changes dictated by NRC but disputedby the applicant. Concurrent with the prepara-tion of the SER, the Advisory Committee on Reac-

tor Safeguards (ACRS) would review and com-ment on the application, and the NRC staff couldissue supplements to the SER to respond to issuesraised by ACRS or to add any information thatmay have become available since issuance of theoriginal SER. During the 1970’s, this review proc-ess culminating in SER might have taken 1 to 2years. The review period could potentially beshortened if an application were filed now withessentially complete design information or astandardized design. Detailed design informationwould be likely to meet the minimum criteria foracceptance of the application with little delay.

A standardized design could indirectly acceleratethe process even more because it is unlikely thatmany new questions would be raised by theACRS or about the SRP after approval of the firstplant using that design.

During this period, the NRC staff also wouldbe reviewing the proposed plant’s environmen-tal impacts and preparing a draft EnvironmentalImpact Statement (E IS) to be issued for review bythe relevant Federal, State, and local agencies andby interested members of the public. After com-ments on the draft EIS were received and any

questions resolved, the staff would issue a finalEIS.

Soon after a CP application was docketed, NRCwould issue a notice indicating that it would hold



146 q Nuc lear Power in an Age of Uncertainty

Figure 32.-NRC Responsibilities in Nuclear Powerplant Licensing

Antitrust decision

Antitrust hearing

Antitrust review

Antitrustreview

Operating licenseissued

Operating license

Hearing (if required)—

R e v i e w b y A C R S —

NRC safetyevaluation

Radiological safetyreview (based onFSAR)

Review by

INRC safety

evaluation1

I acceptance review

I(based on CP

I

I

1

I

I

I

II

I Radiologicalsafety review

I

II

II

I

Final environmental I

IDraft environmental

IEnvironmentalacceptance review I

I

II

QA construction andtesting inspections

QA program review

QA programreview



Ch. 6—The Regulation of Nuc lear Power q147

Figure 33.—Utility Responsibilities in Nuclear Powerplant Licensing and Construction

II

III

II

I

I

I

I

I

I

I

Update environmentreport and file FSA

Onsite QA program

State water quality Icertificate ion

File State PUC File CP applicationapplication

State review

I

II

I

I

Prepare PSAR andprovide site inform,to NRC

Prepare environmental

Develop QA programprovide antitrustinformation to NRC

NRC review

— Begin commercial operation

— Initiate power operation test

— Complete construction, load

program

fuel

—Start plant construction

—Start site preparation

—Make construction arrangements

—Design plant

Issue reactor specifications

Construction-related activities




a hearing on safety and environmental issuesraised by the application. Interested parties couldprovide written or oral statements to the three-member Atomic Safety and Licensing Board(ASLB) as limited participants in the hearings, orthey could petition for leave to intervene as fullparticipants, including the right to cross-examine

all direct testimony in the proceeding and to sub-mit proposed findings of fact and conclusions oflaw to the hearing board.

NRC regulations provide an opportunity at anearly stage in the review process for potential in-tervenors to be invited to meet informally withthe NRC staff to discuss their concerns about theproposed facility. This provision has not beencommonly used; as a result, the safety concernsof the critics have not been considered serious-ly and formally until the hearings. The problemwith this timing is that it places the critics in theposition of attempting to change or modify a deci-sion that already has been made rather than in-fluencing its formulation.

The environmental hearings could be con-ducted separately to facilitate a decision on aLimited Work Authorization (LWA) or could becombined with the safety hearing. The SER andany supplements to it plus the final EISwould bethe major pieces of evidence offered by the NRCstaff at the hearing. The ASLB would consider allthe evidence presented by the applicant, the staff,and interveners, together with proposed findingsof fact and conclusions of law filed by the parties,

and issue an initial decision on the CP. ASLB’sinitial decision would be reviewed by the AtomicSafety and Licensing Appeal Board (ALAB) on ex-ceptions filed by any party to the proceeding or,if no exceptions were filed, on ALAB’s own in-itiative (“sua sponte”). Since Three Mile Island,all ASLB decisions must be approved by NRCbefore they take effect. NRC also considers peti-tions for review of appeals from A LAB decisions.

NRC regulations provide that the Director ofthe NRC Office of Nuclear Reactor Regulationmay issue an LWA after ASLB has made all of the

environmental findings required under NRC reg-uIations for the issuance of a CP and has reason-able assurance that the proposed site is a suitablelocation from a radiological health and safety

standpoint, and after Commission approval. A li-censing board may begin hearings on an LWAwithin a maximum of 30 days after issuance ofthe final EIS.

When construction of a plant had progressedto the point where final design information* andplans for operation were ready, an applicationfor an operating license (OL) would be prepared.The OL process has been very similar to that fora CP. The applicant would submit a Final SafetyAnalysis Report (FSAR), which sets forth the perti-nent details on the final design of the facility, in-cluding a description of the containment, thenuclear core, and the waste-handling system. TheFSAR also would supply information concerningplant operation, including managerial and admin-istrative controls to be used to ensure safe opera-tion; plans for preoperational testing and initialoperations; plans for normal operations, includ-ing maintenance, surveillance, and periodic test-ing of structures, systems, and components; andplans for coping with emergencies. The applicantalso would provide an updated ER. Amendmentsto the application and reports could be submittedfrom time to time. The staff would prepare an-other SER and EIS and, as at the CP stage, ACRSwould make an independent evaluation and pre-sent its advice to NRC by letter. The ASLB wouldalso review the OL application and issue a deci-sion. Until recently, the ASLB has granted all re-quests for OLS. However, in January 1984 theASLB refused to grant an OL to CommonwealthEdison Co. for the two-unit Byron station. As inthe procedure for a CP, this decision will be re-viewed by the ALAB and the Commission.

A public hearing is not mandatory prior to is-suance of an OL. However, soon after accept-ance of the OL application, NRC would publishnotice that it was considering issuing a license,and any person whose interest would be affectedby the proceeding could petition NRC to holda hearing. The hearing would apply the same ad-

judicatory procedures (e.g., admission of partiesand evidence, cross-examination) and decisionprocess that pertain to a CP.

*The final design illustrates how the plant has been built and thus

reflects all amendments and variances issued and backfits orderedby NRC since the CP.



Ch. 6—The Regulation of Nuclear Power q 149

Photo credit: Nuclear Regulatory Commission

The members of the Nuclear Regulatory Commission are meeting with the licensing staff of the NRC to review an

upcoming operating license. The Commission’s meetings are open to the public

A stated goal of NRC (under normal circum-stances and barring any important new safety

issues) is to conclude ACRS and Office of NuclearReactor Regulation reviews and the hearing proc-ess before the utility completes construction ofthe plant. Current NRC regulations authorize thestaff to issue an OL restricted to 5-percent pow-er operation; full power operation must be ap-proved by the Commissioners themselves. Uponreceipt of the low-power OL, the utility couldbegin fuel loading and initial startup. The plantthen would have to undergo extensive testingbefore it could begin commercial operation.Through its inspection and enforcement program,NRC maintains surveillance over construction

and operation of a plant throughout its servicelife. As discussed in chapter 5, this surveillanceis intended to assure compliance with NRC reg-

ulations designed to protect the public health andsafety and the environment.

Other Federal agencies with statutory or regu-latory authority over some aspects of the con-struction and operation of nuclear powerplantsinclude: Environmental Protection Agency, U.S.Army Corps of Engineers, National Oceanic andAtmospheric Administration, Department of En-ergy, Department of the Interior, U.S. GeologicalSurvey, Department of Agriculture, Departmentof Housing and Urban Development, AdvisoryCouncil on Historic Preservation, Department ofTransportation, Federal Aviation Agency, Depart-ment of Defense, Council on EnvironmentalQuality, River Basin Commissions, and Great Lakes

Basin Commission. These agencies review, com-ment on, and administer specific issues undertheir jurisdiction.

25-450 0 - 84 - 11 : QL 3




The Role of Emergency Planning

NRC requires license applicants and licenseesto specify their plans for coping with emergen-cies. NRC regulations established in 1980 specifyminimum requirements for emergency plans foruse in attaining “an acceptable state of emergen-

cy preparedness”, including information aboutthe emergency response roles of supporting orga-nizations and offsite agencies (16). For plants juststarting construction, these plans have to bestated in general terms in the PSAR and submittedin final form as part of the FSAR. Detailed pro-cedures for emergency plans would have to besubmitted to NRC no less than 180 days prior tothe scheduled issuance of an OL. Licensees withoperating plants in 1980 were also required tosubmit detailed emergency plans to comply withpost-TMl regulations.

NRC regulations specify a broad range of in-

formation that must be included in emergencyplans. Utilities must develop an organizationalstructure for coping with radiological emergen-cies and define the authority, responsibilities, andduties of individuals within that structure as wellas the means of notifying them of the emergen-cy. Second, the utility must specify the criteriaon which they determine the magnitude of anemergency and the need to notify or activate pro-gressively larger segments of the emergency or-ganization (including NRC, other Federal agen-cies, and State and local governments). Third, theutility has to reach agreements with State andlocal agencies and officials on procedures fornotifying the public of emergencies for publicevacuation or other protective measures and forannual dissemination of basic emergency plan-ning information to the public. Fourth, programsmust be established to train employees and otherpersons to cope with emergencies, to hold peri-odic drills, and to ensure that the emergency planand its implementing procedures, equipment,

and supplies are kept up to date. Finally, the util-ity must develop preliminary criteria for deter-mining when, following an accident, reentry ofthe facility would be appropriate and whenoperation could be resumed.

The role of emergency planning has becomeincreasingly controversial since the accident atThree Mile Island. Local governments must par-ticipate in the preparation of the emergency planand reach agreements with the utility on publicnotification, evacuation, and other procedures,and they may intervene in the hearings on theplan. This is the principal leverage a local govern-ment has over the operation of a nuclear plant,and it can hold up the issuance of an OL. For ex-ample, at the Shoreham nuclear powerplant, sig-nificant differences in scope between the emer-gency plan proposed by the utility and that de-veloped by the county are the primary issue thatmust be resolved before the utility can obtain anOL. There is a possibility that those differencesmight not be resolved. The adequacy of the util-ity’s emergency planning also has become anissue at the Indian Point Station due to its prox-imity to densely populated New York City.

Such situations are of great concern to nuclearutilities, their investors, and the surrounding com-munities. If emergency plans are not developedin a timely and satisfactory fashion, the plantowners will not be granted a license to operatetheir plant. Moreover, emergency planning prob-lems are difficult to anticipate, and their resolu-

tion is not necessarily assured by prudent man-agement. Thus, they tend to increase the uncer-tainties associated with nuclear plant schedules.This source of uncertainty might be eliminatedif final approval of emergency plans were re-quired much earlier in the licensing process oreven as a condition of State issuance of a Certifi-cate of Public Convenience for a nuclear plant.

STATE AND LOCAL REGULATIONA wide range of State and local legislation, reg- During the last decade, more and more States

ulations, and programs affect the licensing, con- moved to a more thorough consideration of needstruction, and operation of nuclear powerplants. for power and choice of technologies, environ-




mental policy, and energy facility siting. Table 22identifies the various State authorities in theseareas. In most cases, NRC requires State approv-als to be obtained before the Commission willtake any action on a CP or OL application.

Several States (e.g., California, Oregon, Ver-

mont, Wisconsin) also have enacted special re-strictions on the construction of nuclear power-plants on the basis of economic, environmental,or waste-disposal considerations. The U.S. Su-preme Court recently upheld State authority torestrict nuclear power development when it ruledin favor of California’s siting law, which bans newnuclear powerplants pending a method to dis-pose of nuclear waste. The Court held that theAtomic Energy Act does not expressly require theStates to construct or authorize nuclear power-plants or prohibit the States from deciding, as anabsolute or conditional matter, not to permit the

construction of any further reactors.The State regulation of environmental and siting

issues discussed below adds to the complexityand uncertainty in planning and licensing nuclearpowerplants. In a State with multiple layers ofreview within numerous agencies, dozens oreven hundreds of State approval steps may beinvolved. However, the State regulatory processgenerally has been far less difficult to managethan the Federal regulatory aspects related spe-cifically to nuclear health and safety.

Need for PowerPrimary responsibility for regulating electric util-

ities has been vested for many decades in Statepublic utility commissions (PUCs). PUCs set rateschedules designed to meet the cost of serviceand to earn the utility stockholders an appropriatereturn on investment, as discussed in chapter 3.Many PUCs also approve financing for new facili-ties deemed necessary to supply service and issueCertificates of Public Convenience and Necessi-ty (CPCN), which certify that when the facilitygoes into service the capitalized cost will be

added to the rate base.Although the procedures for determining need

for power and issuing a CPCN vary from Stateto State, no utility will proceed beyond engineer-ing to construction without a CPCN or an equiva-

Ient guarantee that it will be allowed to earn areturn on its investment. Furthermore, it is un-likely that a utility would apply to NRC for a CPwithout already having obtained a CPCN or atleast being confident of receiving it.

Environmental PolicySeveral Federal statutes, including the National

Environmental Policy Act (N EPA), the Clean AirAct (CAA), the Clean Water Act (CWA), and theHousing and Urban Development (HUD) 701Comprehensive Planning Assistance program,emphasize the State role in regard to environ-mental issues. Moreover, many States have en-acted their own environmental legislation. Thus,the same environmental aspects of a proposednuclear powerplant often are reviewed by boththe State and the NRC, and in some cases, jointNRC-State hearings may be held on matters of

concurrent jurisdiction.Trad itiona lly the State s have been responsible

for land use, and many States have comprehen-sive land-use planning programs. Energy facilitysiting also can be affected by States, local govern-ments, or regional organizations through compre-hensive planning activities under federally ap-proved coastal-zone management programs andunder the HUD program.

State water management agencies must ap-prove a proposed nuclear powerplant, examin-ing issues related to both the quality and quanti-ty of water supply and to effluent discharge limita-tions. The States have programs to review waterwithdrawals from streams and structures placedin water, and they issue Water Quality Certificatesunder CWA, which include any effluent limita-tions, monitoring, or other requirements neces-sary to assure that the plant will comply with ap-plicable Federal and State water-quality stand-ards. These conditions become part of the NRCpermit or license. In addition, if a nuclear facili-ty will discharge into navigable waters, it mustobtain a National Pollutant Discharge Elimination

System (NPDES) permit. CWA establishes specialprocedures for NPDES permits dealing with ther-mal discharges.

Nuclear plants can have air-quality impactsfrom cooling tower plumes, but neither the States




x x

w

x x x x x x x

x

. .

. .

. .

. .

. .

. .

. .

. .

. .

. .

. .

. .

. .

. .

. .

. .

. .

. .

. .

. .

. .

. .

. .

. .

. .

. .

. .

. .

. .

. .

. .

x x x




nor the Federal Government have standards gov-erning such emissions. Rather, the primary effectsof CAA and State air-quality programs under thatact are through restrictions on the siting of andemissions from fossil-fueled plants, which may in-crease the attractiveness of the nuclear option forelectric utilities.

State Siting Activities

ety of State agencies, each concerned with a sep-arate aspect of the construction or operation ofa plant; State licensing through a “one-stop”agency charged with determining the suitabilityof all aspects of a particular site on behalf of allState regulatory bodies; or State ownership of thesite, with a single agency empowered to admin-

ister the terms of a lease with the utility or con-sortium that owns the plant.

Twenty-five States currently have siting laws.These include “multistop” regulation by a vari-

ISSUES SURROUNDING NUCLEAR PLANT REGULATION

For the last decade, nuclear plant regulationhas been slow, unpredictable, expensive, andfrustrating for many involved in licensing. More-

over, it has failed to prevent accidents such asthose at Three Mile island and Browns Ferry aswell as construction problems like those experi-enced at Diablo Canyon and Zimmer. Even inthe case of the Byron plants where the OL wasdenied by the ASLB, the problems were not actedon until the two plants were nearly complete. Thefrustrations, costs, and uncertainties have resultedin extensive criticisms of the regulatory processand a variety of proposals for changes in thatprocess. The focal points for such criticisms arebackfitting, * hearings and other NRC procedures,

the current two-step licensing process, NRC re-

sponsibilities not directly related to safety, the useof rulemaking, and safety goals.

The principal concerns about nuclear plant reg-ulation expressed by utilities and the industry arethat neither the criteria nor the schedules for sit-ing, designing, building, and operating nuclearplants are predictable under the current licens-ing scheme. The industry and some regulatorsalso complain about the extensive opportunitiesfor public participation in licensing, arguing that

*Although “backfitting” technically refers only to design or reg-

ulatory changes ordered by NRC during plant construction, and“ratcheting” to changes imposed after a plant goes on line, “backfit-

ting” usually is used in the literature to refer to both types ofchanges. Modifications requested by the permit or license holderare termed “amendments” or “variances. ”

such participation prolongs hearings unnecessari-ly without adding to safety. They believe thatthese factors have contributed to higher costs and

longer construction times and may have reducedsafety by requiring the applicant and the regula-tors to focus more of their efforts on the proce-dural aspects of licensing to the detriment ofsubstance.

Nuclear critics, on the other hand, argue thatthe lack of predictability and construction diffi-culties were due to the immaturity of the tech-nology and a “design-as-you-go” approach. Thecritics feel that many of their safety concernswould not have arisen had it not been for therapid escalation in plant size and number oforders that occurred in the 1960’s and 1970’s,utility and constructor inattention to qualityassurance, and inconsistent interpretation and en-forcement of regulations within NRC. While somecritics say that nuclear plants will never be safeenough, others believe that the current regula-tory process could ensure safety if it were inter-preted consistently and enforced adequately.Most critics agree that limiting the opportunitiesfor interested members of the public to partici-pate in licensing will detract from safety.

This section will describe in detail the variousparties’ views (as determined by OTA) on NRC

regulation—what they believe works, what theybelieve doesn’t, and why they hold their views– and how they think the regulatory process couldbe improved. These views were solicited by OTA




at workshops and panels involving a broad spec-trum of interested parties, including nuclear crit-ics and representatives from utilities, vendors, andAEs. These meetings were supplemented by a sur-vey conducted for OTA in which a small sampleof qualified individuals responded to an exten-sive questionnaire(l 5). Suggestions for revisions

to current NRC regulations and procedures havebeen made by a number of interested parties.They will be presented in the following text asoriginally proposed and then evaluated on thebasis of the information available to OTA. Thediscussion of regulatory revisions will focus ontwo legislative packages currently before Con-gress: The Nuclear Powerplant Licensing ReformAct of 1983, submitted by NRC (23), and the Nu-clear Licensing and Regulatory Reform Act of1983, proffered by the U.S. Department of Energy(DOE) (21).

The evaluation of proposals for changes in NRCregulation must depend first on an assessment ofthe goals to be served by regulation and by theindividual changes. The primary goal of NRC reg-ulation as defined in the Atomic Energy Act is toensure that the utilization or production of specialnuclear material will be in accord with the com-mon defense and security and will provide ade-quate protection to the health and safety of thepublic. Therefore, in analyzing proposals forchanges in licensing, the first consideration mustbe whether they are necessary to further the ful-fillment of this goal. If changes would further this

health and safety goal–or at least not detractfrom it–then secondary policy goals might be:

q

q

q

to provide a more predictable and efficientlicensing process in order to assure licenseapplicants that a plant, once approved, canbe built and operated as planned;to increase the effectiveness of public par-

ticipation in licensing; andto improve the quality of NRC decisions inorder to increase public confidence in plantsafety.

Achieving these secondary policy goals prob-ably is a necessary condition in ensuring (whetherfor national security, economic, or other reasons)that nuclear power remains a viable option forutilities in choosing their mix of generating tech-nologies. However, it should be emphasized thatthese goals cannot be accomplished through li-censing changes alone. Rather, they also will re-

quire a commitment to excellence by all partiesin the management of plant licensing, construc-tion, and operation, as well as a commitment toresolving outstanding safety and reliability issues.

Another consideration in evaluating proposalsfor change in the licensing process is whetheramendment of the Atomic Energy Act is necessaryto accomplish a particular change, or whetherit can be accomplished through rulemaking oreven simply more effective implementation of theexisting regulations.

BACKFITTING

The utilities’ and the nuclear industry’s com-plaints about lack of predictability in reactor reg-ulation focus principally on the potential forchanges in regulatory and design requirementsduring plant construction and operation (“back-fitting”).

The present NRC regulations define backfittingas” . . . the addition, elimination or modification

of structures, systems or components of the facili-ty after the construction permit has been issued”(16). Under the present regulations, the stand-ard NRC may use (the language in the regulations

is discretionary) in ordering a backfit is whetherit will “provide substantial additional protectionwhich is required for the public health and safe-ty or the common defense and security.”

NRC never has invoked the backfit definitionformally to amend a permit, license, rule, regu-lation, or order. Rather, it has changed its require-ments through a variety of less formal procedures,

such as bulletins, circulars, regulatory guides, andinformal meetings. While NRC has justified thechanges on a safety basis, the decisionmakingprocess has not been as transparent nor as pre-



Ch. 6—The Reg ulation of Nuc lear Powe r q155

dictable as desired by the industry or its critics.The industry would like to have the backfit rulerewritten and the procedure for invoking it re-vised so that it would provide greater certaintyand stability in terms of costs and schedules andgreater flexibility in implementation. The criticswould like to see NRC follow an established and

documented procedure in ordering backfits to fa-cilitate evaluation of safety considerations.

Specific Concerns

Until recently, NRC’s Office of Nuclear Reac-tor Regulation has been responsible for review-ing and coordinating backfit proposals. The nu-clear industry has perceived that review to be un-systematic, haphazard, and without reference toany regulatory standard. The industry does notbelieve that all of the changes made to plants over

the years have contributed significantly to safe-ty. In fact, it considers some of these changes tohave made plant design and operations morecomplicated, less predictable, and possibly lesssafe (8). Moreover, these changes have absorbeda large share of the utilities’ financial and tech-nical resources. For example, after a decade ofoperation, there were still hundreds of peopleworking to make changes at the Browns Ferrynuclear plants (5). At another utility, the backfitsin 1980 alone required $26 million and 10 to 12staff-years of engineering. In addition, the long-term expenditures associated with Three Mile

island backfits are estimated to cost $74 millionat the same plant (26). Thus, there are powerfulincentives for the nuclear industry to try to havethe backfit rule and its implementation changed.

Nuclear critics counter that the rule would beadequate if it were implemented consistently andunderstandably. They contend that backfits servean important safety function, since many prob-lems arise only after construction or operationhas been initiated. The critics, however, agreewith the industry that it would be more appro-priate to allocate resources to the design phaserather than using them to satisfy safety concernswith backfits. Unfortunately, this is not an optionfor existing plants, but can be applied only to thenext generation of nuclear reactors.

To review these claims about backfitting, OTAundertook a survey of people of all viewpointsconnected with nuclear power, including indus-try representatives, regulators, and critics. Theresults of this survey form the basis for the anal-ysis presented here.

There are certain ways in which backfits havethe potential to adversely affect the safety ofnuclear plants. First, additional equipment canimpair normal operations or safety functions;backfits related to seismic protection often arecited as examples of these problems. As discussedin chapter 4, requirements for additional pipehangers and restraints increasingly have con-strained the layout of the piping systems in nu-clear powerplants. This could contribute to ther-mal stresses in normal operation and make thesystem more prone to failure in accident situa-tions. Another adverse consequence associated

with seismic backfits is that they can lead to over-crowding, making some equipment virtually inac-cessible for inspection or maintenance.

Backfits also can affect safety by disrupting nor-mal plant operations while new equipment is be-ing installed. While this potential problem is lessa result of NRC’s management of backfits thanof utility planning and expertise, it is importantto recognize that there are safety implicationsassociated with installation. Such an incident oc-curred at the Crystal River plant in 1980 whenthe utility attempted to install a subcooling mon-itor while the unit was still operating. This action

triggered a series of unplanned events, eventuallyfollowed by safe shutdown of the plant.

It should be noted that while examples can befound in which backfits have adversely affectedsafety, this does not imply that the overall impacthas been negative. In fact, one recent study in-dicates that modifications made to plants afterThree Mile Island may have reduced the proba-bility of a large-scale accident by as much as afactor of six at some plants (1 3). However, theoverall gain or loss in safety due to backfits hasnot been analyzed in any comprehensive fashion.

OTA concludes that, while most backfits rep-resent safety improvements, they can have anegative impact when deployed in a manner




that does not allow for sufficient analysis of theconsequences of installing or modifying equip-ment and its interaction with other systems. Amore rational and less hurried approach couldimprove this situation for current plants. if thenext generation of reactors is an evolutionary de-velopment of today’s light-water reactors (LWRs),

new plants should be even less troubled by back-fits. New LWR designs will incorporate the les-sons learned from Three Mile Island and the ac-cumulated experience of current reactors, andthey will address unresolved safety issues withstate-of-the-art technology. However, if an alter-native reactor design is selected for commercialdeployment, it maybe impossible to avoid exten-sive backfitting until the technology is fullymat ure.

NRC backfit requirements also have been criti-cized by the nuclear industry as being overly pre-scriptive to the point of being incompatible withpractical design, construction, or operating tech-niques. Rather than establishing general guide-lines or safety criteria and allowing individual util-ities some flexibility in applying them, NRC gen-erally issues detailed and specific requirements.Nuclear powerplant designers must conform tothe regulations and appendices in 10 CFR Part50 as well as 10 other major parts to title 10, over150 regulatory guides, three volumes of branchtechnical positions, numerous inspection and en-forcement circulars and bulletins, proposed rules,and over 5,000 other voluntary codes and stand-ards that may be invoked at any time by regula-tory interpretation. During construction, thesestandards and codes often are interpreted in thestrictest sense possible, with no allowances forengineering judgment. For example, the filletweld, which is commonly used in field construc-tion, varies in width along the length of the weld.Plant designers recognize that some variation willoccur and set the design requirements accordingto an average width. An inspector, by strict in-terpretation of an industry code, may not lookat the average width, but reject an otherwise ac-ceptable weld if it is slightly less in width than

called for by the designer at any point along thelength of the weld. Constructors compensate forsuch anomalies by overwelding, which entailsconsiderable time and expense (19).

It is OTA’s conclusion that the requirementsassociated with the design, construction, andoperation of nuclear powerplants are prescrip-tive and, in some cases, internally inconsistentor in conflict with other good practices. However,while the inconsistencies and contradictions areproblematic, the prescriptive nature of the rules

should not pose insurmountable difficulties forplant owners and designers. Some utilities havebeen able to accommodate to the same prescrip-tive requirements that govern all nuclear con-struction and still complete their plants efficientlyand with few regulatory difficulties. Moreover,NRC is not wholly responsible for prescriptiveregulation. The nuclear industry has developeda large and growing set of voluntary standardsto provide guidance in interpreting NRC criteria.These standards were expanded greatly in themid-1970’s to match the growth in NRC require-ments and often were written with little consid-

eration of their impact. In addition, many of theearly standards were written too rapidly to reflectfield experience and a convergence of acceptedpractices. NRC magnified these problems by in-voking the standards precisely as written ratherthan allowing them to evolve gradually (19).

Another concern about backfitting is that thereare no clear and consistent priorities. Permit andlicense holders argue that they have not alwaysbeen given consistent and stable criteria by whichto construct and operate a plant and, as a result,some less important backfits have been imposedbefore more critical ones. The prime example ofthis cited by utilities and the industry is the ThreeMile Island action plan, in which the NRC gavethe utilities no guidance on the relative prioritiesamong approximately 180 requirements of vary-ing importance. The action plan was developedwith little comprehensive analysis. As discussedabove, if the next generation of plants incorpo-rates a clean-sheet design based on past experi-ence with LWRs, backfits should not be as seriousa problem as they have in the past. While a lackof priorities has been troublesome for plants cur-rently under construction or in operation, it isunlikely that future LWRs will experience thesame degree of difficulty.

A final concern about backfitting is its poten-tial contribution to increases in construction lead-



Ch. 6—The Regulation of Nuclear Power q157

times and plant costs, These issues are dis-cussed in greater detail in chapters 3, 4, and 5and are only summarized here. The most recentplants to obtain CPs from NRC required 30 to 40months after docketing (i.e., not including thepreliminary utility planning phase) to obtain theirpermits, compared to 10 to 20 months between

1960-70. Similarly, construction (the time be-tween issuance of the construction permit andthe operating license) typically takes 100 to 115months, up from the 32 to 43 months in 1960-70(18). Backfitting has been suggested as one of thesources of delay, along with deliberate delays dueto a decrease in the need for power and difficul-ties in financing construction.

In order to examine the impact of regulationon nuclear powerplant construction Ieadtimes,OTA analyzed case studies of the licensing proc-ess, which are detailed in volume 2 of this report

(1). Because it is difficult to separate the effectsof backfitting from other regulatory activities, theywere considered in the context of the entire li-censing process. Based on these case studies, onpublished analyses of the causes of increases innuclear plant construction Ieadtimes, and on ex-tensive discussions with parties from all sides ofthe nuclear debate, OTA has concluded that theregulatory process per se was not the primarysource of delay in nuclear plant construction.Rather, during the 1970’s (when Ieadtimes esca-lated the most), utilities delayed some plants de-liberately because of slow demand growth and

financial problems, Plant size was being scaledup very rapidly and construction was begun withincomplete design information. The increasing-ly complex plant designs meant that more mate-rials—concrete, piping, electrical cable—were re-quired, and constructors often experienced de-lays in delivery of equipment and materials. Atthe same time, worker productivity declined sub-stantially, at least in part because plants weremore complicated and thus more difficult for theutilities to manage and build (3).

Backfits did lead to delays in some plants, es-pecially those subject to the extensive regulatory

changes that followed the accidents at BrownsFerry and Three Mile Island , but in others the e f-

fects of regulatory changes were moderated

through strong management. All plants had to ac-commodate to some backfits that resulted fromthe immaturity of the technology and the overlyrapid scale-up of plant size. In these cases, regu-latory delays must be considered positive. More-over, in some plants that have experienced reg-ulatory delays, such as Cincinnati Gas & Electric

Co.’s Zimmer plant, regulatory actions were anappropriate response to evidence of improperconstruction practices(9). NRC should not be ar-bitrarily limited from imposing backfit require-ments that lead to long delays in such cases, sinceinterest in the public health and safety shouldsupercede concern for minimizing Ieadtime andcost .

In general, OTA concludes that, as in other as-pects of quality control, skillful management bythe utility, its contractors, and NRC is the keyto avoiding delays that otherwise might result

from the licensing process. Thus, licensing ismost likely to proceed without hitches with ex-perienced, committed utility and contractor man-agement personnel; a clear need for power fromthe plant; and a constant and open dialogueamong NRC staff, nuclear critics, and utility andconstruction managers. Since skillful manage-ment has not been a hallmark of NRC administra-tion, changes to make the organization more re-sponsive and efficient should enhance the li-censing process and reduce unnecessary delays.However, such changes cannot substitute forgood utility management and a commitment to

safety in construction and operation.

Proposals for Change and Evaluation

In 1981, NRC created the Committee for Re-view of Generic Requirements (CRGR) to respondto some of industry’s concerns and to reducesome of the burdens that the utilities felt backfit-ting had imposed on them. The CRGR reviewshould guide the industry in assigning priorities,even if it does not solve some of the more fun-damental problems with backfitting. The NRCand DOE proposals for reform attempt to address

the larger issues,In evaluating the proposals outlined below, it

is important to recognize that backfitting cannot




be eliminated entirely, but will continue to be ap-plied to plants under construction or in opera-tion as long as there are outstanding generic safe-ty questions. As discussed in chapter 4, the cur-rent generation of LWRs still is troubled by a num-ber of potentially serious safety issues eventhough they have been studied extensively by

NRC and industry groups. Nuclear critics are con-cerned that the resolution of problems such assteam-generator degradation and cracking in pri-mary system components might be compromisedin the interest of limiting backfits. Therefore, theyare skeptical about proposals that would restrictNRC’s freedom to impose legitimate backfit re-quirements or emphasize cost and efficiency atthe expense of safety.

The debate about backfitting centers on fourmain considerations of the backfitting rule: 1 )whether the current definition and standard needto be revised or simply invoked and enforced;2) if they do need to be changed, how shouldthe new definition and standard be phrased, 3)what criteria should be applied by NRC in order-ing a backfit; and 4) whether any changes thatmay be needed should be made legislatively orthrough rulemaking. As discussed in the previoussection, some change in the manner in whichbackfits are managed and enforced within NRCprobably is necessary so that the primary regu-latory goal of ensuring safety is achieved. More-over, to provide license applicants with more sta-bility and certainty, and to increase the effective-ness of public participation in licensing, the back-fit procedures and criteria at least must be mademore explicit.

Definition

The present definition of a backfit in the NRCrules includes any design or technological changeordered after issuance of the CP. In doing so, itignores the reality that much design informationis not available when the CP is issued, and notall evaluations and modifications of designsshould be considered backfits merely becausethey are postpermitting. From another perspec-

tive, however, the present definition may be toonarrow in that it focuses only on changes in“structures, systems, or components of the facili-

ty,” and thus excludes important institutional andmanagement changes.

One alternative definition has been put forwardby the NRC Regulatory Reform Task Force (RRTF)in its proposed revisions to the NRC rules: “theimposition of new regulatory requirements, or themodification of previous regulatory requirements

applicable to the facility, after the constructionpermit has been issued” (23). Prior to the invoca-tion of a backfit, NRC would set approved de-sign and acceptance criteria for the protectionof public health and safety and national securi-ty. Once a licensee embarked on the design, con-struction, or operation of the reactor and hadcommitted substantial resources to and was act-ing in accordance with the NRC criteria, then,according to the definition above, any proposedchange in those criteria should be considered abackfit and should trigger a special decisionmak-ing process.

A second definition (proposed in the DOE legis-lative package) is “an addition, deletion, or mod-ification to those aspects of the engineering, con-struction or operation of a . . . facility upon whicha permit, license or approval was issued” (21).This definition may be slightly narrower than theRRTF definition in that it applies backfit criteriaonly to the conditions in a license rather than tothe full range of regulatory requirements appli-cable to a facility.

The most important attributes of any NRC re-quirement are explicit criteria and consistent ap-plication of these criteria by NRC management.Thus, either NRC or DOE proposed definitionswould be preferable to the current one, underwhich it is unclear when a change ordered byNRC should be considered a backfit, providedthat the application of the definition by NRC isconsistent and clear to all interested parties. Sucha change should contribute to more predictabilityabout backfits.

The definition of a backfit would be particularlyimportant if it were coupled with a thresholdstandard for triggering it. One approach wouldrequire a backfit to result in a substantial increasein public protection, with benefits from the in-creased protection exceeding both the direct and




indirect costs of the backfit. While such a cost-benefit standard would presumably assure con-sistency, OTA concludes that the available meth-odologies are inadequate to fully quantify im-provements in safety. Thus it is likely that a cost-benefit standard alone (or the use of quantitativesafety goals to justify backfits, as discussed in

detail below) would be unworkable until suchmethodologies are developed further. Rather,some combination of engineering judgment cou-pled with cost-benefit analysis, as has been usedin the past, will be necessary.

Within this context, however, NRC could im-prove the process of evaluating and imposingbackfits by making its standards more explicit andby specifying the relative consideration to begiven to factors such as the effects for orderingbackfits on public and occupational exposuresto radioactivity; the impact on safety given overall

plant system interactions, changes in complexi-ty, and relationship to other regulatory require-ments; the cost of implementing the backfit, in-cluding plant downtime; the resource burden im-posed on NRC; and, for backfits applicable tomultiple plants, the differences in plant vintageand design. While these factors probably are con-sidered in some form in NRC’s current delibera-tions, the decisionmaking process is frequentlyinscrutable.

Other changes could be made in the backfitreview process to ensure that criteria and stand-ards are applied consistently. A centralized groupsuch as CRGR or ACRS could review backfits rou-tinely and judge them according to standards es-tablished by NRC. Alternatively, an independentpanel of experienced engineers drawn from util-ities, the public, and industry (but not from theorganization that did the design) could be set upfor centralized review.

General Procedures

Changes in overall procedures and guidelinesfor backfitting also have been proposed. The DOEbill would shift the burden of proof from industry

to NRC by requiring NRC to demonstrate that abackfit is cost effective. Moreover, the DOE billwould restrict the information that NRC can re-quire from licensees. In addition, it implicity di-

rects NRC to employ a lower standard of safetyfor older plants with shorter remaining operatinglives, even though these are often the plants mostin need of upgrading. Further, the DOE bill wouldapply to breeder reactors and reprocessing plantswhere backfitting is likely to lead to significantimprovements in safety.

These procedural changes in the DOE billwould have the effect of making it more difficultfor NRC to order safety-related improvementsafter a construction permit has been issued. Suchchanges will be controversial without other assur-ances—absent in the DOE bill—that safety canbe assured.

A more general and fundamental change hasbeen proposed by the nuclear industry, whichwould like to see NRC’s prescriptive rules re-placed with a few general criteria. In such a sys-tem, each utility could determine how it mightbest satisfy NRC’s criteria, subject to concurrenceby NRC. OTA finds that the latter proposal hassome merit in that it might encourage innovativeapproaches among the more capable utilities andvendors. Treating the problems generically ratherthan prescriptively also might reinforce the useof owners’ groups and data pooling. However,it should be noted that such an approach alsocould pose severe resource problems for NRC.If NRC staff had to review numerous different pro-posals for changes rather than devise a singlesolution of its own, it would severely tax a systemthat already has difficulty with coordination and

organization.

It generally is agreed that the key to a shift toperformance standards is to make the industry(including utilities, vendors, and AEs) accept fullresponsibility for safety and to design and buildplants according to a consistent regulatory philos-ophy rather than making numerous modificationsas problems rise. Acceptance of this responsibilitycould be demonstrated in part by industrywideimprovements in management practices, quali-ty control, performance records, and event-freeoperations. If the evidence indicates that the in-

dustry has matured sufficiently to be able to con-struct and operate plants safely and reliably, NRCmay be able to allow more flexibility in the inter-pretation of its guidelines. However, as long as




any industry participants demonstrate an inabili-ty to guarantee safe operations, OTA believesthat the current level of detail in the backfit reg-ulation probably is necessary to fulfill NRC’sprimary legislative mandate of protecting publichealth and safety.

Legislation

OTA found that congressional action is notnecessary to change the backfit rule. Changesthat would contribute to reactor safety, and lendstability and certainty to, and increase the effec-tiveness of public participation in this aspect ofregulation can be accomplished better adminis-

tratively, through rulemaking. This would allowgreater flexibility in adjusting to changing con-struction and operating experience and in apply-ing risk assessment and cost-benefit analysis thana backfitting standard rigidly determined by leg-islative action. Because of the extensive publiccomment process associated with rulemaking, it

also might permit greater participation in devel-opment of a backfit rule by all parties. This wasthe rationale followed by NRC in drafting its leg-islative regulatory reform proposal, which did notinclude provisions related to backfitting. NRC per-sonnel reportedly are working on a draft revisionof the current rule, which will appear as a noticeof proposed rulemaking.

HEARINGS AND OTHER NRC PROCEDURES

Hearings and other procedural aspects of NRClicensing and safety regulation, including the con-duct of safety reviews, management problemswithin NRC, the use of rulemaking, and some as-pects of enforcement, are highly controversial.The industry and the utilities perceive the hear-ings and other procedures as contributing mini-mally, if at all, to plant safety, but requiring over-whelming amounts of paperwork and manage-ment resources. Nuclear critics, on the otherhand, see these procedures as their only meansof raising safety concerns, and they strongly ob-

ject to any attempts to limit the process and their

participation.

HearingsThe current licensing process includes adjudi-

catory hearings, * with public participation, beforea CP is issued and optional hearings (generallyrequested) at the OL stage. Formal adjudicatoryhearings probably are not required under theAtomic Energy Act, which does not specify thetype of hearing that the Commission must hold.However, they have been granted for so long that

*A formal adjudicatory hearing is similar to a trial, in that the par-ties present evidence subject to cross-examination and rebuttal,and the tribunal or hearing officer/board makes a determinationon the record. The key ingredient is the opportunity of each partyto know and meet the evidence and the arguments on the otherside; this is what is meant by “on the record. ”

it would be difficult, if not impossible, to inter-pret the act as allowing anything less than a for-mal adjudicatory hearing. Furthermore, trial-typehearings are required under the principles of ad-ministrative law to the extent that the examina-tion of evidence is necessary to resolve questionsof fact, as opposed to issues of law or policy,which can be resolved in legislative-type hearings.

Part of the debate concerning hearings has fo-cused on the appropriateness of using an adju-dication process to resolve technical issues. In-dustry and utility representatives claim that the

current system leads to unnecessary delays andinefficient allocation of resources. On the otherhand, nuclear critics view the hearing process asan opportunity to examine NRC records and raiseissues that might have been overlooked. In thissense, the adjudicatory hearings are appropriatefor NRC licensing because they are designed, le-gally, to illuminate the contested issues of fact andcause the utility and NRC to justify their technicaldecisions more thoroughly than they might in alegislative-type hearing.

Closely associated with the issue of appropriate-ness is the efficiency argument. The industryclaims that hearings have been too long (spreadout over a year or more in extreme cases) andcostly due to the highly technical and complexnature of the subject matter and the inclusion of




issues not directly germane to safety, such asneed for power and alternative means of gener-ating that power, It is possible that changes couldbe made to the hearing process to reduce ineffi-ciencies while preserving the right of the criticsto participate effectively. As discussed below, leg-islative action would not necessarily be required,

since most of the problems could be amelioratedby strict conformance to the NRC rules of admin-istrative procedure.

A final important consideration is the degreeto which critics participate in the decisionmak-ing process and their effectiveness in raising safetyconcerns. Timing is a central issue concerningparticipation. In the past, plant designs have beenso incomplete at the CP hearing stage that it hasbeen virtually impossible to make constructivecriticisms about them. But by the time the OLhearings are held, the final design is complete,

it has been reviewed and approved by the NRCstaff, and the plant is built. Therefore, any con-cerns the critics raise are directed toward a groupthat has already decided upon the plant’s safety.

Another issue related to participation is the ef-fectiveness of the interaction between critics andthe NRC staff. Industry representatives interactwith the staff prior to hearings and reach agree-ments on the major safety issues. When the criticsquestion these resolutions at the hearings, theyfeel that the staff does not give adequate atten-tion to their complaints. Furthermore, the criticsfeel that they have even less influence with the

staff when they are not in an adjudicatory set-ting. The critics cite occasions on which theywere ignored by the NRC staff when they infor-mally raised issues such as emergency core cool-ing, environmental qualification, and fire protec-tion. These issues later proved to be majorconcerns.

Proposals for Change in the HearingProcess and Evaluation

There have been several proposals to addressthe industry’s and critics’ complaints about theNRC hearing process, including changing the for-mat of the hearings, improving management ofthe hearings and other procedures, and chang-ing the structure of licensing so that safety issues

are addressed in a public forum before the CPor OL hearing.

The industry and some regulators would liketo see the hearings restructured to a hybrid for-mat that would combine some of the elementsof adjudication and legislative-type hearings. Ina hybrid hearing, all testimony and evidencewouId be presented first in written form, as in alegislative hearing. Adjudicatory hearings wouldbe granted on issues that present genuine andsubstantial factual disputes that only could beresolved with sufficient accuracy by the introduc-tion of evidence in a trial-like setting.

Both the NRC and DOE legislative packageswould amend the Atomic Energy Act to providefor hybrid hearings. Under the NRC proposal,hearings on CPS would be optional rather thanmandatory, and the Commission could substitutehybrid hearings for adjudication, after providing

the parties an opportunity to present their views,including oral argument on matters determinedby the Commission to be in controversy. Sucharguments would be preceded by discovery, andeach party, including the NRC staff, would sub-mit a written summary of the facts, data, andarguments to be relied on in the proceeding. Thehearing board then would designate disputedquestions of fact for resolution in an adjudicatoryhearing based on the standard described aboveand on whether the decision of the Commissionis likely to depend in whole or in part on the res-olution of a dispute.

The hearings as proposed by NRC and DOEwould be limited to matters that were not andcould not have been considered and decided inprior proceedings involving that plant, site, ordesign unless there was a substantial evidentiaryshowing that the issue should be reconsideredbased on significant new information, The NRCbill defines “substantial evidentiary showing” asone sufficent to justify a conclusion that the plantno longer would comply with the Atomic EnergyAct, other Federal law, or NRC regulations (23).

The DOE bill would require hybrid hearings tobe substituted for adjudication. This maybe con-tracted with the NRC bill, in which the shift toa hybrid hearing would be discretionary. TheDOE bill would allow anyone to introduce writ-




ten submissions into the record. Interested par-ties could petition the hearing board for oral ar-guments, which would be granted on contentionsthat had been backed up with reasonable speci-ficity. As in the NRC bill, oral argument wouldbe preceded by discovery and submission of writ-ten facts and arguments. After oral argument,

each party could file proposed findings that setforth the issues believed to require formal hear-ings. Under the DOE bill, the hearing board’s de-cision as to which issues required adjudicationwould be reviewed by the Commission.

The DOE bill specifies that issues raised andresolved by NRC in other licensing proceedingscould not be heard again unless “significant, newinformation has been introduced and admittedwhich raises a prima facie showing that actionis needed to substantially enhance the publichealth and safety or the common defense and

security.” New issues would not be admittedunless they were “significant, relevant, materialand concerned the overall effect of the plant”on health, safety, or security (21).

Efficiency improvements

Hybrid hearings are intended to increase theefficiency of the hearing process and to improvethe effectiveness of public participation in thatprocess. In terms of efficiency, proposals for hy-brid hearings are directed toward complaints thathearings are too long and costly and tie up toomuch of staff and industry resources without con-

tributing to plant safety. Although it is true thatthe hearings can be unduly long and expensive,OTA found, based on extensive discussions withutility and industry representatives, regulators,and nuclear critics and public interest groups, thatif management of the hearings were tighter,either through enforcement of the existing reg-ulations or through changes in those regula-tions, a formal shift to hybrid hearings wouldbe unnecessary. Most of the problems cited bythe industry that contribute to unnecessarily longhearings can be remedied through better man-

agement control by the utilities and NRC to en-sure that safety issues are resolved early in thelicensing process and through tighter manage-ment of the hearings by the licensing boards orhearing officers without making fundamental and

highly controversial changes in the structure andscope of the hearings themselves. Furthermore,because proposals for a shift to hybrid hearingsinclude more opportunities for requesting hear-ings than under the present licensing process, andmore administrative decisions subject to appeal,it is likely that these proposals actually would in-

crease the amount of time taken up by hearings.Other changes in NRC regulations or in man-

agement of the hearings could contribute to moreefficient hearings. Such changes include: vigor-ously enforcing existing NRC regulations that im-pose time limits in hearings; excluding issues notraised in a timely manner without a showing ofgood cause; requiring all parties to specify thefactual basis for contentions; resolving genericissues through rulemaking once they have beenlitigated in a licensing proceeding; using summarydisposition procedures for issues not controverted

by other parties; excluding issues that were raisedand resolved in earlier proceedings unless ashowing of significant new information can bemade; and eliminating consideration of issues notgermane to safety that are best considered inother forums. Only the last of these changeswould require legislative action.

improvements in Public Participation

The hybid hearings proposed by NRC and DOEalso can be assessed in terms of the effectivenessof public participation. DOE and the industryargue that these proposals would provide moreopportunities for critics to influence the decisionprocess. As stated by Secretary of Energy DonaldHodel:

After a plant is essentially complete, with manyhundreds of millions–or billions–of dollarsalready spent, the view of the public cannot, as

a practical matter, be considered as effectively

as it co uld be at the b eginning of the licensingproc ess. Therefore, [DOE is] p ropo sing a system

with multiple op portunities for public p articipa -

tion early in the process, before firm decisionsare made by the Commission and the applicant

(6).

Under the DOE bill, these opportunities wouldoccur if and when standardized plants are con-sidered for approval, when the specific site is con-




sidered for approval, and when the issuance ofa combined CP/OL is being considered. The NRCbill would allow hearings at these points as wellas on construction permits, operating licenses,and preoperational reviews for plants with CP/ OLs.

In analyzing whether hybrid hearings wouldimprove the effectiveness of public participationin licensing, it is important to distinguish the tim-ing and number of hearings from the scope ofthose hearings. To the extent that the NRC andDOE bills would increase the number of oppor-tunities for public involvement in nuclear plantlicensing before final decisions have been made,they would improve the effectiveness of publicparticipation in the hearing process. However,

if those opportunities are not provided when de-sign decisions are being made and safety issuesare being raised and resolved—all of which cur-

rently occurs in industry-staff interactions fromwhich members of the public are excluded—thenthe public’s ability to have its safety concernsheard will not be improved, and the critics stillwill feel that decisions will have been made priorto the hearings.

Nuclear critics contend that the means pro-posed by DOE and NRC to increase the efficien-cy of the hearings would serve to undercut theeffectiveness of public participation by severelylimiting the scope of that participation. They notethat both bills would weaken the rights of thepublic to cross-examine NRC and utility wit-nesses, which they argue is often the only wayto uncover safety problems and uncertainties thatcould not be revealed through examination ofwritten testimony. Furthermore, the critics feelthat both bills (but especially the DOE bill) maymake it more difficult for members of the publicto raise serious safety issues by raising the stand-ards for admission of evidence.

Nuclear critics also point out that, under thebills’ provisions for hybrid hearings, the hearingboard would have to decide in each case whichevidence is subject to cross-examination—a deci-

sion that often would be appealed, thus lengthen-ing the process rather than shortening it. Underthe DOE bill, the Commission itself would haveto review the hearing board’s decision, plus the

written submissions and oral presentations, andaffirm or reverse the board’s designation on eachissue. The critics are especially cautious aboutNRC dictating to the hearing boards which issuesto consider. They cite quality assurance at Zim-mer and the steam generators at Three Mile Islandas issues NRC previously has taken away from

hearing boards on the grounds that the staff wasworking on them. In the critics’ view, all of thepoints listed above are serious defects that wouldseriously erode public confidence in the effec-tiveness of NRC safety regulation.

OTA concludes that the effectiveness of publicparticipation in licensing can be improved with-out causing the hearing process to negatively af-fect costs or construction schedules. First, theproposals for early design and site approvalswould permit extensive public participation inhearings on safety issues prior to the start of con-

struction of any particular plant. Then, when autility applied for a CP based on an approved de-sign and site, the only questions that would re-main to be heard in the CP hearings would bethe combination of the site and the design, plusany safety issues that were not resolved in thedesign approval. This might alleviate the critics’concern that design-related safety issues are re-solved in private industry-staff interactions. Allow-ing public involvement early and often in utilityplanning for nuclear power also would enhancethe effectiveness of public participation in thelicensing process.

Second, a funding mechanism for public par-ticipation in licensing would ensure that the criticscould make a substantive contribution to designand safety issues by enabling them to devotemore resources to the identification and analysisof reactor engineering and safety. This would re-spond to the industry’s complaint that the criticsare not sufficiently knowledgeable about reac-tor engineering and safety, as well as to the critics’view that the utility, and to a lesser extent theNRC staff, can devote extensive resources to de-fending design decisions.

Funding of public participation has been a partof the rulemaking proceedings in the FederalTrade Commission, the National Highway Traf-




fic Safety Administration, DOE, and the Environ-mental Protection Agency, and of both rulemak-ing and public hearings in the National Oceanicand Atmospheric Administration and the Con-sumer Product Safety Commission. In the mid-1970’s, NRC considered an intervener fundingprogram but did not implement one, arguing that

NRC adequately represents the public interest inreactor safety and that the present method offunding through citizen contributions is a moredemocratic measure of public confidence in howwell NRC does its job. Given the extent of thecriticism of NRC management and expertise ex-pressed to OTA by all parties, any policy pack-age intended to revitalize the nuclear optionshould include reconsideration of an intervenerfunding program and alternatives such as an of-fice of public counsel within NRC.

Changes in the Role of the NRC Staff

The NRC staff currently participates as an ad-vocate of the license application in the hearings.This role is a consequence of the detailed involve-ment of the staff in licensing issues and the resolu-tion of most issues to the satisfaction of the staffand the applicant prior to the hearings. The disad-vantage of this situation is that the staff may beperceived as being less effective in resolving safe-ty problems than it might be. This concern couldbe addressed by limiting the staff’s participationin contested initial licensing proceedings to thoseissues on which it disagrees with the applicant’s

technical basis, rationale, or conclusions. The staffthen would not be perceived as defending a par-ticular plant in a hearing and might be more ef-fective in aiding ASLB.

A related issue is the ex parte rule. Like a courttrial, an agency adjudication is supposed to bedecided solely on the basis of the record so thata participant in an adjudicatory hearing will knowwhat evidence may be used and will be able tocontest it. These rights can be nullified if agencydecisionmakers are free to consider facts outsidethe record without notice or opportunity to re-spond.

The most common problem of extrarecord evi-dence occurs when there are ex parte contacts– communications between any interested party

and an agency decisionmaker that take place out-side the hearing and off the record. The Admin-istrative Procedures Act (APA) prohibits suchcommunications once a notice of hearing hasbeen published for a particular proceeding.When an improper off-the-record contact doesoccur, the A PA requires that it be placed on the

public record; if it was an oral communication,a memorandum summarizing the contact mustbe prepared and incorporated into the record.

Strict interpretation of the ex parte rule effec-tively cuts off communication between the Com-missioners and some parts of NRC during a licens-ing determination or requires that the commu-nication be made public. The Rogovin Reportrecommended more active involvement by theCommissioners in individual licensing determina-tions, but implementation of this recommenda-tion is constrained by the ex parte rule (14). Inits rulemaking options, NRC’s RRTF argued thatthe Commissioners should be allowed to talk tostaff supervisory personnel who are not partici-pating directly in a particular hearing. The exparte rule could be interpreted more liberally toallow such Commission/staff interaction—espe-cially if the role of the staff in hearings is limited—as long as true ex parte communications continueto be made public.

Other NRC Procedural Issues

Additional issues related to NRC procedures in-clude the role of ACRS in the conduct of safety

reviews; management problems within NRC andother aspects of safety reviews; the use of rule-making; and NRC enforcement methods.

Advisory Committee onReactor Safeguards

The Atomic Energy Act requires ACRS to revieweach license application referred to it, at both theCP and OL stages, even if the Committee doesnot judge the review to be merited. Many observ-ers consider ACRS to be particularly adept at re-vealing previously unrecognized safety problems,

but because its members devote only part of theirtime to ACRS activities, it has few resources topursue such problems in depth. Both the RogovinReport and the President’s Commission report on




Three Mile Island recommended that ACRS berelieved of its mandatory review responsibilitiesand be allowed to participate in hearings (7,14).A 1977 NRC review of the licensing process alsorecommended that the ACRS be given discretion-ary authority to decide which license applicationsmerit its review (22).

The DOE bill would make ACRS discretion verybroad; it would amend the Atomic Energy Actto make ACRS review of applications to grant,amend, or renew a CP, OL, combined construc-tion and operating license (COL), or site permitdiscretionary unless the Commission specifical-ly requested a review. Only ACRS review of de-sign approvals or amendments and renewalswould be mandatory under the DOE bill. Further,the DOE proposal specifies that neither the ACRSdecision to review nor the NRC decision to referan issue to ACRS would be subject to judicial re-

view. Under the NRC legislative proposal, ACRSreview of CPS, OLs, site permits, design approv-als, and amendments to any of these would con-tinue to be mandatory.

OTA concludes that the ACRS review of de-signs should be mandatory to ensure that safe-ty problems are identified early in the licensingprocess. If proapproval of standardized designswere implemented, only discretionary ACRS re-view of a CP application should be required be-cause it would be based on a thoroughly studieddesign, Similarly, if site-banking were imple-mented, ACRS reviews of sites also could be dis-

cretionary. Another mandatory ACRS reviewmight be appropriate before granting an OL ordeciding to allow a plant to begin operationunder a COL, since at this stage significant safe-ty issues can arise about compliance with theoriginal design.

Management Control

According to the industry, the primary prob-lem with NRC procedures is lack of managementcontrol within NRC, as reflected in uneven safe-ty and other reviews, in a lack of priorities, andin the problems with backfitting discussed pre-viously. There does not appear to be any true de-cisionmaking process; rather, NRC appears to re-act to immediate, pressing problems. As a result,

small problems can be given proportionatelymore attention than is warranted. Furthermore,the decision path within NRC is virtually untrace-able, making it difficult to knowledgeably critiquethe staff’s analysis and resolution of safetyconcerns.

Another concern that is shared by the NRC staffand the industry is that the regulatory process istoo cumbersome and legalistic in an area thatis primarily technical. This produces requirementsfor an inordinate amount of paperwork and maydivert attention away from the primary missionof ensuring plant safety. For example, the ShollyRule (which requires that a notice be put in theFederal Register before any change—no matterhow trivial–is made in a plant’s technical specifi-cations) requires extensive staff attention and re-sources, but produces little accompanying benefitto the public. Similarly, the industry thinks it has

to report too much to NRC, and that significantsafety issues may get lost in the resulting paper-work.

Another problem concerns consistency of re-views; the SRP helps to even out reviews, but itis limited by the resources of the NRC staff. Thisreview is, of necessity, an audit review, with theratio of hours spent on the design to those spentin review on the order of 10,000 to 1. Consistencywill be increasingly difficult to guarantee as morereview functions are shifted to the NRC regionaloffices.

It is unclear how to address these concerns.Good management cannot be legislated. Addingmore technically qualified staff probably wouldimprove the quality of substantive reviews butwould not necessarily improve management. Fur-ther, it is generally agreed that the ultimateresponsibility for safety rests with the utilities andthe industry. Even the most competent and effec-tive NRC could not make an incompetent or un-willing utility operate safely short of shuttingdown a plant if the utility did not accept thisresponsibility. Financial sanctions other thanfines, such as might be imposed through insurers

or financiers, may be the most effective in thisregard.

As noted above in the discussion of backfitting,it is important that NRC procedures be explicit,

25-450 0 - 84 - 12 : QL 3




workable, and applied consistently, but even thebest written regulations or legislation cannotachieve this if there is not a firm commitment bytop NRC management to ensure that the regula-tions are implemented properly. For example, thecurrent NRC rules of administrative procedure,for the most part, are adequate to increase the

effectiveness of the hearing process but are notenforced by NRC.

The Use of Rulemaking

In other agencies, increasing the use of rule-making, as opposed to bulletins, circulars, no-tices, and regulatory guides, has improved thequality of management decisions due to the ex-tensive opportunities for external review andcomment by all interested parties. However, NRCis not perceived as being particularly good at

rulemaking. Many NRC rules are considered in-comprehensible due to the poor wording thatresults from the cumbersome internal reviewprocess: a rule drafted by the technical staff isrevised by numerous others culminating with thelegal staff-by which time it may be unrecogni-zable—but the technical staff is reluctant tochange the wording lest it has to start the reviewprocess all over again. Thus, the staff tends toavoid rulemaking because fulfilling the review re-quirements is likely to make the final product lookmuch different than the initial intention.

If NRC could streamline its rulemaking proce-

dure, it might be particularly useful in resolvinggeneric issues–those common to more than oneplant. As discussed earlier, one of the factors thatcan contribute to inefficiency in the regulatoryprocess is the consideration of generic questionsduring the licensing or oversight of a particularplant. The Rogovin Report and the President’sCommission on the Accident at Three Mile Islandrecommended the increased use of administra-tive rulemaking procedures to resolve issues thataffect several licensees or plants, as opposed toconsidering such issues during the licensing oroversight of individual plants (7,14). NRC hasbeen heading in this direction over the past dec-ade with its rulings on emergency core coolingsystems and its environmental statement on

mixed oxides. However, considerably more prog-ress could be made in resolving generic issuesthrough rulemaking.

OTA concludes that resolution of genericquestions through administrative rulemakingwould remove a source of regulatory inefficien-

cy if the rulemaking procedure were improved.Moreover, it also would improve the effectivenessof participation by the public (including the in-dustry, nuclear critics, and other interested par-ties) on these issues because of the opportunitiesfor review and comment through publication inthe Federal Register and, often, for public hear-ings on a proposed rule. Furthermore, a rule isan enforceable regulation, and thus is a strictermeans of instituting requirements than notices,circulars, regulatory guides, or bulletins. Also, asdiscussed previously, generic treatment of safe-ty concerns would facilitate industry use of own-

ers’ groups and other management tools.NRC is not particularly adept at rulemaking,

producing poorly worded regulations that are dif-ficult to interpret by those who must implementand enforce them. But, because of the regulatoryand enforcement problems posed by the use ofalternatives to rulemaking for generic issues, NRCwould be better off to improve its ability to writecomprehensible rules than to continue to devel-op solutions to generic problems through licens-ing or notices on individual plants.

RRTF suggested that a generic question be

heard once in a license proceeding and then bepublished as a proposed regulation within 45days after resolution in that proceeding. If the reg-ulation were adopted by NRC following the req-uisite public comment period, it could not be re-Iitigated in subsequent licensing proceedings un-less “special circumstances” were shown. If the“hearing” of the initial rulemaking were in an ad-

judicatory setting, then this proposal would pro-vide for comprehensive discussion of generic is-sues for all parties. However, if the hearing werelimited to a legislative-type proceeding, critics ofthe regulation may not feel that they really have

been heard.

If the RRTF suggestion is not implemented,some other means of involving as many parties



Ch. 6–The Regulation of Nuclear Power q 167

as possible in the development of rules—beforethey are published in the Federal Register forcement—should be devised. NRC is involving theindustry in the development of the generic rulefor radiation protection but not the critics. As aresult, when the draft rule is published, the criticsmay view it with skepticism and distrust of theprocess. Their only recourse would be the publiccomment process and, ultimately, a petition tochange the rule after it has been finalized.

In addition to the concern for ruIemaking pro-cedures, there is another issue relating to con-tent of NRC rules. The current NRC technical reg-ulations have evolved over a 25-year period witheach new rule devised on a largely ad hoc basis.The relative contribution of each of the numerousregulations to safety is undetermined, althoughit is likely to be highly variable. As one utility rep-resentative expressed it:

Many codes and standards were contrived and

written by well-qualified, well-meaning individ-

uals projec ting idea l situations. They never had

any idea that in this day and age of rigid quality

assurance and quality control, the codes and

standards would be enforced to the letter (1 2).

Some industry representatives and regulators

argue that it is now time for a wholesale revamp-ing of NRC’s technical regulations to reflect thecurrent state-of-the-art and the accumulated op-erating experience. Before such a radical step,however, what is needed is a detailed analysisof the existing technical rules. A possible startingpoint would be to initiate a thorough revision ofthe technical regulations related to licensing. Anysuch effort also should examine the relative ad-vantages and disadvantages of a shift away fromhardware-based (prescriptive) standards and to-ward performance criteria, the role of safety goalsand PRA, and the source-term work.

Enforcement

With regard to enforcement, nearly all parties

to the nuclear debate agree that the procedurescould be improved. First, there are some 80 ormore means by which NRC can transmit informa-tion to a utility or the industry, but only two,orders and rules, are mandatory in the sense that

the recipient would be subject to fines or otherenforcement action if he did not respond. Al-though many observers would prefer to see agreater use of rules to change requirements,NRC’s current problems with wording could leadto enforcement problems. inspectors in the fieldhave to enforce a rule based on what was writ-

ten, which may differ from what was meant. In-dividual judgment on the inspectors’ part as towhether the intent of a rule is being met is dis-couraged to prevent the matter from ending upin court. Yet, inspection and enforcement staffrarely are asked to participate in the formulationof regulations, and thus have little contributionto their enforceability.

Second, there is general agreement that thecurrent system of fining utilities for violations doesnot work, although the range of opinions aboutwhy it doesn’t work is quite broad. The utilities

contend that they are less inclined to identify safe-ty concerns when they know that a fine is likelyto follow. Further, they state that the present sys-tem of fines does not distinguish between a one-time simple human failure and continual inatten-tion to problems or negligence. Utilities wouldprefer to see a system in which they could beginby informally negotiating solutions to safety con-cerns with NRC. If the problem is not remediedimmediately, the Commission then could resortto fines and press releases. This procedure cur-rently is followed by some NRC Regional Admin-istrators.

Nuclear critics agree that the present systemof fines is inadequate, but they cite different rea-sons. They point out that a large fine ($500,000)is equivalent to a single day’s outage cost for amajor utility and, in some cases, can be passedon to the ratepayers. They would like to see NRCchange its enforcement policy to include the op-tion of shutting down a plant or denying an OLand making it clear that those options will be in-voked. The current perception that NRC does notenforce the regulations already in place does notbode well for convincing the critics or the indus-

try that strong enforcement is a real threat. Therecent ASLB action in denying an OL for theByron plants may contribute to a change in theperception of NRC’s willingness to enforce itsreguIations.




THE TWO-STEP LICENSING PROCESS

The current two-step licensing process (CP andOL) was instituted before the nuclear industry wasfully mature. There were many first-time licenseapplicants, designers, and constructors with un-proven and incomplete design concepts; at that

time, plant designs needed a final evaluation priorto operation. Now, reactor engineering may havematured to the point where final designs for mostplants can be described at the CP stage. There-fore, the industry argues that a two-step Iicens-ing process no longer is necessary.

The utilities and the nuclear industry contendthat the two-step procedure exacerbates con-struction scheduling problems because the plantdesign, regulatory design review, and hearingsall occur during construction. They would liketo change this to a one-step process that wouldplace all three activities before constructionbegins. They believe this would improve the pre-dictability and efficiency of the licensing processby making scheduling more certain. Also, an OLis perceived in many cases to be pro forma, butit still requires a full EIS and optional but usuallyrequested hearings. * They suggest that a one-stepprocedure might encourage earlier identificationand resolution of licensing issues while continu-ing to accommodate participation by interestgroups and State and local governments.

There are two ways to achieve the equivalentof a one-step NRC licensing process: by combin-

ing the CP and OL, and by banking reactor de-signs and sites. The NRC and DOE legislativepackages include proposals for both of thesemeasures. it should be noted that neither theDOE nor the NRC bills is tied to the use of stand-ardized designs, either in the provisions for com-bined CP/OLs or for design banking. However,in the following discussion of these proposals, itis assumed that plants will be much less custom-ized, relying on only a few standardized and com-plete designs. An earlier OTA study, Nuclear

Powerplant Standardization, found that standard-ization of designs and construction, operation,

*The ASLB refusal to authorize an OL for Commonwealth EdisonCo.’s Bryon plants may indicate a change in approach at NRC. Even

if the decision is overturned by the ALAB, it is unlikely that utilitieswill ever again consider the OL to be a formality.

and licensing practices could alleviate many ofthe nuclear industry’s difficulties in verifying thesafety of individual plants. In addition, standard-ization could facilitate the transfer of safetylessons from one reactor to another and could

help reduce the rate of cost and Ieadtime escala-tions (1 O). As discussed in detail in chapter 4, itis likely that any new plants would try to maxi-mize these advantages by standardizing designsto the greatest extent possible.

The NRC legislative proposal specifies that toget a COL, an application must contain “sufficientinformation to support the issuance of both theconstruction permit and the operating license. ”The NRC staff analysis of the proposal interpretsthis to mean that the application must includean essentially complete design. Under the NRCbill, an optional hybrid hearing could be re-quested before the COL is issued and again be-fore the plant goes into operation for matters thatwere not considered in the first hearing. The finalreview before a plant goes on line would endwith NRC issuing an “operation authorization”that would be the regulatory equivalent of a li-cense for purposes of inspection and judicialreview (23).

This proposal would eliminate the duplicationof detailed environmental and safety reviews thatare currently needed for an OL; otherwise, it isthe equivalent of the present two-step process

with a new name. It is likely that hearings wouldstill be held before construction and again beforeoperation. Moreover, if the plant were a uniquerather than standardized design, this procedurecould take even more time than the current two-step process.

In the DOE legislative proposal, NRC wouldprovide an expedited procedure for COL holdersto start operation by allowing the licensee to cer-tify safety when the plant is virtually complete.NRC would publish notice of the certificationwith a 30-day comment period, and the staffwould have 45 days from the date of that noticeto review the plant for safety, consider the pub-lic comments, and recommend action to NRC.There would then be an additional 30-day peri-




od in which NRC could take action to prohibitor limit operation if the certification was foundto be incorrect. If NRC did not prohibit opera-tion during that period, the plant could goon line.The only opportunity for public hearings wouldbe at the issuance of the initial COL.

The COL proposal is controversial because ofuncertainty about the level of design detail thatwould be required to obtain a combined license,since this is left up to NRC to specify throughrulemaking. In addition, neither bill directs NRCto resolve all outstanding safety issues prior tolicensing. Nuclear critics argue that the numberof design changes still being made between a CPand an OL and the critical safety issues still be-ing uncovered at the OL stage indicate that theindustry and NRC are not yet ready for one-steplicensing. Such a procedure could reduce atten-tion to unresolved safety issues raised at the CP

stage and could be used to restrict NRC’s abilityto order backfits. Regulators and critics especiallyobject to the DOE bill because it allows the li-censees themselves to certify safety, with a limitedtime for the NRC staff to verify that certification,and no real opportunities for citizen participation.

Some utilities are not convinced that a one-stepprocess would be any more predictable than thecurrent two-step process in terms of requirementsfor a license and backfits. Furthermore, it is possi-ble that the proposed COL procedure, when cou-pled with hybrid hearings, would take longer thanthe current CP and OL process. Using procedural

changes to improve the management of the hear-ings and implementing site- and design-banking,which together would serve as a surrogate forone-step licensing, probably would do more toincrease the efficiency and predictability of licens-ing than a switch to a COL.

Current NRC regulations allow for design re-view prior to the filing of the CP application, butthe results of the review are not binding uponthe CP determination. Alternatively, reactor ven-dors can submit generic designs for approvalthrough rulemaking. Many industry analysts ar-

gue that reactor engineering has matured suffi-ciently to allow proapproval of standardized plantdesigns, or of major system or subsystem designs,and both the NRC and the DOE bills include pro-

visions for “design-banking.” Debate continues,however, on the degree of specificity that shouldbe required for proapproval of designs andwhether such approval would act as a disincen-tive to the continued improvement of designs.

Under the NRC legislative proposal, a binding

design approval valid for 10 years could begranted without reference to a particular site andcould be renewed for 5 to 10 years unless NRCfound that significant new safety information rele-vant to the design had become available. Thepublic would have an opportunity to request hy-brid hearings on the design before NRC grantedapproval. Issues related to the design could notbe raised in a subsequent CP, OL, or COL hear-ings unless the combination of a design with aparticular site resulted in new issues that had notbeen addressed in the design approval or therewas convincing evidence that reconsideration of

design issues was necessary.

The DOE bill also would allow utilities tochoose a preapproved plant or major subsystemdesign as an alternative to selecting a uniqueplant design. Design approvals would be subjectto hybrid hearings. Once approved, a designwould be valid for 10 years and then could berenewed for 10 years but would be subject to thesame backfitting requirements as normal plantsunder the DOE bill. Preapproved designs wouldbe incorporated into a CP or COL application,and the review of design issues in the hearings

would be strictly limited. The DOE bill would re-quire NRC to define the level of detail necessaryfor design approvals through the normal rulemak-ing process.

Proapproval of standard designs might make asubstantial contribution to a more efficient andpredictable licensing process by removing mostdesign questions from the licensing of a particularplant, but it is likely to be as controversial as theproposal for a COL. Issues include the degree ofspecificity required for design approval, the con-ditions and procedures under which the utility

or its contractors could deviate from a preap-proved design once construction has begun, andthe ability of NRC to order backfits on approveddesigns.




Nuclear critics are concerned that discussionof new or previously unresolved safety issueswould be foreclosed in the CP or COL hearingson preapproved standardized designs, especial-ly in light of the provisions that prohibit the rais-ing of generic safety issues in the licensing of par-ticular plants and of the provisions that shift the

burden of proof to the public to show that a pre-approved design does not meet current safetystandards. Proponents of this change argue thatproapproval of designs could improve the effec-tiveness of public participation in that it wouldallow earlier and more detailed discussion ofdesign issues in hearings without the time con-straints imposed by the licensing of a particularplant.

The critics also object to the length of time forwhich a design approval would be valid, giventhe frequency with which design changes havebeen instituted in the past, and to the subsidygranted by deferral of the application fee untilthe design is used. Furthermore, there is concernthat once a design has been approved, the vestedinterest in it would remove any incentives to im-prove it. However, as discussed in chapter 7, theindustry argues that its need to remain competi-tive with foreign countries should be incentiveenough.

In the present system, NRC approval of site suit-ability is not initiated until the CP application isdocketed, which places site review on the “crit-ical path” for reactor licensing. The existing NRC

regulations permit review of site suitability priorto filing of the CP application, but the outcomeof this review is not binding in the final CP deci-sion unless a special ASLB decision is obtained.Both the NRC and the DOE legislative packagesrecommend a procedure for binding early site ap-proval that would be independent of a CP appli-cation.

In the NRC legislative proposal, a site approvalthat does not reference a particular nuclear plantcould be granted for up to 10 years, with renewalpossible for 5 to 10 years. Federal, State, regional,

and local agencies, as well as utilities could ap-

ply for site approvals, thus encouraging broaderplanning. In the NRC bill, a site approval wouldnot preclude the use of the site for an alternativeor modified type of energy facility or for any otherpurpose. However, other uses not considered inthe original approval may invalidate the site per-mit, as determined by NRC. The public would

have an opportunity to request hybrid hearingson the site approval, but issues related to the sitewould be excluded from further licensing pro-ceedings unless matching the site with a particularplant design raised issues that were not consid-ered at the time of the site approval.

The DOE proposal is similar to NRC’s, exceptthe site-approval procedure in the DOE bill wouldnot allow alternative uses and would allow CPapplicants to perform limited construction activ-ities before issuance of a permit. A site approvalwould be valid for 10 years, with 10-year renew-

als. Under the DOE legislative proposal, the pub-lic could request hybrid hearings prior to NRCapproval of a site.

As with design approvals, OTA concludes thatsite-banking could improve the efficiency andpredictability of the licensing process by takingsiting out of the critical path entirely. As longas the site-approval process allows adequate op-portunity for public participation and ensuresconsideration of issues related to the combina-tion of a particular site and design prior to issu-ance of a CP, binding early site approval shouldnot be a controversial change. In fact, severingsite approval from the CP could facilitate earlierand more substantive public participation. Theprincipal objections nuclear critics have to thesebills are the length of time for which approvalsare valid (including renewals, 20 years in the NRCbill and an indefinite period in the DOE) and thesubsidy introduced by deferring the applicationfee until the site actually is used or the approvalexpires. Furthermore, the selection of particularsites—whether they are matched with a plant ornot—will remain controversial, as discussed in

chapter 8.




OTHER NRC RESPONSIBILITIES

In licensing a nuclear powerplant, NRC is re-quired to make several determinations that arenot related directly to safety. These include cer-tification of the need for power from the plant(required under NEPA) and of compliance with

antitrust laws,

There is general agreement that NRC is poorlyequipped to judge need for power on a local orregional basis, and therefore that it is a waste ofstaff resources to make such a determination.Moreover, at least 45 States already require otheragencies to determine the need for power eitherin the certification or licensing of powerplants,in rate cases, in the approval of financing, or inan independent planning process (20). Further-more, evaluations of the need for power and thechoice of alternative types of generating technol-

ogies can take up hearing time and staff time thatcould be better spent in the analysis of safety anddesign issues.

Both the .NRC and the DOE legislative packagesprovide for binding NRC acceptance of a needfor power determination made by a Federal, Stateor other agency authorized to do so. The NRCbill also provides for acceptance of other agen-cies’ rulings on alternative sources of generatingcapacity. Only where no other agency is requiredto make such a determination would NRC per-form a de novo review of the need for power.

In both bills, these provisions are embedded with-in the section on a one-step licensing process,but they could be separated out. Because eachagency is required under NEPA to make thesedeterminations, legislative action would be re-quired to delegate that authority to the States orother Federal agencies. it is possible that this pro-vision would result in expanded opportunities forpublic participation in the discussions of the needfor power and choice of technology. However,neither bill sets minimum standards for publicparticipation in delegating this authority to theStates, nor do the bills mandate consideration of

the full range of alternatives, as required in NEPA.Under current practice the Department of Jus-

tice performs a comprehensive review of licenseapplications for compliance with antitrust laws.

Although NRC weighs the opinion of the justiceDepartment heavily in its determination, theCommission remains responsible for the final an-titrust decision. As in need for power, it may notbe appropriate for NRC to devote staff resources

to antitrust law. One option is for NRC to adoptthe Justice Department’s decision on antitrust un-less an affected party objects within a specifiedtime after notice of the decision. If the objectionis found to have merit, then NRC could remandto Justice for further consideration or do an in-dependent review. Legislative action would berequired to delegate this authority to the JusticeDepartment.

The U.S. General Accounting Office (GAO)also has recommended that the NRC provide bet-ter coordination with State and local governments

in NEPA reviews. At least 23 States have statutesrequiring preconstruction environmental reviewssimilar to those required under NEPA, but NRC’sNEPA regulations make no provision for coordi-nation with the States or for eliminating duplica-tion of efforts. GAO recommends NRC work jointly with all the States to identify common legaland procedural requirements as a first step incoordinating environmental reviews (2).

Finally, it has been suggested that introducinga little flexibility into the concept of exclusiveFederal jurisdiction over reactor regulation wouldgo a long way toward alleviating State and localconcerns and improving public acceptance. Forexample, Oregon has a memorandum of un-derstanding with NRC that sets forth “mutuallyagreeable principles of cooperation between theState and NRC in areas subject to the jurisdictionof the State or the NRC or both. ” This memoran-dum is intended to minimize duplication of ef-fort, avoid delays in decisionmaking, and ensurethe exchange of information that is needed tomake the most effective use of the resources ofthe State and NRC. To accomplish these ends,the memorandum provides for potential future

subagreements in areas of mutual concern, in-cluding siting of nuclear facilities, water quality,nuclear plant operation, radiological and environ-mental monitoring, decommissioning of nuclearplants, emergency preparedness, personnel train-




ing and exchange, radioactive material transpor-tation, and other areas. Subagreements adoptedto date include a protective agreement for the

SAFETY

One concept that has attracted much attentionin discussions of backfitting and other changesin the NRC technical regulations is the use of safe-ty goals* to establish safety requirements andgage the need for changes in those requirements.

NRC currently is developing a safety goal poli-cy, and the DOE legislative proposal emphasizesthe importance of this effort by endorsing theCommission’s efforts. The DOE bill would requireNRC to report to Congress within 1 year on itsprogress in developing and implementing a safety

goal policy. The NRC proposal is described be-low.

NRC Safety Goal Proposal

NRC has issued a policy statement on safetygoals for nuclear powerplants that is being usedon an experimental basis (25). It currently playsno part in licensing decisions, and license appli-cants do not have to demonstrate compliancewith it. If the proposed policy receives sufficientlyfavorable response, NRC will consider amending

its regulations to include safety goals in licens-ing decisions.

In developing a safety goal policy, NRC consid-ered qualitative goals that would interpret theAtomic Energy Act’s standard of adequate pro-tection of public health and safety, as well asquantitative goals that could provide a more ex-act standard against which risks could be meas-ured. Qualitative goals were adopted to lend NRCsafety decisions “a greater coherence and pre-dictability than they presently appear to have,”supported by numerical guidelines as goals orbenchmarks (25). The NRC report notes that this

*NRC defines a safety goal as “an explicit policy statement onsafety philosophy and the role of safety-cost tradeoffs in the NRCsafety decisions” (25).

exchange of information, and an agreement onresident inspectors at the Trojan plant, the onlynuclear powerplant in Oregon (11).

GOALS

approach allows it to capture the benefits of qual-itative goals and quantitative guidelines in meas-uring performance while avoiding the vaguenessof qualitative goals without numerical guidance.It does not lock NRC into quantitative goals thatmay not be able to yield technically supportableresults given the uncertainties inherent in quan-titative risk assessment.

The qualitative safety goals established in the

NRC policy statement are:

Individual members of the public should be

provided a level of protec tion from the co nse-quences of nuclear powerplant accidents such

that no individual bears a significant additionalrisk to life and health

Societal risks to life and health from nuclearpowerplant accidents should be as low as rea-

sonably achievable and should be comparable

to o r less than the risks of ge nerating elec tricityby viable competing technologies (25).

The intent of the first safety goal is to requirea level of safety such that individuals living orworking near nuclear powerplants should be able

to go about their daily lives without special con-cern by virtue of their proximity to such plants.The second safety goal limits the societal risksposed by reactor accidents and includes an im-plicit benefit-cost test for safety improvements toreduce such risks.

These goals focus on nuclear powerplant acci-dents that may release radioactive materials tothe environment. They do not address risks fromroutine emissions, from other parts of the nuclearfuel cycle, from sabotage, or from diversion ofnuclear material. The policy statement notes thatthe risks from routine emissions are addressed incurrent NRC practice through environmental im-pact assessments that include an evaluation of theradiological impacts of routine operation of theplant on the population around the plant site. For



Ch. 6—Regulation of Nuclear Power q 173

all plants licensed to operate, NRC has found thatroutine operations will have no measurable radio-logical impact on any member of the public.Therefore, the object of the experimental policyis to develop safety goals that limit to an accept-able level the additional potential radiological riskthat might be imposed on the public as a result

of accidents at nuclear powerplants.In establishing the numerical guidelines to sup-

port these safety goals, NRC noted that progressin developing probabilistic risk assessment (PRA)techniques and in accumulating relevant datasince the 1974 Reactor Safety Study (24) has ledto recognition that it is feasible to begin to usequantitative assessments for limited purposes.However, because of the sizable uncertainties stillpresent in the methods and the gaps in the database–essential elements in gaging whether theguidelines have been met–NRC indicated that

the quantitative guidelines should be viewed asgoals or numerical benchmarks that are subjectto revision as further improvements are made.Many of the participants in the Safety Goal Work-shops held by NRC agreed that quantitative goalswere not feasible at this time, but numericalguidelines could be used to support qualitativegoals. Finally, in setting the numerical guidelines,NRC specified that no death attributable to a reac-tor accident ever will be “acceptable” in thesense that the Commission would regard it as aroutine or permissible event. NRC intends that

no such accidents occur but recognizes that the

possibil i ty cannot be eliminated entirely.With these caveats, NRC established four ex-

perimental numerical guidelines: two for individ-

ual and societal mortality risks for prompt and

delayed deaths; a benefit-cost guideline for use

in decisions on safety improvements that wouldreduce those risks below the levels specified inaccordance with the longstanding regulatory prin-ciple that risks from nuclear power should be “aslow as reasonably achievable”; and a plant per-formance guideline that proposes a limitation onon the probability of a core melt as a provisionalguideline for NRC staff use in reviewing and eval-uating PRAs of nuclear powerplants. These guide-lines are:

The risk to a n individua l or to the po pula tion

in the vicinity of a nuclear powerplant site of

prompt fatalities that might result from reactor

accidents should not exceed 0.1 percent of the

sum of p romp t fata lity risks resulting from otheraccidents to which members of the U.S popula-

tion a re generally exposed.The risk to a n individua l or to the p op ulation

in the area near a nuclear powerplant site of

cancer fatalities that might result from reactor ac-

cidents should not exceed 0.1 percent of thesum of cancer fatality risks resulting from all

other causes.

The b enefit of an inc remental reduc tion of risk

below the numerical guidelines for societal mor-

tality risks should be compared with the associ-

ate d c osts on the b asis of $1 ,000/ ma n-remaverted.

The likelihood of a nuc lear reac tor acc identthat results in a large-scale core melt should nor-

ma lly be less than 1 in 10,000 per yea r of rea c -tor operation (25).

In its experimental safety goal proposal, NRC

left open a number of questions for future con-sideration. These include: whether the benefitside of the tradeoffs should include the economicbenefit of reducing the risk of economic loss dueto plant damage and contamination outside theplant; whether a numerical guideline on availa-bility of containment systems to mitigate the ef-fects of a large-scale core melt should be added;and whether there should be a specific provisionfor risk aversion and, if so, what it should be. Inaddition, the proposal sought further guidanceon developing a detailed approach to implement-ing the safety policy, including decision making

under uncertainty; resolving possible conflictsamong quantitative aspects of issues; the ap-proach to be used for accident initiators that aredifficult to quantify (e.g., seismic events, sabotage,human and design errors); the terms for defini-tion of the numerical guidelines (e.g., median,mean, 90-percent confidence); and identifyingthe individuals to whom the numerical guidelinesshould be applied (e.g., the individual at greatestrisk, the average risk).

Shifting from prescriptive regulation to a safe-ty goal approach could have far-reaching conse-

quences. Such a change might contribute to amore favorable regulatory environment for thenuclear utilities since the number and unpredict-ability of regulatory actions probably would bereduced. Furthermore, utilities would be allowed




to select the least costly route to compliance, withresultant gains in efficiency. Another result of thesafety goal approach might be to encourage di-versity and innovation in developing alternativesfor improving safety. Such activities, however,may not be consistent with the standardizationof nuclear powerplants (4).

The proposed safety goals and numerical guide-lines are not free of controversy. The proposedguidelines have been criticized as being “tooremote from the nitty-gritty hardware decisionsthat have to be made every day by designers,builders, operators, and regulators to be of muchuse” (25). Most regulators and industry repre-sentatives agree that while, in principle, it wouldbe nice to be able to use overall goals to supplantthe myriad specific decisions NRC must makeabout the adequacy of hardware and procedures,they find the proposed goals too generaI and

abstract to provide specific guidance for dealingwith practical questions, and withhold judgmenton whether they will prove useful. As one Com-missioner noted, the only reliable guides to reac-tor safety remain time-tested engineering princi-ples:

redundant and diverse means of protectionagainst core damage, sound containment, suffi-

cient d istance from pop ulated a reas, effec tiveemergency preparedness, and, of course, careful

attention to quality assurance in constructionand operation. To provide guidance to the NRC

technical staff and the nuclear industry, and toinform the pub lic, the Co mm ission should d istillits expe rienc e a nd state c learly a nd succinc tlythat each of these [engineering] principles must

be satisfied separately, and how this is to bedone. Unfortunately the Commission seems tobe on a n op po site course (25).

The nuc lear c ritics objec t more strong ly to the

safety goal proposal, arguing that to adopt goals

with no viable means of confirming their achieve-

ment is a useless exercise. They do not believethere is any immediate prospect of PRA being de-veloped sufficiently to provide a means of con-firmation. Therefore, the critics argue that it is notfeasible to use quantitative guidelines for limitedpurposes, and NRC only misleads the public insaying that PRA calculations will be used to sup-port qualitative goals.

LICENSING FOR ALTERNATIVE REACTOR TYPES

Nuclear powerplant licensing experience in theUnited States, for the most part, is based on the

LWR design concept. The exceptions are the FortSt. Vrain high-temperature gas-cooled reactor(HTGR), which achieved full power in 1981, andthe Clinch River breeder reactor. Yet variationson the LWR and other reactor design conceptsare attracting attention for their possible safetyand reliability advantages over the LWR, asdiscussed in chapter 4. Given the extent of thelicensing and regulation experience with LWRs,it is reasonable to question whether a shift to adifferent design would entail substantial changesin the regulatory process, such that the sameproblems encountered in the regulation of LWRs

would be repeated with alternative reactors, andwhether the development of a licensing processfor such reactors would delay their implementa-tion.

Small LWRs contributed greatly to the originaldevelopment of commercial nuclear power in the

United States. However, as operating experiencegrew, apparent economies of scale motivated util-ities to purchase larger reactors. Today the normis over 1,000 megawatts electric (MWe), but in-terest in smaller reactors is reemerging, primari-ly for financial and system flexibility reasons. Ashift to smaller reactors could not be accom-plished by replicating existing small plantsbecause the designs of those plants do not meetall current safety requirements. NRC has estab-lished a systematic evaluation program specifical-ly to review these older designs and improve theirsafety where possible. New small reactors would

require new designs based on current NRC reg-ulations, although such designs would notnecessarily differ substantially from large LWRsexcept in the size of the core and other plant



Ch. 6—The Regulation of Nuclear Power q175

components. Thus the regulatory process prob-ably would be similar to that for current largeLWR designs, including the potential for backfits,unless small LWRs were standardized within thecontext of proapproval of designs.

The high temperature gas reactor has littleoperating experience in the United States. Theprimary safety concerns are quite different fromthe LWR and have not been studied as intensive-ly. As a result, the potential for the emergenceof significant unforeseen safety concerns prob-ably is higher than for the LWR. On the otherhand, inherent characteristics of HTGRs makethem less susceptible to certain types of accidentsthat can progress more quickly or have more seri-ous consequences in a LWR. This eventually maysimplify the licensing process after any initialproblems are resolved.

During the early 1970’s, several utilities made

CP applications for HTGRs. As a result, NRCmade a significant effort to formalize design re-quirements and establish review plans for theHTGR. Nevertheless, several years would be re-quired to make the regulatory process for thisdesign as mature as that for LWRs. Backfitting re-quirements for the HTGR are uncertain butshould be reduced through the operating expe-

CHAPTER 6

1. Cohen, S. C., Lob bin, F. B. and Thomp on , S. R,“ The Future of Co nventiona l Nucle ar Power: Nu-c lear Rea c tor Reg ulation, ” SC&A, Inc., Dec. 31,1982.

2. Comptroller General of the United States, “Reduc-

ing Nuc lear Powe rplant Lea dtime s: Many Ob sta-

cles Remain, ” EMD-77-15, Mar. 2, 1977.3. Crowley, j. H., and Griffith,, j. “ U.S. Construction

Co st Rise Threa tens Nuc lea r Optio n,” / V u c / e a r

Engineering /ntemationa/, June 1982.

4. Fisch o ff, B., “Ac cep tab le Risk: The Ca se of Nuc le-

ar Power,” Journal of Policy Analysis and Manage-

ment, vol. 2, No. 4, 1983.

5.Freeman, S. D.,

“ The Future of the Nuc lear Op -t i on , ” Environment, vol. 25, No. 7, 1983.

rience of the Fort St. Vrain plant and the 900MWe prototype that DOE is sponsoring.

The heavy water reactor, as represented by thewell-proven Canadian CANDU design, has attrac-tive safety and reliability features, but licensabilityis a major constraint on the adoption of this de-sign by U.S. utilities. The NRC requirements forseismic protection and thicker pressure tube wallswould require design changes in the CANDU thatmight reduce its efficiency and could lead tobackfits until these changes were proven. NRCwould have to establish new design criteria andstandard review plans before a heavy water reac-tor could be licensed.

The Process Inherent Ultimate Safety (PIUS)reactor is the least developed of the alternativedesign concepts. It has readily visible safety ad-vantages, but they might not be accounted forin the regulatory process until significantoperating and construction experience is estab-lished. If PIUS is forced to include all the engi-neered safety features of the LWR, it is not likelyto be competitive. Successful development ofPIUS, therefore, depends on NRC determiningwhat level of safety is appropriate and creditingthe inherent safety features of PIUS during thedesign approval and licensing process.

6.

7.

8.

9.

104

Hodel, D., statement before the Subcommittee onNuclear Regulation of the Committee on Environ-

ment and public Works, U.S. Senate, May 26,1983.Kemeny, j. G., “Report of the president’s Commis-sion on the Ac c ident a t Three Mile Island , ” Oc -tober 1979, Washington, D.C.Mat iel , L., Minnek, L. W., and Levy, S., “Summary

of Disc ussions with Utilities and Resulting Conc lu-sions, ” Electric Power Research Institute, June

1982.

“NRC Pulls the Plug on the Zimmer Nuclear Pow-er Plant,” The Energy Dai/y, vol. 10, No. 215, Nov.

16, 1982.Office of Tec hno log y A ssessme nt, U.S. Co ngress,




11,

12 ,

13 ,

14 .

15 .

16 .

17 .

18 .

19 .

Nuclear Powerplant Standardization–Light Water

Rea c tors, OTA-E-1 34, June 1981.Oregon Department of Energy and the U.S. Nucle-

ar Regulatory Commission, Memorandum of Un-derstanding, Jan. 4, 1980.Patterson, D. R., “ Revitalizing Nuc lea r PowerPlant Design and Construction: Lessons Learned

by TVA,” Oc tob er 1981.

Phung, D. L., “ Light Water Rea c tor Safety Sincethe Three Mile Island Accident,” Institute for En-

ergy Analysis, July 1983.

Rogovin, M., et al., “ Three M ile Island: A Rep ort

to the Commissioners and to the Public,” U.S. Nu-c lear Reg ulatory Co mm ission, NUREG/ CR-l 250,February 1980,

SC&A, Inc., “Nuclear Reactor Regulation and Nu-

clear Reactor Safety, ” Sept. 6, 1983.Title 50 of the Cod e o f Fed era l Reg ula tions, Pa rt50.1 09(a).

Title 10 of the Cod e o f Fed era l Reg ulatio ns, Part

50, Appendix E.United Engineers & Constructors, Inc., “Nuclear

Regulatory Reform,” UE&C 810923, September1981.

United Engineers & Constructors, Inc., “Regula-tory Reform Case Studies,” UE&C 820830, August1982,

20.

21.

2 2

23.

24 .

25 .

26 .

U.S. Depa rtmen t of Energy, “ The Ne ed for Pow erand the Cho ice of Tec hnolog ies: State Dec isionson Electric Power Facilities, ” DOE/EP/ l 0004-1,

June 1981.U.S. Department of Energy, “Nuclear Licensingand Regulatory Reform Act of 1983,” draft bill sub-

mitted to Congress, March 1983.U.S. Nuclear Regulatory Commission, “Nuclear

Power Plant Licensing: Opportunities for improve-ment,” NUREG-0292, June 1977.U.S. Nuclear Regulatory Commission, “NuclearPower Plant Licensing Reform Act of 1983,” draftbill submitted to Congress, February 1983.U.S. Nuclear Regulatory Commission, “ReactorSafety Study,” WASH-1400, October 1975.U.S. Nuclear Regulatory Commission, “Safety

Goals for Nuclear Power Plants: A DiscussionPaper,” NUREG-0880, February 1982.

U.S. Nuclear Regulatory Commission, “A Surveyby Senior NRC Ma nage ment to Obtain Viewpo ints

on the Safety Impact of Regulatory Activities FromRepresentative Utilities Operating and Construct-

ing Nuclear Power Plants,” NUREG-0839, August1981.



Chapter 7

Survival of theNuclear Industry in the

United States and Abroad

Photo credit: Electricite de France I

Four nuclear units at Dampierre, France



Contents

Page Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..179

The Effects in the U.S. Nuclear lndustry of No, Few or Delayed

New-Plant Orders 1983 to 1995..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ....179Reactor Vendors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ...................180Nuclear Component Suppliers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .............182Architect-Engineering Firms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .................183The lmpact on Nuclear Plant Operation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ......183The lmpact on Future New Construction of Nuclear Plants . .....................186

The Prospects for Nuclear Power Outside the United States . ......................189The Economic Context for Nuclear Power . . . . . . . . . . . . . . . . . . ..................189Public Acceptance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..190

Foreign Technical Experience: Plant Construction and Reliability . . . . . . . . ..........191Licensing and Quality Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ....191Overall Nuclear Development. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..,.........194Implications for the U.S. Nuclear Industry. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..197The U.S. Industry in an International Context . ................................199

Nuclear Proliferation Considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .......200Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ................203

Chapter 7 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..........................207

Table No. Page

23. NSSS/AE Combination of Light Water Reactors Under Constructionor On Order As of 1981 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..................180

24. Forecasts of Installed Nuclear Capacity in OECD Countries . ...................18925. Sa m p le Co nstruc tion Time s for Nuc lea r Plan ts in Va rious Coun tries. . ............19226. Operating Performance by Country . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ... ... ... ..19527. Categories of Foreign Nuclear Programs in Western Europe and the Asia-Pacific Region 194

28. Structure of Electric Utility in Main Supplier Countries and the Distribution ofAuthority Over Key Decisions Influencing Financial Health of Utilities . ..........196

29. Estimated Reactor Export Market in Units . . . . . . . . . . . . . . . .. .. .. .. .. ... ... ....20 130. No-Nuclear Weapons Pledges in Effect in 1981 . .............................2027A. Onsite and Offsite Nuclear-Related Job Vacancies at lNPO Member Utilities, Mar. 1,1982204

7B. Nuclear-Generating Capacity Outside the United States . ......................2057C. Usability of Nuclear Materials for Nuclear Weapons or Explosives. . . . . .....206

Figures

Figure No. Page

34. Engineering Manpower . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ...................18135. Estimates of Additional Manpower Requirements for the Nuclear Power industry, 1982-91186

36. Nuclear Engineer Degrees: Foreign Nationals and U.S. Citizens . ................18737. Average Annual Increase in Electricity Demand: France, West Germany,and United Kingdom . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .......190

38. Engineering and Construction Man-Hours per Megawatt of Capacity .. ... ... ... ..194



Chapter 7

Survival of theNuclear Industry in the

United States and Abroad

INTRODUCTIONWhether or not utility executives order more

powerplants (given all the uncertainties and dis-incentives described in earlier chapters) has direct

implications for the U.S. nuclear industry and its

ability to remain viable as a source of nuclearpowerplants both within the United States andabroad. This chapter examines the consequencesfor different parts of the U.S. industry of a longperiod with no orders for new plants or a periodin which orders for new plants follow a long de-lay. The chapter then surveys the prospects for

nuclear power abroad and the likelihood of U.S.exports as well as the possibility that the UnitedStates might be able to turn to foreign suppliersas future sources of the technology.

Although there are no strict parallels betweenthe U.S. nuclear industry and that of any othercountry, there nonetheless is much to be learnedfrom foreign experience. Many of the same prob-lems faced by the U.S. industry are being facedelsewhere: public opposition to nuclear power,slow demand growth, and the difficulty of con-trolling cost and time overruns in nuclear plantconstruction. Understanding how these and otherproblems are being coped with in each country,provides some perspective on the U.S. situation

and information on approaches that might besuccessful in the United States.

THE EFFECTS IN THE U.S. NUCLEAR INDUSTRY OF NO,FEW OR DELAYED NEW-PLANT ORDERS 1983 TO 1995

The nuclear industry may be portrayed as amonolith by its critics. I n fact, however, it hasalways been a loose-knit group of several hun-

dred businesses and organizations, given whatcohesion it has by the demands of a difficult tech-nology and the need to develop a coordinatedresponse to critics. Today, the industry consistsof the 59 public and private utilities that are theprincipal owners of nuclear powerplants in opera-tion or under construction, 4 reactor manufac-turers also known as nuclear steam supply sys-tems (NSSS) vendors, 12 architect-engineering(AE) firms with a specialty in nuclear design andconstruction, about 400 firms in the United Statesand Canada qualified to supply nuclear compo-nents, and several hundred nuclear service con-

tractors. Table 23 shows the combinations ofreactor manufacturers and AE firms for plantsunder construction or on order as of the springof 1981.

Of about 90,000 employees of the nuclear in-dustry, about half operate and maintain commer-cial power reactors (as well as some test and

research reactors), a quarter are engaged in reac-tor and reactor component manufacturing, anda quarter are engaged in design and engineer-ing of nuclear facilities (other than designassociated with reactor manufacture) (4).

Companies and organizations in each of thesesectors must develop strategies for coping withthe likelihood of no new orders for nuclear plantsfor 3 to 5 years and the possibility of no or veryfew orders for 5 or more years after that. In acomprehensive study for the U.S. Department ofEnergy (DOE), the S. M. Stoner Corp. (37) as-

sessed the impact on NSSS vendors and compo-nent suppliers of three possible futures:

qa slowly increasing projection of: no ordersuntil 1986, an average of two to three a year

179



180 q Nuclear Power in an Age of U n c e r t a i n t y

Table 23.—NSSS/AE Combination of Light Water ReactorsUnder Construction or On Order As of1981

Reactor vendors

Architect/ General Combustion Babcock &engineering firms Westinghouse Elec tr ic Eng ineering Wilcox

Bechtel . . . . . . . . . . . . . . . . . 6 10 6 5

Burns & Roe . . . . . . . . . . . . . — 1 1 —

Black & Veatch. . . . . . . . . . . — 2 — —

Brown & Root . . . . . . . . . . . . 2 — — —Ebasco . . . . . . . . . . . . . . . . . 4 1 4 —

Gilbert/Commonwealth. . . . 1 2 — —

Gibbs & Hill . . . . . . . . . . . . . 2 — — —

Gilbert Associates. . . . . . . . — — — —

Utility Owner . . . . . . . . . . . . 7 — 6 —

Fluor Power Services . . . . . — — — —

Sargent & Lundy . . . . . . . . . 8 7 — —

Stone & Webster . . . . . . . . . 5 6 2 2

United Engineers . . . . . . . . . 2 — — 2

Tennessee ValleyAuthority, . . . . . . . . . . . . . 3 6 2 2

SOURCE Nuclear Powerplant Standardization Light WaterTechnology Assessment, OTA-E-134, April 1981)

until 1989, and six to eight orders a year afterthat;q no orders until the early 1990’s; andqno orders until 1988 or 1989 and an average

of one a year for 5 years after that.

The findings of the Stoner study are echoed inthe results of 35 interviews conducted by OTAwith representatives of reactor vendors, nuclearsuppliers, AE firms, utilities with nuclear plants,and industry analysts and regulators. Further in-sights are available from several assessments ofpersonnel needs for the industry (4,9,16,18).

Reactor Vendors

No new nuclear reactors are now being built.The nuclear business for the four reactor vendorscurrently consists of assembling at site, fuel load-ing and services, and the latter two will continueregardless of what happens to new orders. Figure34 shows one vendor’s prediction of the needfor engineering manpower through the 1980’s.Engineers will be needed for services to operatingplants and fuel loading. Manpower to handlechanges in existing plants, and rework in plants

under construction, will initially increase but thendiminish. The need for engineering manpowerto design new NSSS will practically disappear.

Refueling, which occurs in each plant approx-imately every 18 months, is a demanding task re-

Reactors (Washington, D.C. U.S. Congress, Office of

quiring sophisticated skills and a sound knowl-edge of nuclear physics. Used fuel rods are re-moved and new fuel rods are inserted among par-tially used fuel rods, and the array of both freshand older fuel rods is then reconfigured to pro-vide maximum nuclear energy. Vendors expectalso that spent fuel management will also be acontinuing source of business.

Vendors are now competing for the nuclearservice business in an arena once dominated bythe nuclear service consultants. The Stoner reportestimates further that backfits and rework mayrequire 30 to 50 man-years of contracted engi-neering work per operating plant with a total de-mand of 3,000 to 6,000 technical people per year.(37) The vendors are uncertain, however, if thecurrent level of backfits, stimulated largely byrequirements following the Three Mile Island ac-cident, will continue beyond the next few years,and, at the same time realize that over the longrun the continued cost of backfits will discouragenew orders.

The only current new plant design activities are joint ventures by both GE and Westinghouse withJapanese companies. The Westinghouse project

is being aimed at both the domestic and exportmarkets and is being developed in consultationwith the Nuclear Regulatory Commission (NRC).(See ch. 4.) The GE project is being developedfor Japan only and not for future U.S. licensing.



Ch. 7—Survival of the Nuclear Industry in the United States and Abroad • 181

Figure 34.-Engineering Manpower

100

75

50

25

0

Operating plant services

Nuclear fuel

Changes and delays

Original nuclear steamsupply system scope

Year supplier,

SOURCE A S Program (Washington, D C.: Department of Energy, November 1981)

Both companies hope that an export marketwill sustain some of their design and manufac-turing capability. The likely export market (de-scribed later in this chapter), however, has shrunkto only a fraction of what was projected 5 yearsago and is currently substantially less than whatcan fully use worldwide manufacturing and de-sign capability.

For U.S. companies to compete successfully re-quires not only continued technical success (as

is being attempted in these joint ventures) but alsopossible modifications in U.S. export financingand nonproliferation policies (25). There is evi-

dence that some orders have already been lostbecause U.S. vendors are losing their reputationfor up-to-date technology. As a Finnish sourcetold Nucleonics Week: “Why should Westing-house put in millions (of dollars) for R&D if theydon’t have business prospects. That is one reasonwhy we are not studying their (U. S.) reactors ina [plant-purchase] feasibility study” (29).

Moreover, future export orders are likely to in-volve reduced U.S. manufacturing demand since

many of the countries most likely to pursue nu-clear programs have nuclear import policies de-signed to promote domestic industries. Other ad-

25-450 0 - 84 - 13 : QL 3




vanced countries with nuclear programs, espe-cially Japan, are also likely to bid successfully forcomponent manufacturing business (25,37).

The current backlog of NSSS manufacturingwork is scheduled for completion in 1984. AllU.S. vendors have taken steps to close or moth-

ball many of their manufacturing facilities, or toconvert them for other uses. It has been estimatedthat announced facilities closings and consolida-tions have already reduced by two-thirds the U.S.capacity to supply nuclear powerplants. Somevendors are maintaining their technical capabil-ity with nuclear work for the U.S. Navy, DOE,or research and development (R&D) sponsoredby the Electric Power Research Institute (EPRI)(37).

Vendors are already feeling the effects of ashrinkage in nuclear component suppliers onwhich they base future standardized NSSS de-signs. Currently, each vendor purchases compo-nents from about 200 qualified nuclear suppliers.Several vendors estimate that the number of sup-pliers will drop by two-thirds in 3 tos years, leav-ing the vendors dealing with a much higher pro-portion of sole source suppliers (37). Vendorsfaced with this situation are considering variousresponses, such as manufacturing their owncomponents, encouraging less qualified suppliersto upgrade their products and get them certified,and developing new sources of foreign supply.

Nuclear Component SuppliersThe impact of the shrinkage in new orders is

most dramatic on component suppliers. Somecompanies supply components used both fornew plants and for backfit and spare parts forplants in operation. These companies expect tokeep their businesses going. Many companies,however, supply only components for newplants. Some of these produce nuclear compo-nents that are identical or very similar to non-nuclear components except for quality-controldocumentation. These companies can be ex-pected to maintain their nuclear supply lines.Others, however, produce very specializednuclear components that require separate testingand manufacturing facilities. Many of these facil-ities are now closed or mothballed (37).

At present the number of component suppliersappears to be declining slowly. One clear signis the decision by suppliers not to renew the “N-stamp,” a certificate issued by the American Soci-ety of Mechanical Engineers (ASME) for the man-ufacturer of nuclear plant components. N-stampsare not specifically required by the NRC for the

manufacture of safety-related nuclear-plant com-ponents. However, they are required by someStates, and their use certifies that certain NRCquality-assurance requirements have been met.

The number of domestic firms holding N-stamps has dropped by about 15 to 20 percentsince 1979, the year of the accident at Three MileIsland, and the drop would probably be greaterif the renewal were annual instead of triennial(21). By contrast, foreign N-stamp registration has

held stead y. By the end of 1982, some 400 com-panies in the United States and Canada held

about 900 N-stamps, according to ASME. An ad-ditional 50 companies held about 100 certificatesfor Q-system accreditation on nuclear-grade ma-terials. Overseas, about 70 companies held about100 N-stamps, and about 20 companies heldabout so Q-system certificates (21).

Maintaining an N-stamp requires both person-nel and money. Thus, in the absence of new nu-clear business, many smaller companies have de-cided they cannot justify the costs. In additionto the $5,000 to $10,000 that must be spent forASME certification (renewable at the same cost

every 3 years), there is also the need to dedicatepart of the plant and at least one or twoemployees to the intricate paperwork that accom-panies each N-stamp component. In total, costestimates for maintaining a stamp range from$25,000 to $150,000 a year (21). Suppliers saythat no other work, including contracts for theNational Aeronautics and Space Administration(NASA) and the U.S. Navy nuclear program, re-quires such a detailed paper trail. “1 make a valvethat sells for about $300” one supplier said. “Ifit has an N-stamp I have to charge $4,000 for thesame valve. And with low volume, I suppose I

should charge even more” (21).So far, the reduction in N-stamps has not been

as rapid as the lack of new orders might suggest.Part of the reason may be a habit of looking tothe future that has been characteristic of the



Ch. 7—Survival of the Nuclear Industry in the United States and Abroad q 18 3

nuclear industry since the beginning. Some sup-pliers evidently believe that the N-stamp impartsa certain status to a nuclear supplier’s operations,and those who must consider letting their certifi-cation expire say they would do so reluctantly.“It’s a nice marketing tool,” one supplier said,“even when you’re selling non-nuclear items.

And it’s good discipline for a company to haveit” (21). Some suppliers and utilities report thatthey must persuade their subcontractors to keepthe stamp. “We’re giving companies [with N-stamps] our nonnuclear business, just to helpthem along, ” one utility executive said. Anotherchallenge is to prevent market entry by foreigncompanies. “If equipment from overseas be-comes standard,” one supplier said, “we’ll neverget that business back” (21).

For many suppliers it will be almost impossi-ble to obtain nuclear qualification for new prod-

uct lines. For some product lines, 1 to 5 yearswouId be needed to carry out the necessary tests.Maintaining an older nuclear-qualified productline alongside a newer nonnuclear product linewill be difficult for those suppliers with a prepon-derance of nonnuclear business. Since nonnucle-ar business is likely to respond more quickly toan increase in general business investment of therecession than is nuclear business, there may bepressure to drop the nuclear product lines. Theexistence of nuclear components in 35 gigawatts(GW)* of partially completed but canceled nu-clear plants is viewed as a further damper on the

nuclear component business even though onlysome of this equipment is expected to be suf-ficiently maintained and documented enough tobe usable (see advertisement). For all thesereasons, there may be a far more rapid decreasein suppliers over the next 3 to 5 years than overthe past 3 years, possibly down to a third of thepresent number (37).

Architect= Engineering Firms

AE firms have substantial work for the next fewyears finishing the plants under construction, in-

stalling backfits and dealing with special problemssuch as steam generators. One promising con-

———.“One GW = 1 GWe = 1,000 MWe (1 ,000,000 kWe) or slightly less

than the typical large nuclear powerplant of 1,100 to 1,300 MW.

cept for interim survival involves “recommis-sioning” nuclear stations—installing some newcomponents to extend their operating lives by 10to 20 years. Like the reactor vendors, AE firmscomplain of reduced sources of supply for nu-clear-grade components and materials. And, likethe reactor vendors, they are moving outside their

specialties to bid on nuclear services (e. g., emer-gency planning) and rework proposals.

Most AE firms also have large amounts of busi-ness stemming from major construction projectsother than nuclear: cogeneration, geothermal,and coal technologies; petrochemical plants; in-dustrial process heat applications; and conven-tional fossil powerplants. During the 1981-83recession, business in these areas was no morerobust than the firms’ nuclear work. One AE ex-ecutive said, “As it is now, we can’t move ournuclear people to nonnuclear projects just to

keep them in-house. There isn’t much nonnucle-ar work around either” (21). Several firmsreported they expected their nonnuclear workto pick up long before their nuclear work (21).

Much of the project management and con-struction skills used on other types of large con-struction projects are also required for nuclearprojects. These skills will be available as long asthe AE firms have experience in major construc-tion projects. The design and project manage-ment skills unique to nuclear projects area smallproportion of the total work force.

Some firms are taking losses to keep theirskilled nuclear people employed because theyestimate that retraining would ultimately costmore. Architects are working as draftsmen, forexample, and skilled machinists are cutting andstacking sheet metal. Layoffs have not been nec-essary, one AE executive said, because employ-ees are retiring early or quitting to move to fieldswith more growth potential and less regulation,such as military R&D (21).

The Impact on NuclearPlant Operation

The halt in nuclear plant orders and uncertainprospects for new orders have had discernible



Ch. 7—Survival of the Nuclear Industry in the United States and Abroad q 185

effects on the utilities’ experience in keeping theirexisting nuclear plants staffed and maintained.The effects are most noticeable in two areas: ob-taining component parts and services, and fillingcertain key jobs.

Component Costs and Delays. -With the

decrease in nuclear component suppliers de-scribed above, utilities report an increase in thenumber of sole source suppliers and a resultingupward pressure on prices. One utility reportedthat sole source suppliers received 40 to 50 per-cent of 1982 contract dollars. A more typicalrange reported by utilities was 25 to 30 percent,an increase from 15 to 20 percent a decade ago(21).

In a few cases, utilities report that prices of serv-ices and components are falling because of in-creased competition. Generally, however, pricesare expected to rise partly because of lack of com-

petitive pressures on the increased number ofsole source suppliers and partly because the fixedcost of nuclear quality assurance must be spreadover dwindling sales.

Delays are also expected to be more of a prob-lem for similar reasons. With less nuclear workto do, suppliers are more likely to arrange pro-duction schedules to use qualified craftsmen andtheir ‘special machinery only when a number oforders are in hand, postponing work on someprojects for months. Or they could require pre-miums for deadlines that are more convenient

for the utilities. “He’d get the part for you, whenyou wanted it,” one utility executive said of a sup-plier, “but you’d have to pay for a whole shiftto go on overtime” (21). Suppliers report that util-ities are placing more “unpriced” orders, forwhich the supplier alone sets the costs, andchoosing other than the lowest bid to get theschedule and quality they need (21).

In addition to possible increased prices and de-lays, utilities are also experiencing some greaterconfusion in the bidding process for rework andnuclear services as more and more firms attempt

to diversify in the face of falling profits. “Anythingin an RFP [request for proposal], that’s at all re-lated to our business, we’ll bid on it,” one nuclear

consultant said. “We’ve got to try for anythingout there, just to survive” (2]).

Skill Shortages.–Utilities are also having trou-ble recruiting certain categories of employees andthis may get worse in the future. According toa personnel study by the Institute of Nuclear Pow-

er Operations (INPO), the overall vacancy ratewas 12.5 percent of all nuclear-related positions.However, for nuclear and reactor engineers, forradiation protection engineers, and health physi-cists (technical specialists in health effects of radia-tion), the vacancy rate was more than 20 percent(see app. table 7A). The average turnover rate forengineers is almost 7 percent a year, and, for mostcategories of engineers, quitting their jobs in util-ities means leaving the industry altogether (16).For the nuclear utilities as a group, an estimated6,000 additional engineers will be needed be-tween now and 1991. Almost 5,000 of these will

be needed to replace those that leave the indus-try (see fig. 35). About 3,000 technical level healthphysicists will be needed, about 2,000 of thesefor replacement (18).

Despite the availability of ample jobs for nucle-ar specialists, degrees and enrollment in nuclear-related fields are stable or declining (see fig. 36for nuclear engineering degrees). There is someevidence that students are being discouragedfrom enrolling in programs leading to employ-ment in nuclear power by a perception that theindustry is declining and by parental concern and

some peer pressure against nuclear power ca-reers. A recent DOE study of personnel for thenuclear industry contrasted steadily increasingenrollment in medical radiation physics programswith declining enrollments in technically similarhealth physics and radiobiology programs aimedat work in the field of nuclear electricity genera-tion (9).

INPO which has developed demanding train-ing requirements for utility personnel has alsotaken some modest steps to help with recruitingby setting up a fund for graduate nuclear train-

ing (see ch. 5). Individual utilities have also takensteps to fund nuclear programs at local communi-ty colleges. More, however, may be needed if




Figure 35.–Estimates of Additional Manpower Requirements for the Nuclear Power Industry, 1982-91

I Professionals I Technicians I Operators I Skilled craft I

SOURCE: Ruth C. Johnson, Manpower Requirements in the Nuclear Power Industry, 1982-91, September 1982, ORAU-205.

current turnover and recruiting trends continue,or worsen.

The Impact on Future New

Construction of Nuclear PlantsThere are several ironies in the current situa-

tion of much of the nuclear industry. Many com-panies are sustaining themselves on backfits andrework, which over time increases the cost of nu-clear power to utilities and consumers alike andmakes it less likely that utilities will place orderssoon for more nuclear powerplants. Some com-panies are maintaining their nuclear businessbecause the rest of their business has not yet beenaffected by the improvement in the economy. Asbusiness investment picks up, the rate of com-

panies leaving the nuclear business may accel-erate. The recovery is likely to create work and jobs in most other industries before utilities seetheir reserve margins shrinking and begin order-ing again (see ch. 3). As one supplier said, “If theeconomy revives, I’m not sure I can wait around

for the nuclear contracts to roll in. I maybe do-ing something else in the meantime” (21).

Some effects of a hiatus in new plant orders areinevitable even if the optimistic projection of nu-clear orders after 3 to 5 years occurs. There are

likely to be fewer reactor manufacturers (two orthree rather than four) and some initial delays forvendors in securing all the necessary suppliersand encouraging their renewal of N-stamps. Evenunder this optimistic projection, foreign sourceswould probably be used for some specific areasof supply.

Component suppliers estimate a delay of 1 or2 years in obtaining N-stamp qualifications andadditional delays once they are operating be-cause of unfamiliarity with support services.“Right now,” one said, “my people know just

who to call for an interpretation of the regula-tions. They know which seismic stress labs arethe best. If we had to start over [several yearshence], a question that takes an afternoon on thephone to answer today would take three to fourmonths to answer” (21).




800

700

600

500

400

300

200

100

(

Figure 36.—Nuclear Engineer Degrees: Foreign Nationals and U.S. Citizens

Key

B.S.

1973 1974 1975 1976 1977 1978 1979 1980

YearSOURCE: A Study of the Adequacy of Personnel for the U.S. Nuclear Industry (Washington, D. C.: Department of Energy, November 1981).

Despite these difficulties, it is probable that vide designs for initial plants that were developedeven after a hiatus, a realistic 8-year projectschedule (under conditions assuming no hiatus)would be delayed only about a year, and perhapsless, if utilities were to freely allow overseas pur-chasing. If utilities insisted on U.S. sources for allor most components, the delay could be longer(37).

With a hiatus of 10 years or longer, there willbe a much bigger problem in new plant construc-tion. Unless there have been at least some over-seas orders, the reactor vendors are likely to pro-

overseas in joint ventures with foreign countries,perhaps even as licensees of foreign companies(37). With a longer passage of time, there will bemore critical areas with no qualified U.S. supplierand more dependence on overseas componentsuppliers. Since licensing and quality-control re-quirements for foreign nuclear programs are quitedifferent it could prove time-consuming and dif-ficult to obtain nuclear qualification and licens-ing for the design and components of the initialplant (37). Under these circumstances it is unlike-ly that there would be more than two U.S. ven-




dors. It is also conceivable that foreign companiesmight bid directly for the design and supply ofthe initial units (37).

After a period of 10 years or more without nu-clear plant construction some skills would still beavailable from other industries. Control and in-strumentation designers and workers would prob-

ably be available from the electronics and aero-space industries. Construction contractors expectthat semiskilled construction and maintenanceworkers would also be available. Of the construc-tion skills, a shortage of welders with nuclear cer-tification might pose the greatest staffing prob-lem. But one AE executive said that the biggestdifficulty “would be administrative;” learningagain to control “the thousands of decisions andtasks” needed to construct and test a nuclearplant (21).

Still another possibility, a very low volume of

orders beginning in the late 1980’s, is the situa-tion most likely to encourage evolution in the nu-clear industry structure to permit the necessaryeconomies of scale in design and constructionmanagement experience (37). In this situation,it is likely that utilities or others will form regionalor national nuclear generating companies to ob-tain the economics of scale from multi-unit sitesand standardized construction. This is also thesituation that is likely to encourage “turnkey”construction, a practice used for the earliestplants constructed in the late I %0’s and still usedfor some exports of nuclear plants (e.g., a plant

being constructed by Westinghouse in the Philip-pines). In turnkey construction, a company orconsortium, often headed by an NSSS vendor,offers to construct and warranty an entire nuclearisland, * or even a complete nuclear plant, for afixed price, ready for the operating utility to “turnthe key” and operate the plant. Such fixed-priceagreements may be the only way for vendors toconvince utilities that their costs for nuclear

*Nuclear island refers to all the equipment that directly or indi-rectly affects the safety of nuclear operations. In addition to thereactor vessel itself and the primary cooling system it usually in-cludes the secondary cooling system and the steam generators (in

a pressurized water reactor).

power are predictable. It is quite possible thatforeign vendors might offer turnkey plants in theUnited States. It is perhaps more likely that U.S.vendors may attempt to form consortia with for-eign designers and component suppliers to of-fer turnkey plants.

Conclusion.–As of 1983 the nuclear industry

is still intact although somewhat reduced from3 to 4 years ago. The industry probably wouldsurvive a short hiatus of 3 to 5 years in new orderswith only some increase in costs and delays inobtaining some components from U.S. sourcesand perhaps little or no increase in costs anddelays if foreign component sources are used.

Predicting the consequences of a hiatus of 10years or more is more difficult but it is unlikelyto mean the end of the nuclear option in theUnited States. If vigorous, economical, and safenuclear programs survive in several foreign coun-

tries, they are likely to provide designs and somecomponents for the initial plants of a new roundof nuclear construction if one occurs. (The sur-vival of foreign nuclear programs is the subjectof the next section of this chapter.) Many U.S.businesses would still supply nuclear componentsbecause they supply very similar nonnuclearcomponents. Many others probably would getrecertified to supply nuclear components. U.S.vendors of NSSS will still have large nuclear serv-ice and fuel-loading businesses and probablysome foreign nuclear work as well. They are likelyto be active in any consortia or joint ventures in-

volving foreign sales of nuclear powerplants inthe United States.

Under some circumstances, AE firms could endup with less nuclear business after a long hiatus,depending on what restructuring might occur inthe industry. A shift to turnkey construction ofentire plants or the formation of a few generatingcompanies with their own design and construc-tion management staffs would sharply reduce therole of the architect-engineer. The number of util-ities directly involved in nuclear constructionwould also be drastically reduced under such cir-

cumstances.




THE PROSPECTS FOR NUCLEAR POWER OUTSIDETHE UNITED STATES

Many of the problems that have threatened thenuclear power industry in the United States havealso weakened nuclear power prospects abroad.

However, a few countries–with somewhat differ-ent institutional structures for producing electrici-ty and stronger motivation to avoid dependenceon energy imports-may be able to nurture theirnuclear industries to survive the 1980’s in strongercondition than the U.S. industry. This sectionsurveys the highlights of the foreign nuclear ex-perience–economic, technical, and political–and points out a few aspects of foreign experiencethat provide a perspective on U.S. experience.The section also assesses the likely competitivesituation of the U.S. industry vis a vis its com-petitors abroad.

The Economic Context forNuclear Power

Worldwide forecasts of the future role of nu-clear power have experienced the same boomand bust cycles as have U.S. forecasts. In 1975,OECD* countries forecast a total of 2,079 GWof nuclear power by the year 2000. As of 1982

*OECD means Organization for Economic Cooperation and De-velopment, a Paris-based organization of industrialized countries.

the OECD countries forecast for the same yearhad fallen by 75 percent, to only 455 GW of nu-clear power (table 24). The reasons for this dras-

tic reduction in expected nuclear capacity are fa-miliar to anyone acquainted with the U.S. nuclearindustry: slower-than-expected electricity de-mand growth, high interest rates that increasedthe cost of capital for nuclear powerplants andstronger-than-expected public opposition inmany countries.

Just as in the United States, the rate of growthof electricity demand slowed from the 1960’s tothe 1970’s in France, West Germany, and theUnited Kingdom (except for demand from Frenchhouseholds) (10) (see fig. 37).

Given the slower-than-expected growth ratesin electricity demand, many countries are nowlowering their forecast growth rates for 1990 and2000 and finding themselves with adequate gen-erating capacity, West Germany expects to neednew powerplants only if oil and gas capacity isto be replaced (24). A Government commissionin France estimated that completion of the pres-ent construction program should provide mostof the electricity forecast to be needed before2000. As of mid-1983, the Government had not

Table 24.–Forecasts of Installed Nuclear Capacity in OECD Countries (GW)

Installed public generatingForecasts for the year 2000 Nuclear capaci ty instal led capacity of all types

Regions and countries OECD, 1975 INFCE, 1980a OECD, 1982a 1980 1990b 1979C

Western Europe: 798 341 214 43.8 126.6 371France . . . . . . . . . . . . . . . . 170 96 86 16.1 56.0 47Germany. . . . . . . . . . . . . . . 134 63 34 8.6 23.5 66United Kingdom. . . . . . . . . 115 33 31 6.5 11.1 69

North America: 1,115 384 173 60.2 131.2 671Canada . . . . . . . . . . . . . . . . 115 59 22 5.2 15.0 72United States . . . . . . . . . . . 1,000 325 151 55.0 116.2 599

West Pacific: 166 130 68 15.8 28.0 153Japan . . . . . . . . . . . . . . . . . 157 130 68 15.8 28.0 126Australia/New Zealand . . . 9 0 0 0.0 0.0 27

OECD total 2,079 855 455 119.8 285.8 1,195

aFlgu~~S are the averages of high and low estimates.bcapaclty installed or due to be commissioned by Jan 1, 1990 Source SpRIJ turbine 9enerator data bank

csource IJnlted Nat Ions, 1981.

SOURCE Mans Lonnroth, “Nuclear Energy In Western Europe,” a paper for the East-West Center, Honolulu Conference on Nuclear Electric Power In the Asia Paclf!cRegion, January 1983.




Figure 37.–Average Annual increase in ElectricityDemand: France, West Germany, and United Kingdom

Industry House- Industry House- Industry House-holds holds holds

1980-70 1970-80

SOURCE: OECD.

yet accepted the report because of the drastic im-plications for the future of the nuclear powerindustry.

Other conditions, familiar to observers of theU.S. nuclear industry, were described by MansLonnroth, and William Walker in a 1979 paper,The Viability o f the Civil Nuc lea r Industry (26).

Although the rapid increases in oil prices after1973 made nuclear power appear relatively lessexpensive over the long run, it made it harder

to finance in the short run because the resultinghigh rates of inflation increased interest rates. Inthose countries where the government approveselectricity rates it became harder to get politicalsupport for rate increases to compensate for in-flation. Inflation and the several recessions of thedecade also put pressure on governments provid-ing financing to electric utilities to restrict the ex-tent of their support (26).

Public Acceptance

In most European countries and some other

countries, there has been considerable public op-position to nuclear power. This sometimes arisesfrom specific concerns about plant siting, designand management, and sometimes from much

broader philosophical concerns about the futuredirection of a society based on high technologywhich requires extensive central control (24,30).

In a few countries, public opposition to nuclearpower has become directly involved in politicaland administrative decisions that affect the

growth of nuclear power. The most dramatic ofthese is Sweden. in a referendum held in March1980, 57 percent of the public voted that 3 plantsunder construction should be completed but thetotal of 12 reactors then in existence should beoperated only until economical means are foundto replace them and should not be replaced withmore nuclear powerplants if there is any feasi-ble alternative. Parliament subsequently adoptedthis position as official Government policy. A sim-ilar referendum, which halted nuclear power-


In West Germany, construction of nuclear plantshas been stopped in the courts while in France themore centralized decision-making system has keptnuclear construction going without delays, despite

public opposition.



Ch. 7—Survival of the Nuclear Industry in the United States and Abroad . 191

plant construction in Austria, was passed follow-ing a period of widely publicized debates amongnuclear experts. In West Germany, citizens cansue in State courts to stop the construction ofpowerplants. Four plants were stopped in themid-1970’s and brought nuclear construction inGermany to a virtual halt. Subsequent extensive

Parliamentary Commissions have recommendedcaution but not a halt. In the United Kingdom,a public inquiry was conducted throughout 1983to consider the general adoption of a modifiedWestinghouse light water reactor (LWR) design.(This is the Sizewell B design, discussed in ch.4.) This technology would be in addition to theexisting series of advanced gas-cooled reactorsthat until now have formed the basis of theUnited Kingdom’s nuclear technology.

In France, there have been infrequent Parlia-mentary discussions of policy with respect to

nuclear power; otherwise decision making hasbeen treated as a technical and administrativerather than a political matter (31). Public opposi-tion has been expressed in anti-nuclear demon-strations, in demonstrations at particular sites, andin the formation of ecology parties which havechallenged candidates in local and regional elec-tions. None has had any substantial impact onthe French nuclear program. In part, this appearsto be because public opinion surveys have shownincreasing support for nuclear power (36).

Foreign Technical Experience:

Plant Construction and ReliabilityMany foreign countries have experienced de-

lays in building nuclear powerplants similar tothose experienced in the United States, but somehave built all their plants as fast or faster than anyU.S. plant (see table 25). In France, the slowestplants have taken 7 or 8 years from reactor orderto commercial operation, while the fastest plantstake 5 to 6 years (14). In Japan, slower plants havetaken 9 or 10 years while faster plants have alsobeen built in 5 or 6 years. In Japan most of thedelay occurs at the site-approval stage, prior to

the start of construction. For the other countrieswith at least several nuclear plants the fastest con-struction times are comparable (6 to 8 years) withthe fastest construction schedules in the UnitedStates.

Typical nuclear plants in several foreign coun-tries can also be constructed more cheaply thantypical U.S. plants. Based on information ob-tained in an NRC survey of foreign licensing prac-tices, U.S. costs are comparable with those ofSweden and West Germany but about 30 per-cent higher than in Japan and about 80 percent

higher than in France (38). In France, a nuclearplant can be constructed for about half the man-hours/kW required for a nuclear plant in theUnited States. The other three countries useabout 30 percent fewer man-hours/kW than inthe United States (fig. 38). Most of the savings oc-curs in two categories: the nuclear incrementover man-hours needed for constructing a non-nuclear powerplant, and the engineering man-hours used during construction, which is almost10 times higher in the United States than in anyother country (38) (see ch. 3).

Performance of U.S. reactors falls at, or slight-ly below, average in cumulative load factors forworld reactors* (see table 26). Several countries,such as Finland, Switzerland, Belgium, and theNetherlands, with only one to four reactors, havevery high average cumulative load factors. Can-ada’s nine heavy water CANDU reactors have thehighest load factors of all, averaging over 80 per-cent (see ch. 4). In other countries, however, withlarge numbers of reactors, average reactor per-formance does not differ substantially from theUnited States. At the same time, a smaller shareof U.S. reactors can be found among the top-

ranking reactors. Among the top 25 percent are96 percent of Canada’s reactors, 32 percent ofSweden’s reactors, 44 percent of West Germany’sreactors, 16 percent of Japan’s reactors, but only10 percent of U.S. reactors. Many U.S. reactorscan be found at the bottom of the list; 27 per-cent of U.S. reactors rank in the lowest 25 per-cent of world reactors.

Licensing and Quality Control

West Germany has as complex a licensingprocess as the United States. Licensing of nuclear

plants is governed by seven State (Lander) licens-

*World reactors excluding reactors in Eastern Europe, theU. S. S. R., and several third-world countries for which cumulativeload factor data is not available (see appendix table 7B for listingof nuclear capacity in all countries).




Table 25.—Sample Construction Times for Nuclear Plants in Various Countries

Faster Slower

Date of com Years Sine; Date of com Years s inceoperation reactor order operation reactor order

France:St. Laurent B1, B2 . . . . . .Gravelines C5 . . . . . . . . .Dampierre 2,3,4 . . . . . . .

Cattenom 2 . . . . . . . . . . .Germany:KKU, Unterweser. . . . . . .KKI-1, Ohu . . . . . . . . . . . .KKG, Grafenrheinfeld . . .KKK, Krummel . . . . . . . . .KBR, Brokdorf . . . . . . . . .

Italy:Caorso . . . . . . . . . . . . . . .Montalto 1 . . . . . . . . . . . .

Japan:Ikata 2. . . . . . . . . . . . . . . .Fukushima 11-2 . . . . . . . .Fukushima 11-3 . . . . . . . .Genkai 2 . . . . . . . . . . . . . .Ohi 1,2 . . . . . . . . . . . . . . .Fukushima 11-1 . . . . . . . .

Sweden: Barseback 1 . . . . . . . . . . .Barseback 2. . . . . . . . . . .Ringhals 3,4 . . . . . . . . . .Forsmark 1 . . . . . . . . . . . .

Spain:Almaraz 1 . . . . . . . . . . . . .Valdecaballeros 2 . . . . . .Lemoniz 1 . . . . . . . . . . . . .

Switzerland:Goesgen. . . . . . . . . . . . . .Leibstadt . . . . . . . . . . . . .

Taiwan:Kuosheng 1 . . . . . . . . . . .Chin-Shan 1 . . . . . . . . . . .Maansham 2 . . . . . . . . . .

Canada:

Pickering 1 . . . . . . . . . . . .Bruce 5 . . . . . . . . . . . . . . .Gentilly 2 . . . . . . . . . . . . .Darlington 1 . . . . . . . . . . .

United Kingdom:Torness 1, 2 . . . . . . . . . .Heysham 3,4 . . . . . . . . . .Dungeness B-1, B-2. . . . .Heysham 1,2 . . . . . . . . . .

United Statesa:St. Lucie 2 . . . . . . . . . . . .Hatch 2. . . . . . . . . . . . . . .Diablo Canyon 1 . . . . . . .Midland 1 . . . . . . . . . . . .

19811984 —

—

197819771981 — —

1978 —

1982198419851981 — —

19751977 — —

1981 — —

1979 —

19811979 —

19711983 — —

19861986 — —

19831979 — —

—

766

— —

8 —

5655

— —

65

— —

9 — —

6 —

89

—

67 — —

87

— —

67

— —

— —

1981

1986

19831987

1986

19791982

19821980

19881983

1984

1985

—

19831989

—

19821983

—1984

1985

— —

— — —1112

—

12

— — — —

910

— —109

—

1311

—

14

— —

11

— —

912

— —

1713

— —

1613

%arting time s construction permit rather then reactor order.

SOURCES: December 19S1 Atomic Induatrlal Forum List of Nuclear PowerPlants Outside the United States; ch. 5, tablefor U.S. nuciear plants.




Photo credit: Electricite de France

Each of the four identical 925-MW units at Tricastin in France averaged 6.5 years construction time. Work force a ndengineering experience with the standard design and with the site accounted for the short construction time




Figure 38.-Engineering and Construction Man-Hoursper Magawatt of Capacity (typical nuclear plants)

19,160

United West Sweden Japan FranceStates Germany

Engineering:

Design &analysis

Construction:

Normal

SOURCE: J. D. Stevenaon end F. A. Thomee, Selected Review of Forwign LlcemwIng Practlcea for Nuclear Power P/8nt8, NUREG/CR-2e64. Theae esti-mates were made from estlmatea of staffing pattema, project dura-tion, end other Information obtained from ntervlewa with about 50 pao.ple in the Unltad States end foreign countrlea. The Intewlews are listedIn the front of the raport.

ing agencies each of which in turn depends onone of seven independent inspection agencies(TUV) for technical licensing requirements. In the1970’s, West German licensing and quality-con-trol requirements imposed on nuclear plants in-creased in much the same fashion as they did inthe United States (19). Engineering work tripled,quality-assurance documentation quintupled,and there was a 50-percent increase in the timerequired for component manufacture.

In France, Japan, and the United Kingdom, theconsideration of siting and environmental issues

is clearly separated from the consideration of safe-ty issues. In all three countries the only oppor-tunity for public intervention occurs at the earliestsite-approval stage. This process can take 2 to 3years in Japan (17). In France, the site-approvalprocess can take place as much as 8 years beforeapplication for a construction license. In fact, sites

have been approved for all nuclear constructionuntil 1990 (36). In England, there is an option tohold a public inquiry at this early stage. The pub-lic inquiry for the Sizewell B covers a broad rangeof issues concerning development of pressurizedwater reactors (PWRs) in England (32).

Once site approval has been obtained, the li-censing process in all three countries involves atechnical exchange of information between li-censing entity and licensee. In Japan and theUnited Kingdom, there is only one license (a con-struction license) but there is a series of technical

requirements—for tests, safety analyses, and oper-ating procedures—that must be met before theplant is allowed to operate. In practice, thisamounts to a multistage licensing procedure. InFrance, there is an additional operating license,but the process is similar. The plant must passa series of tests and have proposed operating pro-cedures approved before an operating licensecan be issued (36).

Overall Nuclear Development

There are significant differences among coun-

tries in the probable future course of nucleardevelopment. Table 27 classifies countries withactual or possible nuclear power programs intosix categories based on the prospects for furthernuclear construction.

Five countries–France, Japan, Canada, Tai-wan, and Korea—have Government-supportedprograms of constructing and operating nuclearplants. Three—France, Japan, and Canada–haveemphasized standardization to minimize con-struction cost and increase reliability. All threehave ambitions for major export programs in the

1990’s when demand picks up again, althoughCanada has had several setbacks in its efforts tomake the CANDU heavy water reactor (HWR)a viable export, when sales failed to go forwardin Mexico, Korea, and Rumania. In addition, Can-




Table 26.—Operating Performance by Country (average reactor cumulative load factor, to end of 1933)

Reactor types

PWRs BWRs Magnoxs Heavy water HTRs &AGRs

Load Number of Load Number of Load Number of Load Number of Load Number ofCountry factor, 0/0 units factor, % units factor, 0/0 units factor, % units factor, 0/0 units

United States . . . . . . . . . . . 54.8 48 56.1 23 — — 17.9Argentina . . . . . . . . . . . . . . . —

— — — — — — 75.3 1

Belgium . . . . . . . . . . . . . . . . —

75.6 3 — — —

Brazil. . . . . . . . . . . . . . . . . . . 2.3

— — —

1 — — —Canada . . . . . . . . . . . . . . . . . — — — — — —

— —80.1 9

Finland . . . . . . . . . . . . . . . . .

—

75.0 2 62.7 2France . . . . . . . . . . . . . . . . . 56.0 23 — —

Germany, West . . . . . . . . . . 71.2 7 44.1 4Great Britain . . . . . . . . . . . . — — — — 58.5 18 - - –40.6India . . . . . . . . . . . . . . . . . . . — — 49.1 2 — 33.4 2Italy. . . . . . . . . . . . . . . . . . . . 44.1 1 30.3 1 59.3 1 —

Japan . . . . . . . . . . . . . . . . . . 56.2

— —

11 61.0 12 62.6 1 49.8 1

Netherlands . . . . . . . . . . . . . —

77.9 1 — — —

South Korea . . . . . . . . . . . . — — —

55.2 1 — —

Spain . . . . . . . . . . . . . . . . . . — — —

36.8 2 62.0 1 72.0 1 –Sweden. . . . . . . . . . . . . . . . .

— —38.3 2 65.1 7 —

Switzerland . . . . . . . . . . . . . — — —

75.1 3 80.5 1 —Taiwan . . . . . . . . . . . . . . . . . —

— — —51.9 4 —

Yugoslavia . . . . . . . . . . . . . . 53.1 1 — — —

— — — — — —

World . . . . . . . . . . . . . . . . . . 56.7 106 56.6 57 57.8 26 76.0 13 38.0Notes: 1. All plants operating less than a year as of July 1983 were excluded from this calculation.

2. The USSR and several countries in Eastern Europe with substantial numbers of nuclear plants are not listed, (See appendix table 7B)3. Graphite-water and fast breeder reactors are not included,4. Plant load factors were weighted by plant rated capacity to get country and reactor-type averages.

SOURCE: Nuclear Engineering International, “Nuclear Station Achievement 1983,” October 1983.

1 — —

— — — — —

4

— — — — — — — — — —

5

Table 27.—Categories of Foreign Nuclear Programs in Western Europe and theAsia-Pacific Region

Category Countries in category

I.

Il.

Ill.

IV.

v.VI.

More nuclear plants planned and under-construction backed by Government policyMore nuclear plants planned but may bestopped by public oppositionMore nuclear plants planned but delayeddue to economic difficultiesNuclear plants in existence with

de facto and de jure halt on furtherconstructionNuclear plants begun but indefinitely haltedGovernment policy prohibits nuclear plants

France, Japan, Taiwan,Canada, KoreaUnited Kingdom, WestGermany, ItalySpain, Yugoslavia, GreecePortugal, TurkeySweden, Switzerland

Philippines, IndonesiaAustralia, Austria, NorwayIreland, Denmark

SOURCES: Off Ice of Technology Assessment categorization based on papers presented at conference on Nuclear-ElectrlcPower In the Asia Pacific Region Jan 24-28, 1982; and Mans L6nnroth and William Walker Nuc/ear PowerSfrugg/es. /ndustr/a/ Corrrpet/t/orr and Pro//ferat/on Corrtro/, George Allen and Unwln, London, 1983.

ada has entered into negotiations with U.S. util-ities to build plants whose output is primarily in-tended for the U.S. market (20,24). Several char-acteristics of the nuclear industry in Canada,

France, and Japan make it far easier to maintainmomentum in the nuclear industry than it is inthe United States. In Canada and France, thenuclear-using utilities are Government-owned(see table 28). With only one nationalized utility

in France and three nuclear-owning utilities inCanada, the institutional coordination for an ef-fective standardization program is fairly easy. InJapan, the nine utilities are investor-owned and

depend on private financing. However, planningfor nuclear power development takes place with-in the overall framework of private-public coop-eration established by the Ministry of interna-tional Trade and Industry (MITI). Of these coun-




Table 28.—Structure of Electric Utility in Main Supplier Countries and the Distribution of AuthorityOver Key Decisions Influencing Financial Health of Utilities

Ownershipof utilities

United States: Large number of

utilities, mainlyprivately owned

Choice of External financing Rategenerating mix of investments regulation Comments

Fragmented authorityover utility financialhealth. Role of FederalGovernment very weak

In principle utility,but, State governmentstend de facto to influencedecisions

Capital market (bonds,stocks). Bonds rated byindependent ratingagencies; State finance

for public utilities

Public utilitycommissions in eachState

France:State owned,

EdFGovernment, at

recommendation fromEdF

National budget,international capitalmarket (e.g. US) forbonds

National Governmentapproves ratechange

National Governmentcontrols financial health

West Germany:Several utilities,

the larger ones hav-ing mixed State(land) and privateownership

Utility, but regional govern-ment makes finallicensing decisions

Capital market,regional government

Federal Governmenthas to approve ratechanges

Local, regional, and Federalgovernments allinfluence, and haveinterest in, financialhealth of utilities due toownership and ratingresponsibilities

Canada:Mainly provincially

owned (Ontario,Quebec, etc.)

Utility recommendation,provincial governmentfinal decision

Budget of provincialgovernment, capitalmarket (bonds)

Provincial governmentapproves ratechanges

Provincial governmentcontrols financial health

United Kingdom: State owned,

CEGB and SSEBNational Government, after

recommendation fromgenerating boards

Budget of NationalGovernment

National Governmentapproves ratechanges

National Governmentcontrols financial health

Japan:Private investor

owned (9)Safety assessments carried

out by central govern-ment, environmentalassessment by localgovernment. Finalauthority rests withPrime Minister

Mixed (bonds,equity ... )

Ministry ofInternational Tradeand Industry (MITI)

Financial health generallygood. Utilities withnuclear investments havedeveloped substantialin-house technicalcapabilities

Sweden: State owned (50)%)

Privatelyowned ( 3 5 % )

Localgovernment (15°10)

National Governmentfinal Iicensor after pro.posals from utilities

National budget (forState-owned utility),bonds for utilities not

Almost none.Electricity producersallowed to competefor large-scalecustomers and

distributors

Financial health generallygood, due to large shareof inflation resistanthydro. Competition be-tween utilities said to

hold rates down

owned by the state

SOURCE. Mans Lonnroth and William Walker, The V~ab~l/fy of the CIvI/ Nuclear Industry, a working paper for the International Consultative Group on Nuclear EnergyPubl}shed by the Rockefeller Foundation and the Royal Institute of International Affairs In 1979.

tries, France and Japan have substantial domesticmarkets for nuclear powerplants and are thus instronger position to sustain nuclear industries.

in West Germany, a group, called a “convoy,”of six similar nuclear powerplants was startedthrough licensing review in the spring of 1982 (7).There are indications that political and public op-position may have peaked although there havebeen no changes in legal structure (24). Construc-tion stoppage is still a possibility, however, be-cause citizens retain the legal right to sue to stopthe plants and the courts are independent.

The United Kingdom, West Germany, and Italyhave nuclear powerplants underway, but thesecould still be stopped by public opposition. Fur-ther construction of LWRs in the United Kingdomwill depend on the outcome of the Sizewell B

Inquiry. The case for new construction is greatlyweakened by the very low electricity demand In Italy, two nuclear plants are under construc-growth in the United Kingdom over the decade tion in addition to four in operation. Local oppo-of the 1970’s (32). sition to the two under construction caused ex-




tensive delays. The State electricity corporation,EN EL, will begin a vigorous campaign of localpublic education at each site of four plants pro-posed to be built by 1990. Success of such ef-forts in avoiding site delays is still unknown;similar site public relations efforts in France weretargets of protestor bombings.

Of the countries with nuclear plants in exist-ence and nearing completion, but with a holdon further construction, Sweden has the most ex-plicit moratorium. Switzerland also appears tohave halted further construction beyond plantsscheduled to begin commercial operation in1985. Several countries have avoided nuclearpower in developing a national energy policy.Despite possessing some of the world’s richesturanium supplies, Australia is dedicated to bas-ing its electricity generation on coal. Austria andNorway have decided against building nuclear

plants but are able to use abundant hydropower.New Zealand has surplus electricity from hydro-power and coal; and Ireland will increase coal-fired electricity (12).

Implications for the U.S.Nuclear Industry

Implications for the U.S. industry can be drawnfrom the experience of other countries in devel-oping nuclear power. Probably the most power-ful lessons will come from countries more likeourselves.

The West German “Convoy” Experiment.–The German licensing system for nuclear powerappears every bit as cumbersome as the U.S. sys-tem; in fact it involves even more regulatory man-years per regulated megawatt (38). With sevenState licensing authorities, assisted by seven dif-ferent independent engineering review organiza-tions (TUVs), the West German system adds State-to-State inconsistency to the several stages ofhearings and the independent court reviews ofthe U.S. system.

In an effort to halt the cycle of delays, require-

ments for rework, and increasing engineeringmanpower and paperwork, Kraftwork Union(KWU), the chief German reactor vendor, has ne-gotiated a plan with State licensing authorities and

technical agencies (TUVs) for a series of power-plants to be ordered and licensed in groups offive or six or “convoys” over the next 10 to 12

yea rs (19). The b asic proc ess is mo de led on the

successful French program. Each series would

have a standard design. Improvements on the

design would be saved for a subsequent series.

The various parties to the construction and li-censing of powerplants have agreed to severalchanges in procedure designed to reduce the costand delays in plant construction. Documentationrequirements have been simplified. Technicalreviews of different aspects of the convoy plantshave been assigned each to a separate technicalreview agency. KWU will make maximum useof computer-assisted design and develop a con-voy management system that controls the criticalfeatures of the design of each plant.

The legal framework has not changed in WestGermany in order to facilitate the convoy con-cept. As KWU concluded in its report on the con-cept, “without the broad consensus of all the par-ties involved (namely the customers, licensingauthorities, authorized inspection agencies, andmanufacturers) the concept will remain nothingmore than a collection of odds and ends, withevery chance of real success denied it” (19).

Although the institutions are different, the prob-lems of getting a large number of different organi-zations to work together to streamline an increas-ingly cumbersome process is similar to what

would have to be accomplished in the UnitedStates. Success of the West German effort woulddemonstrate that such an effort is possible.

Backfits.—ln both West Germany and Francethere are policies that restrict backfits. In WestGermany, utilities are supposed to be compen-sated for the costs of implementing any backfitsafter the State licensing authority has given its ap-proval (38). In practice, this provision has notbeen carried out very often and has not pre-vented the escalation in required engineeringman-hours described above.

In France, backfits are restricted once eachstandardized design has been approved. Occa-sionally, certain backfits (e.g., several followingthe Three Mile Island accident) may be judged

25-450 0 - 84 - 14 : QL 3




important and then they are implemented on allplants of a certain design (36,6).

Standardized Training.–In West Germanythere is a single institution for certifying power-plant operators. This school, called the Kraftwerk-schule is owned by a joint organization of 116

utility members in six countries. Operators com-plete a 3-year course including supervised opera-tion of an actual powerplant. Such training is aminimum requirement for a deputy shift super-visor. The shift supervisor must be an engineer(33).

Siting of Nuclear Powerplants.–ln Japan and

Italy, land is constrained, and finding sites for nu-clear plants is difficult. In Japan, the most difficultpart of the licensing process is the series of nego-tiations with local governments. MITI has in itsbudget about $60 million for public works grants

to local governments that accept nuclear power-

plants nearby. Additional funds are used to re-duce electric bills of local residents (12,36,38).More funding is available if power is exportedfrom the local area.

A similar approach is used in France whereelectric bills in areas surrounding nuclear plantsare reduced by 15 percent, and funds to buildhousing, schools, and other public facilities arelent by the utility to nearby towns which repaythe loan in utility property tax abatements. InItaly, there is an 18-month site review and ap-proval process. Special public education centersare set up at each proposed site well in advanceto help educate the public about the benefits andrisks of nuclear power (12).

Financial Risk.—There is far less financial risk

to utilities investing in nuclear power in othercountries than in the United States. In West Ger-

many, utilities set their own electric rates subject




to general Federal approval for household ratesand subject to antitrust provisions for industrialrates. Financing is provided not only by the pri-vate sector market but also by regional govern-ments, which are pat-t-owners of the largest util-ities. In Sweden where utilities also use privatefinancing for investments, electric rates are gen-

erally not regulated, and utilities are in goodfinancial shape (see table 28) (26). Similarly, inJapan, privately financed utilities are strong finan-cially. Electric rates in Japan are the highest inthe world and reflect the full costs of producingpower, in order to encourage conservation (12).In France, on the other hand, the Governmentapproves electric rates; in 1982 they were not in-creased enough to prevent a deficit of about $1billion in the account of the electric utility (36).

Timing and Balance. —There are some indica-tions that public opposition outside the United

States has been intensified by very rapid develop-ment of nuclear power in Sweden and West Ger-many (26). At the same time, public acceptancefor nuclear power in many countries seems tobe more solid if nuclear power is included as partof an overall national energy plan that includesa strong emphasis on energy conservation, re-newables, and other sources of electricity. Suchplans have been formally announced in Japan,France, West Germany, and Italy, all countrieswith potentially important roles for nuclear power(1 2).

Shift to a New Reactor Technology.–As con-sideration is given in the United States (see ch.4) to the desirability of shifting to a whole newreactor technology—heavy water reactors (HWRs),gas-cooled reactors (GCRs), or “forgiving” LWRs,several lessons can be learned from foreign ex-perience. One is that it is quite possible to shiftto a new technology. France shifted from GCRsto PWRs for plants ordered in the early 1970’s.The United Kingdom is now considering the shiftto PWRs in a formal public inquiry. In both cases,the shift has been towards a “standard” technol-ogy now in use in the United States and else-

where, Clearly, this is a different situation froma shift to a relatively untested technology de-scribed as a possibility in chapter 4.

The public inquiry into the Sizewell B PWRnow underway in the United Kingdom has beenone of the most extensive public debates on nu-clear power ever held. Beginning in January 1983,it lasted most of 1983. Much of the argument fo-cused on the economics of the proposed project.In all but one of the five electricity demand sce-

narios proposed by the Central Electricity Gen-erating Board (CEGB), electricity demand in theUnited Kingdom is forecast to decline between1980 and 2000. The argument for building theSizewell B powerplant is that cheaper nuclearpower will substitute for increasingly expensiveoil and coal, but there is much official skepticismabout the economic plan from other governmentagencies. The public inquiry has addressed ques-tions of the likelihood of cost overruns and ofpublic pressure for expensive safety improve-ments to match those being required in theUnited States and West Germany (32).

The Sizewell B debate should provide a thor-ough exploration of many of the issues now fac-ing the U.S. nuclear industry. Furthermore, it willprovide one more example of a possible ap-proach to involving the public in decisionmak-ing on nuclear power. Conceivably such a fullinquiry could precede the launching of a “con-voy” of new advanced design LWRs when orderspick up again.

The U.S. Industry in anInternational Context

Although the United States has by far the largestnumber of nuclear plants of any country, futureprospects for the U.S. nuclear industry are re-garded as dimmer than those of several othercountries, at least by some observers. Mans Lonn-roth, a Swedish author of a book on the world-wide nuclear industry (25), describes the com-ing decade as tough for all nuclear countries sincethe industry will have to demonstrate that it is a“safe, reliable and economic energy source. ”Success will depend in part on the coherence ofeach country’s response to this challenge (24).

The French have been very successful at con-structing nuclear plants quickly and cheaply.




They have a large design, manufacturing, andconstruction capacity much of which has noother alternative use. Among French industries,the nuclear industry has been successful, andthere will be considerable pressure to keep itoperating even in the face of slower growth inelectricity demand than anticipated. The French

nuclear industry, however, has had less successin exporting sales than has West Germany or theUnited States. This is the big challenge for thenext decade (24).

By contrast, Lonnroth claims, in the UnitedStates a “collapse of the nuclear industry wouldnot seriously affect either the public authorities,the industry at large, or the American society”(24). The U.S. social and political system is morefragmented than either the French or West Ger-man system and for this reason may have difficul-ty developing a coherent long-range strategy in

the absence of full consensus. West German gov-ernmental processes are as complex and openas those in the United States, but the Germannuclear industry acts far more as a single industry.It “views itself as one industry, with one collectivewill and one identity” (24). For this reason Lonn-roth suggests that the technical and economiccoordination necessary for a long-range nuclearstrategy in West Germany is possible within a se-ries of “interlocking ownerships among the mainindustrial actors” under the long-range guidanceof key West German banks (24).

Japan has had its difficulties with nuclearpower, including prolonged siting processes andconstruction delays. However, Japan has verystrong motivation to develop its nuclear industrybecause it has no indigenous fossil fuels. Japanhas also demonstrated the ability to coordinatelong-range industrial strategy in other areas andto develop an export-oriented strategy. Japan hasbecome the single most important exporter ofheavy electrical equipment and is expected toprovide about 25 percent of all exports in this sec-tor from 1975-87 (24). This gives Japan a verystrong industrial base on which to develop a

nuclear exports business. For the moment thenuclear industry still needs government support.However, such support will not be availableindefinitely.

Thus West Germany, France, Japan, and theUnited States–and perhaps Canada as well–willbe competing for a nuclear export market in the1980’s and 1990’s. They will be competing fora market that is far smaller than worldwide nucle-ar industrial capacity and far smaller than pre-vious estimates had projected (see table 29 and

appendix table 7B). Several importing countries—Korea, Taiwan, Argentina, and Spain–are work-ing to develop domestic supply capabilities andwill be curtailing imports. Many countries are im-porting far less equipment and engineering andconstruction management services than they didearlier in the decade.

Given the softness of the export market, it willbe difficult for suppliers of nuclear equipment andservices to be sustained through the next two dec-ades unless they have some domestic base. Al-though U.S. firms are involved in 31 plants still

under construction, all major components willhave been delivered by the end of 1984. Forthose overseas plants engineered by U.S. AE com-panies, the basic design is now complete (37).If U.S. suppliers begin to get new domestic ordersin about 5 years, they will provide a new basisfor maintaining design teams and manufacturingfacilities capable of sales abroad. If new reactororders are delayed for 10 years or more, U.S.firms may find themselves looking to the Japa-nese, West Germans, or French for joint venturesor licensing arrangements in which the foreigncompany is an equal, or even dominant, partner.

This would be the reverse of the situation in eachof these countries early in the history of the nu-clear industry.

Nuclear Proliferation Considerations

U.S. and worldwide efforts to restrict prolifera-tion of nuclear military technology are an impor-tant influence on the development of interna-tional trade in civilian nuclear power (25). Theperceived link between the civilian and militaryuses of nuclear power has stimulated much of

the opposition to nuclear power (see ch. 8). Thereasoning is straightforward: commercial nuclearpower requires the production of nuclear fuels,and some of these fuels and facilities that pro-




Table 29.-Estimated Reactor Export Market (1983-87 inclusive) in Units

Hardware Software Previous suppliersOrders Industrial market market 1960-80

L ow H ig h ca pa bi li ty L o w H i gh Low High (no. of units)

Europe: North . . . . . . . . . . . . . . . . . . . . .

Belgium . . . . . . . . . . . . . . . . . . 0Finland . . . . . . . . . . . . . . . . . . 0United Kingdom . . . . . . . . . . . 0

South . . . . . . . . . . . . . . . . . . . . . . (3Greece/Turkey/Portugal. . . . . 0Italy . . . . . . . . . . . . . . . . . . . . . 2Spain . . . . . . . . . . . . . . . . . . . . 1Yugoslavia . . . . . . . . . . . . . . . 0

Latin America: oArgentina . . . . . . . . . . . . . . . . 0Brazil . . . . . . . . . . . . . . . . . . . . OMexico . . . . . . . . . . . . . . . . . . . O

As/a and Pacific: 2

China . . . . . . . . . . . . . . . . . . . . OIndonesia . . . . . . . . . . . . . . . . 0Pakistan . . . . . . . . . . . . . . . . . 0South Korea . . . . . . . . . . . . . . 2Taiwan . . . . . . . . . . . . . . . . . . . O

Africa and Mid-east: oEgypt . . . . . . . . . . . . . . . . . . . . OIsrael . . . . . . . . . . . . . . . . . . . . OSouth Africa . . . . . . . . . . . . . . O

Total . . . . . . . . . . . . . . . . . . . 5Average annual rate . . . . . . . . 1.0

123)1

1

9)1

4

22

5

1

2

2

102

1

1

4

2

5

2

1

2

326.4

‘/2

2‘/2

3

1

1

2

2

2

2/3

2/3

32/3

‘/2

2

3

3

2/3

o00

0000

000

0000.50

0000.50.1

00.5 0.25

1

001

0.51

1.5

1.51

0.751

1

21

1.515.53.1

0 0.50 0.750 0.5

0 10.5 10.25 0.50 1.5

0 0.750 1.50 2

0 20 10 11 20.75 1.5

0 20 10 22.5 22.50.5 4.5

U.S. (6), France (2)Sweden (2), USSR (2)

—

U s . ( l ) –

U.S. (12), Germany (2)Us. (1)

Germany (2), Canada (1)Germany (2), U.S. (1)Us. (2)

Canada (1)U.S. (6), France (2), Canada (1)U.S. (6)

France (2)

SOURCE: Mans Lonnroth, “Nuclear Energy in Western Europe,” based on research for Nuclear Power Struggles, Allen and Unwin, London, 1983.

duce them can be used for nuclear weapons. Thefundamental premise of U.S. and worldwide ef-forts to avoid proliferation to additional nationshas been to keep civil uses of nuclear energy dis-tinctly separated from military applications, and

to try to erect barriers to prevent diversion or mis-use of civil nuclear materials and facilities.Technical and institutional aspects of nuclear pro-liferation were the subject of a previous OTAstudy and a recent Congressional Research Serv-ice paper which is reprinted in full as an appen-dix to this report (1 1,34).

Although it is technically possible to makecrude nuclear weapons from the plutonium inspent fuel from nuclear power reactors, it is morelikely from the higher grade plutonium manufac-tured in spent-fuel reprocessing, and in the opera-

tion of breeder reactors (see appendix table 7C).Economic prospects for commercial reprocess-ing have decreased worldwide as well as in theUnited States and no nation is currently pro-ducing and using plutonium commercially for

nuclear fuel. A few countries, most notablyFrance, are working with plutonium for use inbreeder reactors.

Current worldwide and U.S. concern aboutnuclear power and proliferation stems from theseconsiderations about the potential use of civiliannuclear fuels. One fear is that a rapidly industrial-izing state with a nuclear power base in a trou-bled part of the world might be tempted to useits civilian program as a base for developing nu-clear weapons. The second concern is that someunderdeveloped countries with nominal nuclearpower programs might obtain enough technologyand equipment on the world market to build fa-cilities to produce weapons-usable materials. Thegrave concern in the 1960’s that many nuclearpowerplants worldwide would give rise to the

wholesale spread of nuclear arsenals has givenway in the 1980’s to the concern that a few nu-clear powerplants and related facilities scatteredamong some developing countries could bringthem much closer to an ability to make nuclear




weapons. It is this concern that drives proposalsfor restrictions on international nuclear cooper-ation and trade.

Concern about the spread and use of nuclearweapons has led to several international initia-tives. The Non-Proliferation Treaty (NPT) pledges

its members that do not have nuclear weaponsto forego future acquisition of them, and requiresverification of the use of civilian facilities by in-ternational inspection. It further stipulates that allthe nuclear facilities in signatory states withoutnuclear weapons will be safeguarded, even thosethat are developed indigenously. This treaty rep-resents a significant departure from practices priorto 1970, and is an important element of the in-ternational proliferation regime. However, it hasnot been as effective as it might have been be-cause a number of nations have refused to par-ticipate in the treaty. As shown in table 30, the

nonsignatory states include India, Pakistan, Brazil,Argentina, and South Africa.

After NPT took effect, other events stimulatedfurther proliferation concerns. In 1974, India

tested a nuclear explosive that was derived from

civilian facilities. Shortly thereafter, France and

West Germany agreed to supply enrichment and/

or reprocessing plants to nations (Pakistan and

Brazil) whic h ha d refused to sign the NPT. By the

late 1970’s, the nuclear supplier countries had

become concerned enough to agree informallyto exercise additional restraint in nuclear coop-eration and trade, particularly in the area of thetransfer of sensitive technology. The United Statesimposed even more severe restrictions than theother suppliers with the passage of the NuclearNon-Proliferation Act of 1978.

U.S. policies and controls that guide nuclearcooperation and exports include the following

Table 30.-No-Nuclear Weapons Pledges in Effect in 1981

Treaty and data of entry into force

Treaty ProhibitingAntar ct ic Limi ted Te st Nuclear Weapons Nuclear Non-Treaty, Ban Treaty, in Latin America, Proli feration Treaty,

State 1961 1963 1 966a 1970

Argentina. . . . . . .Australia . . . . . . .Belgium . . . . . . . .Brazil . . . . . . . . . .Canada . . . . . . . .Cuba. . . . . . . . . . .Egypt . . . . . . . . . .

F.R.G, , . . . . . . . . .India . . . . . . . . . . .Iran . . . . . . . . . . . .Iraq . . . . . . . . . . . .Israel . . . . . . . . . .Italy . . . . . . . . . . .Japan . . . . . . . . . .Libya . . . . . . . . . .Netherlands. . . . .Pakistan. . . . . . . .South Africa . . . .South Korea . . . .Spain . . . . . . . . . .Sweden . . . . . . . .Taiwan . . . . . . . . .Yugoslavia. . . . .,

PPPPP

PP

P

P

sPPPP

PPPPPPPPPsPPPPPP

Sb

PPs c

P —

P — —

PP

P

PP

PPPP

PPPPP

P = Party.

S = Signatory.aNot ye t h force for all signatories.bRatified Subject to preconditions not Yet met.cAdditional Protocol II applylng to Dutch territories in Latin America.dThere is some difference of opinion as to whether one small unsafeguarded laboratory should be considered a faCiiity.

SOURCE: W. Donnelly and J. Pilat, “Nuclear Power and Nuclear Proliferation: A Review of Reciprocal Interactions,” Congres-sional Research Service for the Office of Technology Assessment, April 19S3.



Ch. 7–Survival of the Nuclear Industry in the United States and Abroad q 203

items: restrictive conditions for licensing exportsof nuclear materials, equipment, and reactors;restrictive conditions for providing technical as-sistance and transferring technology; restrictionson export of dual-use items that can be appliedto weapons programs as well as to legitimate nu-clear power programs; post-export controls, or

prior rights over what may be done with or toU.S. nuclear exports, such as reactors or fuel; andcutoff of nuclear cooperation and exports tostates which violate safeguards agreements withthe United States. These restrictions are em-bedded in U.S. law, particularly in the Non-Prolif-eration Act of 1978 mentioned above, the AtomicEnergy Act of 1954, and the Symington andGlenn amendments to the Foreign Assistance Actof 1961.

Nuclear cooperation and trade has been cir-cumscribed in many aspects by the restrictive

conditions and controls intended to prevent thedevelopment of nuclear weapons. Specifically,the supply of sensitive nuclear technology tocountries that have little visible economic needfor it is discouraged by the nuclear suppliers, andparticularly by the United States. The supply ofitems that can be used for both civilian and mili-tary purposes has been little affected in the past,but is likely to be more restricted in the futuresince export control lists are being made moredetailed and specific. Importing countries mostlikely to feel the effects of such additional re-strictions include Argentina, Brazil, India, and

pakistan.Pressures from both formal and informal non-

proliferation regulation can be viewed as stimu-lating some nuclear customer countries to seekindependence of the major suppliers, either bybuilding up their own nuclear industries indig-

enously or by finding suppliers who offer less de-manding conditions. This could give rise to theemergence of a second tier of suppliers from themore industrialized developing nations, such asArgentina and India, who might not comply withthe guidelines of the major suppliers. This couldsignificantly change the character of nonprolifera-

tion control.In part because of fears of loss of U.S. nonpro-

liferation influence and trade, the Reagan admin-istration has shifted emphasis somewhat from thepolicies of the Carter administration. Rather thanemphasize across-the-board denial of nuclearsupply, the Reagan administration has promotedthe concept that the United States is a reliablesupplier of nuclear equipment to trusted coun-tries who are not viewed as proliferation risks.

ConclusionFor economic, political, and technical reasons,

the 1980’s will be a difficult decade for the nu-clear power industry in all countries. Those mostlikely to survive are those with the political andindustrial cohesion to develop a long-range strat-egy for demonstrating that nuclear power is asafe, reliable, and economic energy source.France, Japan, and possibly West Germany andCanada have a combination of national motiva-tion and institutional coherence that makes itquite possible they will survive the decade witha more viable nuclear industry than will theUnited States. In the worst case, therefore, if theU.S. industry emerges weakened from a long pe-riod with no new orders, these countries may re-verse earlier roles and provide some of the exper-tise and hardware to U.S. companies during theearly years of a revival of the U.S. industry.




Appendix Table 7A.–Onsite and Offsite Nuclear-Related Job Vacancies at INPO Member Utilities Mar. 1, 1982

Vacancies

Occupations Positions a Number Percent of positions

Managers and supervisors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5,765Engineers:

Chemical . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179Civil . . . . . . . . . . . . . . . . . ., . . . . . . . . . . . . . . . . . . . . . . . . . 872Electrical . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1,518

Instrument and control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 506Mechanical . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2,844Nuclear and reactor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1,427Quality assurance/control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 791Radiation protection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140All other engineers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2,229

10,506Scientists:

Biologists . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 144Chemists . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 269Health physicists . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 404Other scientists . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 235

1,052Training personnel

SRO/Relicensed/certified instructors . . . . . . . . . . . . . . . . . . . . 405Other technical/scientific instructors . . . . . . . . . . . . . . . . . . . . . 576

Other instructors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 188Support staff... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135

1,304Operators:

Shift technical advisors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 416Shift supervisors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 735Senior licensed operators (SRO) . . . . . . . . . . . . . . . . . . . . . . . . . 385Licensed operators (RO). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1,094

2,214

Non-licensed operators assigned to shift . . . . . . . . . . . . . . . . . . . 2,286Other non-licensed operators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 351

2,637

Individuals ingraining for SRO licenses. . . . . . . . . . . . . . . . . . . 495Individuals ingraining for Relicenses.. . . . . . . . . . . . . . . . . . . 878Individuals ingraining for non-licensed positions . . . . . . . . . . 838

7,478Technical and maintenance personnel:

Chemistry technicians. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1,004Draftsmen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1,209Electricians . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1,609Instrument and control technicians . . . . . . . . . . . . . . . . . . . . . . 2,463Mechanics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3,554Quality assurance/control technicians . . . . . . . . . . . . . . . . . . . . 793Radiation protection technicians . . . . . . . . . . . . . . . . . . . . . . . . . 1,792Welders with Nuclear Certification . . . . . . . . . . . . . . . . . . . . . . . 415Other technical and maintenance personnel . . . . . . . . . . . . . . . 3,883

16,742All other professional workers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1,304Other technical personnel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1,061

Total . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ................45,212

432

3040

23991

32728714730

420

1,611

6 37 83 28

154

109100

5217

278

93119117230

7.5

16.84.6

15.7

18.011.620.118.621.418.8

15.3

4.213.820.514.6

14.6

26.9 17.4

27.7 12.6

21.3

22.416.230.421.0

466

242102

21.1

10.629.1

344

2266

246

13.1

4.47.5

29.4

1,237

16198

172320244

88266

48312

16.6

16.08.1

10.713.06.9

11.114.811.68.0

1,709125114

10.29.6

10,7

5.660 12.5

%h{sincludes pereons employed by INPOrnember utilities, Including holdlngcornpany positions allocated tothe utilities, vecant positions, and contractor posi-

tions used in normal operations.

Note: Fifty-five utilities providing offsite information; Onsite data comes from 82 plants representing 58 utilities, except onsite vacancy data, which was provided byonly81 piants representing57 utilities.




Appendix Table 7C.—Usability of Nuclear Materials for

Nuclear Weapons or Explosives

Usability

Material and form Direct Indirect Processing required

Plutonium:Metal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . YesOxide . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . PossibilityOxide with uranium in nuclear fuels ..... NoOxide or other forms in spent nuclear fuels. . . No

Thorium: Ore, metal, chemical forms . . . . . . . . . NoUranium-235:

Normal ore, metal, chemical forms. . . . . . . . . . NoSlightly enriched (3-6 percent) metal . . . . . . . . No

Oxide in nuclear fuels. . . . . . . . . . . . . . . . . . . . . NoOxide or other forms in spent nuclear fuels. . . No

Moderately enriched (20 percent) metal . . . . . . Unlikely

Oxide . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . NoOxide in nuclear fuels. . . . . . . . . . . . . . . . . . . . . NoOxide or other forms in spent nuclear fuels. . . No

Highly enriched (90 percent):Metal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . YesOxide . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . YesOxide in nuclear fuel. . . . . . . . . . . . . . . . . . . . NoOxide or other forms in spent nuclear fuels. No

Uranium-233:Metal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . YesOxide . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . YesOxide in nuclear fuels . . . . . . . . . . . . . . . . . . . . . No

Oxide or other forms in spent nuclear fuels. . . No

YesYesYesNo

NoYes

YesYes

Yes

YesYes

Yes

YesYes

Yes

Yes

Chemical separationDoReprocessing

Chemical processingto produce uraniumhexafluoride forenrichment toweapons grade

DoReprocessing,

chemical processingand enrichment

Chemical processingand enrichment

DoDo

Reprocessing,chemical processingand enrichment

Chemical separationReprocessing

Chemical processingand enrichment

Reprocessing,chemical processingand enrichment

SOURCE. W. Donnelly and J. Pilat, “Nuclear Power and Nuclear Proliferation: A Review of Reciprocal Interactions,” Congres-sional Research Service for the Office of Technology Assessment, April 1983.



208 . Nuc lea r Power in an Age of Llncetiainty

34<

35

36,

Standatiization: L i gh t W a te r Reac to rs , OTA- 37E-1 34, April 1981.

Offic e of Tec hnolog y Assessme nt,Nuc/ear Prolif-eration and Safeguards, OTA-E-48, June 1977.Summary, OTA-E-148, March 1982. 38

Ove rsea s Elec trica l Industry Survey Institute, Inc .,Electric Power Industry in Japan 1981, December1981. 39

Procter, Mary, “Notes on the French Nuclear Pro-gram,” OTA me mo to the files, ba sed qn conver-sations with the French nuclear at tachq a trip to

the Tricastin nuc lear p lant a nd other sources, July

19, 1983.

S. M. Stoner Corp., Nuclear Supply Infrastructure: Viability Study, prepa red for the Argonne NationalLaboratory, contract #31-109-38-6749, November

1982.

Stev enson, J. D., and Thomas, F. A., Se/ected Re- view of Foreign Licensing Practices for Nuclear Power P/ants, NUREG/CR-2664, April 1982.

Surrey, John, and Thom as, Steve , “ Worldw ide

Nuc lea r Plant Performanc e: Lessons for Tec hno l-ogy Policy,” Futures, February 1980.



Chapter 8

Public Attitudes TowardNuclear Power

Photo credit: OTA staff

Buttons and bumperstickers from the controversy about nuclear power



Contents

Page

Introduction: Public Opinion and Its Impact on Nuclear Power . . . . . . . . . . . . . .......21 1

Actors in the Nuclear Power Debate . . . . . . . . . . . . . . . . . . . . . ...................214The imp ac t of Public Op inion o n Nuc lea r Power . . . . . . . . . . . . . . . . . . . . . .........215

The Exp erts’ View . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..........................217The Controversy Over Safety Stud ies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ........218The lm p a c t of Risk Assessme nts on Pub l ic O p inio n . . . . .. . . . . . . . .. . . . . . . . .. . . . .220

Factors Influencing the Public’s View on Nuclear Safety . . . . . . . . . . . . . . . . . . . . . . . ...221

Perceptions of Risk and Benefit . . . . . . . . . . . . . . ...............................221Psyc holog ica l Fac tors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ......................222

Values and Knowledge . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..................226Which is More Im porta nt? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .............226

Values . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ....227

The Role of the Media. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .....................231The Exten t o f M ed ia Co ve rag e o f Nuc lea r Pow er . . . . . .. . . .. . . .. . . .. . . .. . . .. . . .231The Co ntent o f Media Covera ge . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .............232

What Would lt Tak e To Inc rease Pub lic Ac c ep tanc e o f Nucle ar Pow er . . . . . . . . . .. ...234Enhance Nuclear Advantages. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ...............234

Reduce Concerns Over Nuclear Accidents. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .236

Min imize L inkage Between Nuclear Power and Weapons . . . . . . . . . . . . . . . . . . . . . . .239

Case Study 1–Maine Yankee: Economics as the Key to Public Support . . . . . . . . ......240Case Study 2–Diablo Canyon: Environmentalism and Industry Credibility . . . . . . ......242

Case Study 3–Publ ic Perception of lmproved Management . . . . . . . . . . . . . . . . . . . . . . .244

Cha pter 8 Refe renc es . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..................245

Tables

Table No. Page

31. History of Statewide Referendum Votes Dealing With Nuclear Powerplants . ......212

32. Major National Groups Influencing Public Opinion For and Against Nuclear Power 21533. State Laws and Regulations Restricting Construction of Nuclear Powerplants .. ....216

Figures

Figure No. Page

39. Trend s in Pub lic Op inion on Nuc lea r Pow er . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..21140. Relationship Between Judged Frequency and the Actual Number of Deaths

per Year for 41 Causes of Death . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. ...223

41. Exp ert an d Lay Jud gm ents of Risk Plotted A ga inst Tec hnic al Estima tes of

Annual Fatalities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..................224



Chapter 8

Public Attitudes TowardNuclear Power

INTRODUCTION: PUBLIC OPINION ANDITS IMPACT ON NUCLEAR POWER

Public attitudes toward nuclear power havebecome increasingly negative over the past twodecades, with the most recent polls indicatingthat a slight majority of Americans opposes fur-ther construction of reactors. During the 1950’s,nuclear power was still in the early states of de-velopment, and pollsters did not even bother tosurvey the public on the issue. In the early 1960’s,a few scattered protests against local plants gainednational attention, but opinion polls indicatedthat less than a quarter of the public opposed nu-clear power (41 ). From Earth Day in 1970 through

the mid-1970’s, opposition levels averaged 25 to30 percent, indicating that substantial majorities

of the public favored further nuclear develop-ment. However, by 1976, anti-nuclear referen-da appeared on ballots in eight States.

Polls taken between 1976 and 1979 indicatedthat slightly over half of the American publicfavored continued construction of nuclear plantsin the United States in general, while about 28percent were opposed and 18 percent unsure.The accident at Three Mile Island (TMI) in April1979 had a sudden and dramatic impact on theseattitudes. As shown in figure 39, the percentageof people who had been in favor of or uncertainabout continued construction of reactors de-creased immediately following the accident whilethe number opposed increased (57). In subse-

quent months, there was some return to previous-ly held opinions, but opposition levels remainedmuch higher than they had been. National pollstaken since mid-1 982 indicate a continued slowerosion in support for nuclear power. About a

Figure 39.—Trends in Public Opinion on Nuclear Power,100

90

80

20

10

0 I 1 i I I t I I I I 01975 1976 1977 1978 1979 1980 1981 1982 1983

Year of survey

Question asked: “Do you favor or oppose the construction of more nuclear powerplants?”

SOURCE Cambridge Reports, Inc.

211




third of the public now supports construction ofnew plants in general, while over so percent areopposed (6, 10, 18). The accident at TMI appearsto have accelerated a trend of even greater op-position to construction of new plants near towhere those polled live. By the end of 1981, alarge majority of those polled opposed construc-

tion of new plants in or near their communities.When compared with other energy options, in-cluding offshore oil drilling and coal plants,nuclear is now the least favored alternative.

Despite the trend of declining support, the pubIic’s overall current attitude toward nuclearpower can best be described as ambivalent. Forexample, a 1983 poll indicates that about 40 per-cent of the public thinks currently operating reac-tors are “mainly safe” while slightly over halfthink they are dangerous and 5 percent are “notsure.” There is some evidence that the public

looks to nuclear power as one solution to the Na-tion’s long-term energy problems. In a recentsurvey, the majority of respondents believed thatmost U.S. energy needs would be supplied pri-marily by nuclear” and solar over the next twodecades, and over a third of those polled ex-pected nuclear power to provide most of the Na-

tion’s energy after the year 2000 (14). The majority of Americans favor neither a halt to all newconstruction nor a permanent shutdown of all op-erating reactors. Opinion polls on this questionhave been verified by State ballot initiatives. Asshown in table 31, most of the nuclear moratori-um initiatives, and all referenda that would haveshut down operating plants were defeated in1976, 1980, and 1982. However, more of theseinitiatives have been approved in recent years,and many restrictions on nuclear waste disposalhave been passed, reflecting public doubts aboutthe technology.

Table 31.—History of Statewide Referendum VotesDealing With Nuclear Powerplants

Year State Proposal Outcome Vote split.1976 Arizona 30-70 %

California 33-66%ColoradoMontanaOregonOhioWashington

1978 Montana

1980 Maine

Missouri

Oregon

1981 Washington

1982 Idaho

Maine

Massachusetts

Would halt new constructionand reduce operations untilsafety systems were foundeffective, liability ceilingslifted, and waste disposal wasdemonstrated

DefeatedDefeatedDefeatedDefeatedDefeatedDefeatedDefeated

Same as ’76 referenda

Would shut down Maine Yankee

Would prevent Callaway plantsfrom operating until safetysystems were found effective,liability ceilings were lifted,and waste disposal was available

Prohibits new construction untilwaste disposal is available andvoters approve in a statewidereferendum

Prohibits issuance of new bondsneeded to complete WPPSSUnit 3

Prohibits legislation limitingnuclear power unless approvedby voters in a referendum

Would phase out Maine Yankeeover 5 years

Prohibits new construction andwaste disposal unless certainconditions, including voterapproval in a referendum. are met

Approved

Defeated

Defeated

Approved

Approved

Approved

Defeated

Approved

29-71 %

42-58%42-58%32-68%33-67%

60-40%

41 -59%

39-61%

52-48%

56-44%

60-40%

44-56%

66-33%

Total restrictive referenda placed on ballots: 14Total approved: 4

SOURCES: Atomic Industrial Forum, State Codes.



Ch. 8—Public Attitudes Toward Nuclear Power q 213

The public’s ambivalent attitude toward nucle-ar power is due to a variety of factors includingthe ongoing debate among experts over reactorsafety, individual perceptions of the likelihood ofa catastrophic reactor accident, changing per-sonal values, and media coverage of the tech-nology. Underlying all of these factors is increas-ing doubt about the technical capabilities and thecredibility of both the nuclear industry and itsgovernmental regulators. As discussed in chapter5, weak utility management has led to poor op-erating performance at some reactors as well asskyrocketing costs and quality-assurance prob-lems at other plants under construction. Theseproblems have led to accidents at operating re-actors, causing great public concern.

As early as 1966, when large majorities of thepublic supported nuclear power, a design errorcaused blockage of coolant, leading to meltingof a small part of the core at Detroit Edison’s Fer-mi breeder reactor (3). Although no radioactivitywas released, and the event received relativelylittle publicity at that time, nuclear critics andsome members of the public became concerned.They pointed to a University of Michigan studyconducted prior to construction of the plant,which indicated that if the plant had been largerand had been operating at full design power forat least a year, a complete breach of containmentcombined with the worst possible weather condi-tions might have led to as many as 60,000 deaths(26). Nearly a decade later, public discussion of

the accident increased in response to the 1975publication of the book, We Almost Lost Detroit (25).

In 1975, a fire started by a worker using a can-dle to test for air leaks spread through the elec-trical system of the Tennessee Valley Authority’s(TVA’s) Browns Ferry plant in Alabama. The firecaused some loss of core coolant in one unit ofthe plant, and disabled the reactor’s safety sys-tems (1 1). Because of confusion about how to putit out, the fire burned out of control for 7 hoursbefore being extinguished. Again there was no

loss of life and no release of radiation, but theincident was reported in the national media, in-creasing public fears of an accident. Critics feltthat the Nuclear Regulatory Commission’s (NRC’s)news release on the event—which emphasized

the safe shutdown of the reactor while downplay-ing the failure of the Emergency Core Cooling Sys-tem— misrepresented the nearness of a disasterand ignored the lack of foresight which the acci-dent demonstrated.

While these earlier accidents had an adverse

impact on popular support for nuclear technol-ogy, it was not until 1979 that a single accidenthad a direct, measurable impact on public opin-ion as reflected in national opinion polls. Thatspring, poor maintenance, faulty equipment, andoperator errors led to a loss of coolant and par-tial destruction of the core at the TMI Unit 2 reac-tor located in Pennsylvania. Radioactive waterspilled onto the floor of an auxiliary building,releasing a small amount of radioactivity to theenvironment, although the total radiation dosereceived by the population in the vicinity was farless than their annual exposure to natural andmedical radiation (31 ). On March 30, GovernorThornburgh advised pregnant women and pre-school children to leave the area within a 5-mileradius of TMI. This advisory was not lifted untilApril 9. Conflicting statements from authoritiescombined with obvious confusion at the reactorsite before and during the evacuation shookpublic confidence in the nuclear industry andState and Federal officials. Following the accident,majority support for nuclear power was lost, atrend that continues today. Local opposition tosome reactors around the country also increasedafter the accident, while local attitudes towardother reactors remained favorable (see CaseStudies at the end of this chapter).

Opinion polls taken after the accident at TMIindicated that at least half of those polled thoughtmore such accidents were likely. Since that time,other incidents, such as the rupture of steam gen-erator tubes at Rochester Gas & Electric’s Ginnanuclear plant in January 1982, have occurred atoperating reactors. There is some evidence thatthe public views these incidents, along with theTMI accident, as precursors to a catastrophic ac-cident that might kill thousands (67).

The handling of reactor safety issues by thenuclear industry and the NRC has led many peo-ple to the conclusion that both have seriouslyunderestimated safety problems. For example,

25-450 0 - 84 - 15 : QL 3



Ch. 8—Public Attitudes Toward Nuclear Power . 215

Table 32.-Major National Groups Influencing Public Opinion For and Against Nuclear Power

Groups supporting nuclear power Groups opposing some aspects of nuclear power

Category 1: Large organizations with a focus on nuclear Category 1: Groups with a focus on nuclearenergy targeting a broad audience. energy and alternatives to it.

— U.S. Committee for Energy Awareness — Union of Concerned Scientists — Atomic Industrial Forum — Critical Mass Energy Project of Public — American Nuclear Society Citizen, Inc.

Category 2: Lobbying organizations with a primary or secondary — Nuclear Information and Resource

focus on nuclear energy. Service — Americans for Nuclear Energy — Safe Energy Communications Council — American Nuclear Energy Council Category 2: Large environmental groups that — Americans for Energy Independence participate in lobbying and public criticism of

Category 3: Trade and professional associations that support nuclear energy.commercial nuclear energy. —

— Edison Electric Institute —

— American Public Power Association —

— National Rural Electric Cooperative Association —

— Institute of Electrical and Electronics Engineers —

— American Association of Engineering Societies —

— Health Physics Society —

— Scientists and Engineers for Secure EnergyCategory 4: Industry research organizations indirectly influencing

public opinion. — Electric Power Research Institute — Institute for Nuclear Power Operations

— Nuclear Safety Analysis Center

Sierra ClubNational Audubon SocietyNatural Resources Defense CouncilFriends of the EarthEnvironmental Policy CenterEnvironmental Defense FundEnvironmental Action, Inc.

SOURCES Terry Lash, “Survey of Major National Groups Influencing Public Opinion Against Nuclear Power, ” Office of Technology Assessment contractor report,April 1983, M & D Mills, “Activities of GrOUpS Which Influence Public Opinion in Favor of Nuclear Power, ” Off Ice of Technology Assessment contractorreport, May 1983

about nuclear power and much more committedto their beliefs than the general public. They acton these beliefs both in seeking to influencenuclear power policies at the State and Federallevel and in attempting to convince the publicof their point of view.

The nuclear establishment sometimes blamesnuclear critics for the growth of public opposi-tion to nuclear power. However, to some extentthese individuals simply are reflecting the con-cern of the wider public which has grown inresponse to reactor accidents and the increasingfinancial problems of the utility industry. In ad-dition, the success or failure of both advocatesand critics depends in part on public responseto their arguments. A 1983 opinion poll indicatesthat Ralph Nader, a leading environmentalist, isconsidered very believable on energy matters (9).Electric utility trade associations are consideredsomewhat less believable, and nuclear industry

associations have much lower credibility amongpoll respondents. Thus, it appears that the publicmay be more willing to listen to and accept thearguments of nuclear critics than those of advo-cates.

The Impact of Public Opinionon Nuclear Power

Public concerns about reactor safety, nuclearwaste disposal, and rising construction costs havehad a particularly notable impact on State policiesaffecting nuclear power. As discussed in chapter6, State Public Utility Commissions (PUCs) mustgrant a license certifying the need for power priorto construction of any type of new powerplant.Because PUCs have veto power over new plants,based on economic and financial criteria, Statelaws essentially can halt further development ofnuclear power. While critics and advocates havebeen involved in voter-initiated referenda restrict-ing further licensing of nuclear plants, it is ulti-mately the voters of the State (the “general pub-lic”) who decide whether or not to approve theserestrictions. Table 31 provides a history of Statevotes on nuclear energy referenda. Overall, thetrend appears to reflect accurately the trends

shown in public opinion polls, declining from alarge margin of suppt for nuclear power in 1976to an ambivalent position today. While all sevenrestrictive proposals were defeated in 1976,voters in Oregon and Massachusetts approved




initiatives limiting new reactor construction in1980 and 1982.

In California, State legislators approved a lawrestricting nuclear power development in 1976to head off a more stringent Statewide referen-dum with similar provisions that was then turneddown only a few months later. The law passedby the legislature was upheld by the U.S. Su-preme Court in April 1983. Other State legisla-tures and a few PUCs have limited further con-struction of nuclear plants by legislation or regula-

tion. A complete list of State laws and regulations(including those enacted by voter referenda) af-fecting nuclear power is given in table 33. Be-cause of these laws and regulations, utilities in10 States cannot obtain State licensing of pro-posed nuclear reactors until certain conditions,such as a clear demonstration of high-level waste

disposal, are met.Even in those States where nuclear power de-

velopment is not limited by law or regulation,State politics can influence utility decisions about

Table 33.—State Laws and Regulations Restricting Construction of Nuclear Powerplants

YearState Type of action approved Citation Provisions

California

Connect i cut

Kentucky

Maine

Maryland

Massachusetts

Montana

Oregon

Vermont

Wisconsin

Washington

Law-byIegislaturea

Law-by

legislature

Law-bylegislatureResolution-bylegislatureLaw-bylegislature

Law-bylegislature

Law-byreferendum

Law-byreferendum

Law-byreferendum

Law-bylegislature

Regulation-byPublic ServiceCommission

Law-byreferendum

1976

1979

1983

1982

1977

1981

1982

1978

1980

1975

1978

1981

Cal Pub Res Code, No licensing of new plants until Federal Govern-Sees. 25524.1-25524.2

H-5096, approved

June 18

H.B. 5237

HR-85,adopted March 26Me Rev Stat Ann,Tit 10, Sees. 251-256 (West 1980)

Ann Code Md,Health-Environ-mental, Tit 8,Sec. 402Question 3,Approved Nov. 2(Chap. 503, Acts

of 1982)

Mt Code Ann,Tit 75, Sees.20-201, 20-1203

Measure No. 7,Approval Nov. 4

Vt. Stat Ann,1970, V.8, Tit 30,Sec. 248cDkt No 05-EP-1,Wis Pub ServComm, Aug. 17

Chap. 80.52, Rev.Code ofWashington

ment approves a demonstrated high-levelwaste disposal technology and fuel rodreprocessing technology is available.

No licensing of a fifth plant until Federal

Government approves a demonstrated high-Ievel waste disposal technology.

Limits construction costs of Millstone 3 torate-payers to $3.5 billion.

Declares the State’s intention to prohibitconstruction of plants.

No licensing of new plants until FederalGovernment demonstrates high-level wastedisposal and a majority of voters approve in areferendum vote.

No licensing of new generators of nonmedicallow-level waste until Federal Governmentdemonstrates waste disposal or an interstatecompact is in effect.

No licensing of new plants or nonmedicallow-level radioactive waste disposal sites untila Federally approved storage facility is

operating, and other conditions, includingvoter approval, have been met.No licensing of new plants until all liability

limits for an accident are waived, a bond isposted against decommissioning costs, andother conditions, including voter approval, aremet.

No licensing of new plants until FederalGovernment provides high-level waste disposaland a majority of voters approve in areferendum.

No licensing of new plants without GeneralAssembly approval.

No licensing of new plants without progresson waste disposal, fuel supply, decom-missioning, and other economic issues.

No issuance of bonds for major new energyfacilities (including nuclear plants) withoutvoter approval.

aon Apr. ZI, 1983, the Us. Supreme Court upheld the COnSI iIUIiOditY Of this law

SOURCES: Atomic Industrial Forum, State Codes, NRC Office of State Programs.




nuclear plants. Whether Public Utility Commis-sioners are directly elected or appointed by anelected Governor, they are sensitive to State pol-itics and broad public opinion. Public concernsabout nuclear power may lead Utility Commis-sioners to disallow rate increases needed tofinance completion of plants under construction

or to simply deny a license entirely. Public op-position at the local level, too, can discourageutilities from implementing planned nuclearplants. For example, Portland General Electric inOregon canceled its planned Pebble Springs reac-tor in 1982 following a lengthy siting controver-sy that made the project less economically attrac-tive. The approval of a State referendum in 1980banning licensing of new plants until waste dis-posal technology was available contributed to theutility’s decision.

Over the past few years, public concern about

reactor safety in reaction to the accident at TMIhas encouraged additional NRC safety studies andnew regulatory requirements, increasing nuclearpower costs and making it less attractive to util-ities. (A more detailed analysis of the costs of reg-ulatory requirements is included in ch. 6.) Thistrend is partially a continuation of increasingpublic concern about environmental quality thatbegan in the late 1960’s. Translated into laws andregulations, those concerns drove up the priceof both nuclear and coal-fired powerplants as util-ities were required to incorporate more pollution

control technology into new and existing plants.Negative public perceptions may also affect theavailability of financing for new nuclear plants.The financial problems caused by the accidentat TMI discouraged some investors and brokersfrom investing in utilities with nuclear plantsunderway, driving up the cost of capital for those

utilities. Finally, negative public attitudes affectnuclear power’s future in less tangible ways: Themost gifted young engineers and technicians maychoose other specializations, gradually reducingthe quality of nuclear industry personnel. And, util-ities simply may not choose nuclear plants if theyperceive them as bad for overall public relations.

The future of nuclear power in the UnitedStates is very uncertain due to a variety of eco-nomic, financial, and regulatory factors outlinedin other chapters of this report. Both parties tothe nuclear debate are bringing these factors be-

fore the broader public. Some may argue that theissues are too complicated for the general publicto contend with. However, as Thomas Jeffersonsaid, “When the people are well informed, theycan be trusted with their own government. ”None of the conditions seen by utilities as a re-quirement for a revival of the nuclear industry–regulatory stability, rate restructuring, andpolitical support-can be met without greaterpublic acceptance. Thus, unless public opiniontoward nuclear power changes, the future pros-pects for the nuclear industry will remain bleak.

THE EXPERTS’ VIEW

I n c o n t r a s t t o t h e p u b l i c , m o s t “ o p i n i o n

leaders,” particularly energy experts, support fur-

ther deve lopm ent of nuc lear po we r. This supp ort

is revealed both in opinion polls and in technicalstudies of the risks of nuclear power. A March1982 poll of Congress found 76 percent of mem-bers supported expanded use of nuclear power(50. In a survey conducted for Connecticut Mu-tual Life Insurance Co. in 1980, leaders in religion,

business, the military, government, science, edu-cation, and law perceived the benefits of nuclearpower as greater than the risks (19). Among thecategories of leaders surveyed, scientists were

particularly supportive of nuclear power. Seventy-four percent of scientists viewed the benefits ofnuclear power as greater than risks, comparedwith only 55 percent of the rest of the public.

In a recent study, a random sample of scien-tists was asked about nuclear power (62). Ofthose polled, 53 percent said developmentshould proceed rapidly, 36 percent said develop-

ment should proceed slowly, and 10 percentwould halt development or dismantle plants.When a second group of scientists with particularexpertise in energy issues was given the same




choices, 70 percent favored proceeding rapidlyand 25 percent favored proceeding slowly withthe technology. This second sample included ap-proximately equal numbers of scientists from 71disciplines, ranging from air pollution to energypolicy to thermodynamics. About 10 percent ofthose polled in this group worked in disciplines

directly related to nuclear energy, so that theresults might be somewhat biased. Supportamong both groups of scientists was found toresult from concern about the energy crisis andthe belief that nuclear power can make a majorcontribution to national energy needs over thenext 20 years. Like scientists, a majority of engi-neers continued to support nuclear power afterthe accident at Three Mile Island (69).

Of course, not all opinion leaders are in favorof the current U.S. program of nuclear develop-ment. Leaders of the environmental movement

have played a major role in the debate aboutreactor safety and prominent scientists are foundon both sides of the debate. A few critics ofnuclear power have come from the NRC and thenuclear industry, including three nuclearengineers who left General Electric in order todemonstrate their concerns about safety in 1976.However, the majority of those with the greatestexpertise in nuclear energy support its furtherdevelopment.

Analysis of public opinion polls indicates thatpeople’s acceptance or rejection of nuclear

power is more influenced by their view of reac-tor safety than by any other issue (57). As dis-cussed above, accidents and events at operatingplants have greatly increased public concernabout the possibility of a catastrophic accident.Partially in response to that concern, technicalexperts have conducted a number of studies ofthe likelihood and consequences of such an ac-cident. However, rather than reassuring the pub-lic about nuclear safety, these studies appear tohave had the opposite effect. By painting a pic-ture of the possible consequences of an accident,the studies have contributed to people’s view of

the technology as exceptionally risky, and thedebate within the scientific community about thestudy methodologies and findings has increasedpublic uncertainty.

The Controversy Over Safety Studies

The Atomic Energy Commission (AEC) com-pleted its first major study of the consequencesof a reactor accident involving release of radio-activity in 1957. Commonly known as WASH-740, the study was based on a very small (by

today’s standards) 165-megawatt (MW) hypo-thetical reactor. In the worst case, an accidentat such a plant was estimated to kill 3,400 peo-ple (5). While the study itself did not become asource of public controversy, its findings contrib-uted to concern about the impacts of an accident.

In 1964, AEC initiated a new study to updateWASH-740 based on a larger, 1,000-MW reactor.The study team found that a worst-case accidentcould kill as many as 45,000 people but was un-able to quantify the probability of such an acci-dent. Rather than publish these disturbing find-ings, AEC chose to convey the results to Congress

in a short letter. Nuclear critics were very dis-turbed by this action, which they viewed as anattempt to keep the facts away from the public(22). In recent years, awareness of AEC’s handl-ing of this early safety study has added to publicskepticism about the credibility of both that agen-cy and its successor, the NRC.

In 1974, AEC published the first draft of the Re-actor Safety Study, also known as WASH-1400or the Rasmussen report. A panel of scientistsorganized by the American Physical Society (APS)found much to criticize in this report. The panel

noted that AEC’s fatality estimates had consideredonly deaths during the first 24 hours after an ac-cident, although radioactive cesium released inan accident would remain so for decades, expos-ing large populations to adverse effects. The mostserious forms of illness resulting from a reactoraccident, the APS reviewers argued, would beforms of cancer that would not show up untilyears after the accident. Other APS reviewersfound fault with the Rasmussen report’s methodsused to predict the performance of emergencycooling systems (23).

On October 30, 1975, the NRC, which had as-sumed the regulatory functions of the formerAEC, released the final version of WASH-1400.Again, there was an extensive, widely publicized




debate over the document. The Union of Con-cerned Scientists released a 150-page report cri-

tiquing the study, and in June 1976, the HouseSubcommittee on Energy and Environment heldhearings on the validity of the study’s findings(71). As a result of these hearings, NRC agreed

to have a review group examine the validity of

the study’s conclusions.Three ye a rs late r, in Sep tem be r 1978, the re -

view group co ncluded that although the Reac -

tor Safety Study represented a substantial advance

over previous studies and its methodology was

basically sound, the actual accident probability

estimates were more uncertain than had beenassumed in the report (35). The panel also wascritical of the executive summary, which failed toreflect all of the study findings. The followingJanuary, the NRC accepted the conclusions of thereview panel. In a carefully worded statement,the agency withdrew its endorsement of the nu-merical estimates contained in the executive sum-mary, said that the report’s previous peer reviewwithin the scientific community had been “inad-equate,” and agreed with the panel that the dis-aster probabilities should not be used uncritical-ly (47).

Two studies published in 1982 continued thedebate over the validity of accident probabilityestimates included in the Rasmussen report. Thefirst, conducted by Science Applications, Inc.(SAI) for the NRC, was based on the actual op-erating history of U.S. reactors during the 1969-79

period. By examining the frequency of precur-sors that could lead to an accident involving coredamage or meltdown, SAI estimated that theprobability of such an accident during the pre-TMI decade was much greater than suggested bythe Rasmussen report (43). In response, the In-stitute for Nuclear Power Operations (lNPO—anuclear industry safety research group) publisheda report arguing that SAI’s probability estimateswere about 30 times too high, and that the ac-tual probability of a core-damaging accident wascloser to the 1 in 20,000 reactor years estimatedin the Rasmussen report (28). This controversy

has not yet been resolved.

While debate over the SAI report was limitedto a small community of safety experts, a morerecent study aroused a widespread public con-

troversy that continued for several weeks. Thisanalysis, known as the Sandia Siting Study, wasinitiated to determine the sensitivity of the con-sequences of reactor accidents to local site char-acteristics (2). While the Sandia team did notstudy accident probabilities in depth, they esti-mated the probability of a “Group 1“ or (worst-

case) accident involving a core meltdown, failureof all safety systems, and a large radioactive re-lease, at 1 in 100,000 reactor years. The conse-quences of this and other less severe hypotheticalaccidents were estimated for 91 U.S. reactor sitesusing local weather and population data and as-suming a standard 1, 120-MW reactor. At the cur-rent site of the Salem, N. J., reactor on the Dela-ware River under the most adverse weather con-ditions and assuming no evacuation of the localpopulation, a Group 1 accident at the hypotheti-cal reactor was estimated to cause 102,000“early” deaths within a year of the accident. If

the hypothetical reactor were located at Bu-chanan, N. Y., where the Indian Point plant nowstands, a Group 1 accident under the worst-caseweather conditions (the accident would be fol-lowed by a rainout of the radioactive plume ontoa population center) might cause $314 billion inproperty damage, according to the study esti-mates.

Although the Sandia report itself did not includeestimates of the “worst-case” accident conse-quences, background information containing theestimates and a copy of the draft report were

leaked to the press on November 1, 1982. Mediaaccounts that day highlighted the high death andproperty damage estimates, while downplayingthat part of the analysis which indicated that con-sequences of this severity had only a 0.0002-percent chance of occurring before 2000 (51).Some accounts suggested that the worst-caseconsequences had the same probability as theGroup 1 or worst-case accident, which was esti-mated to have a 2-percent chance of occurringbefore the end of the century.

That same day, the NRC held a press confer-

ence to clarify the purpose and findings of thestudy, and on November 2, Sandia National Lab-oratory issued a statement saying that wire serv-ice accounts “seriously misinterpret the conse-quences of nuclear power reactor accidents. The




probability of a very severe nuclear power reac-tor accident is many thousands of times lowerthan stated in these accounts” (63). The nuclearindustry took out full-page ads in major nationalpapers to try to counteract the story. At the sametime, however, nuclear critics emphasized thatthe Sandia draft report itself had excluded the

worst-case consequence data and argued that“the NRC is once again feeding selective data tothe public on the theory that they know best whatinformation the public should have” (73). Whilenuclear advocates argued that the report’s find-ings on accident consequences had been great-ly overstated by the press, critics charged thatdata were used incorrectly in developing thoseestimates. Examining the same information on ac-cident probabilities at individual plants used bythe Sandia team, the Union of Concerned Scien-tists found that the likelihood of an accident in-volving a release of radioactivity might be much

greater than assumed in the Sandia report (65).This debate, too, has not been resolved.

The Impact of Risk Assessmentson Public Opinion

The release of the Rasmussen report raised par-ticular concerns about nuclear power for somepeople because of the public disagreementsamong the “experts” that resulted. In June 1976hearings held by the House Interior Committee,scientists from the Massachusetts Institute ofTechnology and Princeton and Stanford Univer-

sities, as well as a high-level official of the En-vironmental Protection Agency, testified aboutthe methodological weaknesses and limitationsof the report. Thus, as Princeton physicist FrankVon Hippel pointed out at the hearings, “Insteadof dampening the fires of controversy, the pub-lication of the Rasmussen report has had the ef-fect of adding fuel to them” (71).

The controversy over the Rasmussen report,like the rest of the nuclear debate, contains manyelements of “disputes among experts” ascharacterized by sociologist Alan Mazur: argu-

ing past one another instead of responding towhat the opposing expert has actually stated; re- jecting data that develop the opponent’s case;interpreting ambiguous data differently; and, con-

sequently, increasing polarization (41 ). Both crit-ics and supporters of the study focused on themethodology and quality of data. The debateover the study continues today, with critics argu-ing that NRC’s 1979 statement was a “rejectionof the report’s basic conclusion,” “repudiatingthe central finding of the Rasmussen report” (23).

Meanwhile, INPO challenges the methodologyand data of SAI’s more recent safety study, argu-ing that the Rasmussen report’s probability esti-mates are still valid.

Although the general public is uncertain aboutnuclear power, most people have more faith inscientific “experts” than in any other source onnuclear power questions (20,39,57). Because ofthis faith, public disputes among scientists andother energy experts, as in the case of the Ras-mussen report, have a particularly negative im-pact on public acceptance of nuclear power.

Rather than attempting to follow the debate andsort out the facts for themselves, many peoplesimply conclude that nuclear technology has notyet been perfected. in other words, if the “ex-perts” cannot agree on whether or not nuclearpower is safe, the average citizen is likely toassume it is probably unsafe. In Austria, thegovernment attempted to resolve the growingcontroversy over nuclear power by structuringa series of public debates among scientists withopposing views. Rather than reassuring the pub-lic, the debate led to increased public skepticismand ultimately to a national referendum that

killed that country’s commercial nuclearprogram.

If public debates about nuclear safety studieshave only fueled the fires of controversy andadded to public skepticism, what can be doneto make nuclear power more acceptable to thepublic? In order to answer that question, weneed a better understanding of the public’s per-ceptions of nuclear power. In particular, it willbe useful to compare the public’s view of the risksof nuclear energy with the risks estimated by mostnuclear experts. For example, if public percep-

tions of risk were based on misinformation, im-proved public education programs might be anappropriate response. However, this does not ap-pear to be the case.




FACTORS INFLUENCING THE PUBLIC’S VIEWOF NUCLEAR SAFETY

Perceptions of Risk and Benefit

Studies of risk perception reveal a gap betweenlay people’s judgment of nuclear hazards and the

risks estimated by technical risk assessments. Ina 1979 study by Decision Research two smallgroups of informed people in Eugene, Oreg. (col-lege students and members of the League ofWomen Voters) were asked to compare the ben-efits and risks of a variety of activities, rangingfrom smoking to vaccinations to swimming. Thebenefits of nuclear power were viewed as negli-gible, and the risks were judged to be almost asgreat as motor vehicle accidents, which claimabout 50,000 lives each year (68).

Although the two groups estimated that the

number of deaths from nuclear power in an av-erage year would be fewer than the number ofdeaths from any of the other activities or tech-nologies, they used a very high multiplying fac-tor to indicate how many deaths would occur ina “particularly disastrous” year. Almost 40 per-cent of the respondents estimated more than

10,000 fatalities would occur within 1 year, and

more than 25 percent guessed there would be100,000 fatalities. Many of the respondents ex-pected such a disaster within their lifetimes, whilethe Sandia study suggests that there is only oneU.S. reactor site–Salem, N.J.–at which an acci-

dent might cause as many as 100,000 “early” fa-talities and estimates that these consequenceshave only a 10-6 or 0.000001 chance of occur-ring at that site. That analysis suggests that theaverage American (who does not live near thatsite) has an even lower probability of being killedwithin a year of a reactor accident. In general,it appears that public perceptions of the possi-ble consequences of an accident correspondsomewhat with the findings of the most recenttechnical studies, but that the probability of suchconsequences is greatly overestimated.

Data from the Netherlands confirm the public’sperception of nuclear power as uniquely hazard-

ous. Over 700 adults of varying ages, living atvarying distances from industrial activities wereasked to judge the “riskiness” of a wide range

of activities in 1978 and 1979. Nuclear power was judged to be more risky than most of the otheractivities and technologies, including drunk driv-ing, transporting chlorine by freight train, andworking as a big-city policeman (74).

Several factors appear to enter into people’sviews of nuclear power as particularly risky. First,respondents in both the Netherlands and Oregonwere concerned about the size of a potential ac-cident and the lack of individual control in pre-venting an accident. In the Oregon study, nuclearrisks also were seen as “unknown to the publicand to science” and as particularly severe anddreaded. Both the Oregon study and opinion sur-veys show that about 40 percent of the Americanpublic believe that a nuclear plant can explode

like an atomic bomb, even though such an ex-plosion is physically impossible. Familiarity alsoplayed a role in people’s judgments. In the Ore-gon study, nuclear risks were perceived as greaterbecause they were unfamiliar. Another factor en-tering into risk perceptions was people’s difficultyin assessing the probability of a reactor accident.

People’s opinions about nuclear power andother “hazardous” activities and technologies arenot determined by perceptions of risk alone. Theperceived benefits offered by a technology mustbe weighed against the perceived risk in deter-

mining how acceptable the technology is. Mostactivities, including development of nuclear pow-er, are undertaken initially in order to achievebenefits, not avoid losses, and for many activitiesthe expected benefits far outweigh the potentiallosses.

in the case of nuclear power, perceptions ofbenefit may have played an important role in thetrend of public opinion. During the 1950’s and1960’s, when electricity demand was growingrapidly, the development of nuclear energy waspromoted as a means to meet future demand and

there was little apparent opposition to the tech-nology. As electricity demand slowed in the1970’s and 1980’s some people may have seennuclear power as less vital to economic growth,so that concerns about risk became more prom i-



Ch. 8–Public Attitudes Toward Nuclear Power • 223

Figure 40.— Relationship Between Judged Frequency and the Actual Number of Deathsper Year for 41 Causes of Death

disease

I

I

II

I

I1

1u

1 10 100 1,000 10,000 100,000 1,000,000

Estimated number of deaths per year

NOTE Respondents were told that about 50,000 people per year die from motor vehicle accidents. If judged and actual frequencies were equal, the data would fall onthe straight line. The points and the curve fitted to them represent the averaged responses of a large number of lay people. While people were approximately ac-curate, their judgments were systematically distorted. To give an idea of the degree of agreement among subjects, vertical bars are drawn to depict the 25th and75th percentile of individual judgment for botulism, diabetes, and all accidents. Fifty percent of all judgments fall between these limits. The range of responsesfor the other 37 causes of death was similar.

SOURCE: SIovic, et at. (68). Reproduced by permission of P. Slovic.

to recall. Moreover, the “availability” of an event These findings help explain the increased pub-

in people’s memories may be distorted by a re- lic opposition to nuclear power reported in opin-

cent disaster or vivid film. Because life is too short ion polls over the past decade. Nuclear power’s

to actually experience all the hazards shown in historic connections with the vivid, imaginable

figures 40 and 41, people tend to focus ondangers of nuclear war lead people to associate

dramatic and well-publicized risks and hence to the technology with catastrophe. Accidents suchoverestimate their probability of occurrence (68). as the fire at Browns Ferry and the near-meltdown




Figure 41 .—Expert and Lay Judgments of Risk Plotted AgainstTechnical Estimates of Annual Fatalities

I I I I I

Technical estimate of number of deaths per yearNOTE: Each point represents the average responses of the participants. The broken lines are the straight lines that best fit the points. If judged

and technically estimated frequencies were equal, the data would fall on the solid line. The experts’ risk judgments are seen to be moreclosely asociated with technical estimates of annual fatality rates than are the lay judgments.

SOURCE: Slovic, et al., (68). Reproduced by permission of P. Slovic.





In May 1979, 2 months following the Three Mile Island accident, there was a demonstration of about 200,000 people onthe U.S. Capitol Grounds. Speakers urged the U.S. Congress to curtail further nuclear construction

at TMI have added to the image of disaster in peo-ple’s minds. The publicity surrounding theseevents, movies and books such as “The ChinaSyndrome” and We A/most Lost Detroit, and theestimates of deaths in various safety studies havefurther enhanced the “availability” of nuclearpower hazards in people’s minds, creating a false“memory” of a disaster that has never occurredat a commercial nuclear reactor. Public educa-tion about nuclear safety systems, by identifyingthe various hazards those systems are designedto guard against, may only serve to increase theperceived risks of the technology.

The Decision Research analysts also comparedlay judgments of risks with the judgments of na-tionally known professionals in risk assessment(see fig. 41). The judgments of experts were muchcloser than those of the lay people to statistical-

ly calculated estimates of annual fatalities asso-ciated with various risky technologies and activ-ities (21 ). However, while the experts knew morefacts, their risk assessments also were found tobe greatly influenced by personal judgment.Thus, expert studies also are subject to errors, in-cluding overconfidence in results and failure toconsider the ways in which humans can affecttechnological systems. This latter problem wasdemonstrated clearly during the TMI accident,which was caused in part by human error.

Because technical experts, like the general pubIic, face limitations in evaluating the risks posed

by nuclear power, it appears that there may beno single right approach to managing the tech-nology. instead, it is most appropriate to involvethe public in order to bring more perspectivesand knowledge to bear on the problem. There




are at least two other reasons for involving thepublic in decisions about nuclear energy. First,without public cooperation, in the form of politi-cal support, observance of safety rules, and rea-sonable use of the court system, nuclear powercannot be managed effectively. Second, as ademocracy, we cannot ignore the beliefs and

desires of our society’s members (2 I).While public perspectives already are reflected

to some extent in NRC decisions, further effortscould be made to involve the public. More re-search could be conducted to define and quanti-fy public opinion, and dialog with nuclear criticscould be expanded with more attention paid tothe substance of their concerns. Perhaps the most

important step in reducing public fears of nuclearpower is improved management of operating re-actors to eliminate or greatly reduce accidentsand other operating difficulties. Even though ac-cidents at commercial reactors in the UnitedStates have never caused a civilian death, thepublic views both accidents and less serious

events at operating reactors as precursors to acatastrophe. An accident with disastrous conse-quences already is viewed as being much morelikely than technical studies and experts project,and any continuation of accidents or operatingproblems will tend to confirm that perception.Approaches to increasing public acceptance arediscussed in the conclusion of this chapter.

VALUES AND KNOWLEDGE

Which is More Important?

Some analysts of public opinion have arguedthat basic values—those things that people viewas most morally desirable—play a relatively smallrole in influencing people’s attitudes toward nu-clear power. For example, Mazur has argued thatmost of the general public, unlike energy activists,do not “embed their positions for or against atechnology in a larger ideological framework ofsocial and political beliefs” (41). In addition,nuclear advocates sometimes suggest that it isprimarily a lack of knowledge which leads peo-ple to oppose nuclear energy, and that better ed-ucation programs would increase public accept-ance (1 3,42). However, as discussed below, theavailable evidence calls both arguments intoquestion.

Although energy experts who are very knowl-edgeable about nuclear power generally supportthe technology, studies of the effects of slightlyincreased knowledge on attitudes among thebroader public have yielded mixed conclusions.Two studies found greater support for nuclear

power among more knowledgeable persons, butanother found the opposite, and several studieshave found no significant relationship (46). Forexample, a 1979 survey conducted just prior tothe TM I accident revealed only a very weak rela-

tionship between knowledge about nuclear pow-er and support for the technology among the gen-eral public. These findings supported a “selec-tive perception” hypothesis in which thosestrongly favoring or strongly opposing nuclearenergy selected and used information to bolstertheir arguments. Attitudes toward nuclear poweramong all respondents were influenced heavilyby preexisting political beliefs and values (58).These results could help to explain why the ac-cident at TMI appeared to have little impact onsome people’s opinions about safety. For those

who already were firmly convinced that nuclearpower was safe, the accident confirmed the ef-fectiveness of safety systems. For those who wereskeptical, it reinforced uncertainties.

A recent analysis of national survey data pro-vides additional evidence that people’s valuesand general orientations may be stronger deter-minants of nuclear power attitudes than specificknowledge about energy or nuclear power issues.In this study, sociologist Robert Mitchell testedthe strength of the correlation between various“irrational” factors—such as belief that a nuclear

plant can explode–and people’s assessments ofreactor safety. While the analysis showed that thepublic was generally misinformed about energyissues, this lack of knowledge appeared to havelittle effect on attitudes toward nuclear safety and



Ch. 8—Public Attitudes Toward Nuclear Power Ž 227

hence on overall attitudes toward nuclear power.When Mitchell went onto test the correlation be-tween values and attitudes toward reactor safe-ty, he found a much stronger relationship. Envi-ronmentalism was associated closely with con-cern about nuclear safety among women, whileskepticism about whether the future benefits of

scientific research would outweigh the resultingproblems appeared to have a strong influence onmen’s concerns about reactor safety (46). Severalother studies also indicate that values have playedan important role in both the growth of the anti-nuclear movement and continued support fornuclear power (8,27,32).

Values

A value has been defined as “an enduringbelief that a specific mode of conduct or end-stateof existence is personally or socially preferred”

(61). While all people share the same values tosome degree, each individual places different pri-orities on different values. This ordering of theabsolute values we are taught as children leadsto the development of an integrated set of beliefs,or value system. Because values have their originsin culture and society, changes in societal expec-tations can, over time, lead to changes in an in-dividual’s value system.

During the 1960’s and 1970’s, the emergenceof new social movements both reflected and encouraged changes in the priority some Americans

placed on different values. Values that appearedto become more prominent to many people in-cluded equality of all people, environmentalbeauty, and world peace. Critics of nuclear pow-er (and, to a lesser extent, proponents) were suc-cessful in attracting broader public support by ap-pealing to these emerging values, and by linkingtheir organizing efforts with the related peace,feminist, and environmental movements. Untila convincing case is made that nuclear power isat least as consistent with these values as otherenergy sources, it will have difficulty gaining ac-ceptance with those who place a high priority on

these values.Overall, Americans are very supportive of sci-

ence and technology, viewing them as the bestroutes to economic progress (39). The public’s

enthusiasm is reflected in the current computerboom and in the emergence of a flood of sciencemagazines such as Omni, Discover, Science 83,and Technology as well as new television pro-grams including “Cosmos,” “Life on Earth,” and“Nova,” and the reliance on high technology byboth major political parties. However, this sup-

port is tempered by a growing concern about theunwanted byproducts of science, including accel-erating social change, the threat of nuclear war,and environmental pollution. The National Opin-ion Research Center recently compared a nation-al poll of adult attitudes toward science taken in1979 with a similar survey conducted in 1957.They found that, over the 20-year period, anincreasing number of survey respondents be-lieved that “science makes life change too fast”or “breaks down people’s ideas of right andwrong.” The percentage of respondents who be-lieved that the benefits of science outweigh the

harms declined from 88 percent in 1957 to 70percent in 1979 (46). This curious duality of at-titudes may help to explain the public’s ambiva-lent attitude toward nuclear power. While ac-ceptance of the technology has declined, the rateof change has been slow, and votes on referendahave demonstrated that Americans are unwillingto forego the nuclear option entirely.

One of the most undesired products of modernscience is the threat of nuclear war. Because ofthe technological and institutional links betweennuclear power and nuclear weapons, opposition

to buildup of nuclear weapons leads some peo-ple also to oppose development of civilian nu-clear energy. AEC, which developed and testedweapons after World War II, was the original pro-moter of commercial nuclear power. In the late1950’s and early 1960’s, growing public concernabout radioactive fallout from AEC’s atomicbomb testing provided a context for increasingfears about the possibility of radioactive releasesfrom nuclear powerplants. Some prominent sci-entists spoke out against both nuclear power andnuclear weapons, and links developed betweengroups opposing the arms race and nuclear pow-er (48). However, after the United States and theSoviet Union signed the Nuclear Test Ban Trea-ty in 1963, concerns about both nuclear falloutand nuclear power temporarily subsided.




Since the mid-1970’s, rapid international de-velopment of nuclear power and growing globaltensions have led to increasing concern about thepossible proliferation of nuclear weapons fromnuclear power technologies. Organizers in thepeace movement and nuclear critics have builton this concern in an attempt to renew the early

linkages between the two movements. Case stud-ies of the Maine Yankee and Diablo Canyon nu-clear plants indicate that concern about nuclearweapons contributed to opposition to theseplants during the late 1970’s and early 1980’s (seeCase Studies).

Today, a single Federal agency–the Depart-ment of Energy—still is responsible for researchand development of both nuclear weapons andcommercial nuclear power. in addition, some pri-vate firms are involved in both nuclear energyand nuclear weapons. These connections en-courage a linkage in people’s minds between thepeaceful and destructive uses of nuclear energy.Due to growing concern about the rapid buildupof nuclear weapons, some groups critical of nu-clear energy are shifting resources toward weap-ons issues. However, most local groups and na-tional organizations continue their efforts to im-prove the safety of nuclear power as they expandtheir focus to include nuclear weapons. Thelinkages between environmental and energygroups and anti-nuclear weapons groups maystrengthen the environmental groups and helpthem maintain their criticism of commercial

nuclear power (33).On the pronuclear side of the debate, orga-

nizers have emphasized the importance of nucle-ar power to national energy independence whichis in turn linked with national security. Analysisof 10 years of public opinion polls indicates thata view of nuclear power as an abundant Ameri-can resource which could reduce foreign oil de-pendence is a very important factor in favoringcontinued development of nuclear power (57).

One of the most important values to affectopinions on nuclear power is environmentalism.

A 1972 poll by Louis Harris&Associates indicatedthat many Americans believed that the greatestproblem created by science and technology waspollution (46). Polls taken in 1981 indicate that

most Americans continue to strongly support en-vironmental laws despite recessions and an in-creasing skepticism about the need for govern-ment regulation of business (4).

The role of the environmental movement in co-alescing and leading the criticism of nuclearpower has been well-documented. The first na-tional anti-nuclear coalition (National Interveners,formed in 1972) was composed of local environ-mental action groups (40). By 1976, consumeradvocate Ralph Nader, who later became alliedwith the environmental movement “stood as thetitular head of opposition to nuclear energy” (30).Today, all of the major national environmentalgroups are opposed to at least some aspects ofthe current path of nuclear power developmentin the United States. While some of the groupsdo not have an official policy opposing nuclearpower, their staffs stay in close communication,and there is substantial cooperation and supporton nuclear energy issues (33). A list of thesegroups is shown in table 32.

Both sides of the nuclear debate have em-phasized environmental concerns to influencepublic opinion. in the late 1960’s and early1 970’s, opponents of local nuclear plants mostfrequently pointed to specific environmental im-pacts, such as thermal pollution, low-level radia-tion, or disruption of a rural lifestyle as reasonsfor their opposition. During that same period,nuclear proponents increasingly emphasized theair-quality benefits of nuclear power when com-

pared with coal (41 ). Today, environmentalist op-ponents of nuclear power are more concernedabout broad, generic issues such as waste dis-posal, plant safety, weapons proliferation, and “aset of troublesome value questions about hightechnology, growth, and civilization” (30). Thisevolution of concerns is demonstrated in twocase studies of local opposition to nuclear plants.In these cases, environmentalists at first did notoppose the local nuclear plant because it offeredenvironmental benefits when compared with acoal plant. However, those positions were laterreversed (see Case Studies).

Along with environmentalism, people’s generalorientations toward economic growth appear toinfluence their attitudes about nuclear power. In




1971, political scientist Ronald Inglehart identifieda shift in the value systems of many Americansaway from a “materialist” emphasis on physicalsustenance and safety and toward “post-materi-alist” priorities of belonging, self-expression, andthe quality of life. At that time, he hypothesizedthat this shift could be attributed to the unprece-

dented levels of economic and physical securitythat prevailed during the 1950’s and 1960’s.Based on analysis of more recent surveys, lngle-hart argued in 1981 that, despite economic un-

certainty and deterioration of East-West detente,

post-materialist values are still important to many

Americans. And, he says, those who place a high

priority on post-materialist values “form the core

of the opposi t ion to nuclear power” (27).

Like Inglehart, psychologist David Buss and hiscolleagues have observed two conflicting value

systems or “worldviews” among Americans (71).Using in-depth interviews with a random sampleof adults from the San Francisco area, they iden-tified “Worldview A“ which favors developmentof nuclear power as an important component ofa high-growth, high-technology, free enterprisesociety, and “Worldview B“ which includes con-cern about the risks of nuclear power along withan emphasis on a leveling off of material and tech-nological growth, human self-realization, and par-ticipatory decisionmaking.

While different priorities within Americans’ val-

ue systems appear to influence attitudes towardnuclear power, it is important to recognize thatthe public is not completely polarized. lnglehartnoted that “post-materialist” values can only begiven priority when basic human needs are met,making both priorities essential to individuals andto American society (27). An extensive nationalsurvey of attitudes toward growth conducted in1982 indicates that the public may be develop-ing a new perspective that includes both a desireto ensure opportunities for development and con-cerns about environmental quality (60). In thissurvey, few respondents could be classified as

totally favoring either resource preservation orresource utilization, and the majority appearedto be quite balanced in their views on economicgrowth. Those who leaned toward resource pres-ervation were more opposed to nuclear powerthan those who favored resource utilization.

However, even among those who most strongly

supported resource uti l ization, so percent in-

dicated that no more nuclear powerplants shouldbe built.

Views about “appropriate technology” as de-fined by the British economist, E. F. Schumacher,

may also affect attitudes toward nuclear energy.Most members of mainstream environmentalgroups share this view, which endorses tech-nologies that are inexpensive, suitable for small-scale application, and compatible with people’sneed for creativity (44,64). In 1976, Amory Lovinsbrought nuclear energy into the middle of thistechnology debate with publication of his article,“Energy Strategy: The Road Not Taken” inForeign Affairs magazine. In that article, Lovinsargued that America’s energy needs could be metby the “soft energy path” of conservation, renew-able energy and other appropriate technologies,

and rejected nuclear energy as unneeded, cen-tralizing, and environmentally destructive (37).These concerns were found important in the localopposition to one case study plant (see CaseStudies). Residents of that rural area at first ob-

jected to the plant on the basis that its electricitywas not needed locally, and that the localityshould not have to bear the impacts of plant con-struction when it would not reap the benefits.Later objections were based on the contentionthat the electricity produced would not beneeded anywhere in the surrounding three States.

Advocates of the appropriate technology phi-losophy fear that increased use of nuclear powerwill lead to a loss of civil liberties and individualfreedom, and decreased world stability due toweapons proliferation. The extent to which theseviews have been accepted by the American pub-lic is difficult to ascertain. National opinion pollsshowing that the majority of Americans prefersolar energy to all other energy sources and viewnuclear power as the least-favored energy optionwould appear to reflect such values (57). A re-cent survey conducted in the State of Washingtonshows that large majorities there share Lovins’

view that it is possible to have both economicgrowth and energy conservation (54). In addition,many people, even those skeptical of renewableenergy, share Lovins’ distrust of large centralizedorganizations (utilities and the government) thatpromote nuclear energy.

25-450 0 - 84 - 16 : QL 3




Another shift in American value systems thatmay help to account for increased opposition tonuclear power is a growing distrust of institutionsand their leaders in both the public and privatesectors. The Vietnam War and the Watergate in-vestigation contributed to growing public cyni-cism about the Federal Government during the

1960’s and 1970’s. By 1980, an extensive surveyof Americans revealed a dramatic gap betweenthe public and leaders in both government andindustry on questions of politics, morality, andthe family. Religious values were found to be ofprofound importance to the majority of Ameri-cans, and respondents indicated that they placeda greater emphasis than before on the moralaspects of public issues and leadership (59).

Some early supporters of nuclear power, in-cluding prominent environmentalists, felt be-trayed by the nuclear establishment when new

information about the uncertainties of the tech-nology became known. The nuclear industry’searly denials of the possibility of accidents, andthe Government’s handling of safety studies havecontributed to the critics’ and broader public’sskepticism. Some critics have expanded their ac-tivities from examination of technical safety issuesto include critiquing the nuclear regulatory proc-ess, and groups that formerly were concerned pri-marily with “watch dogging’ Federal agencieshave entered the nuclear debate. These activities,and Daniel Ford’s recent book, The Cult of the Atom, which focuses primarily on regulatory

“misdeeds” in the early nuclear program, maycontribute to the public’s disillusionment withgovernment in general and the NRC in particular.

The American public’s growing concern withleadership applies to business as well as govern-ment. Americans increasingly are skeptical of theability of both the public and private sectors toproduce quality work. According to Loyola Uni-versity professor of business ethics Thomas Don-aldson, survey data indicate that, despite an im-proving corporate record, the public has becomeincreasingly disappointed with corporate ethics

over the past 20 years. Corporations now areviewed as “part of the overall social fabric thatrelates to our quality of life,” not merely as pro-viders of goods and services (1 5).

This growing skepticism about industry andgovernment was reinforced by the accident atThree Mile Island. Post-TMl polls indicate that lessthan half of the public were satisfied with the waythe accident was handled by Pennsylvania Stateofficials and the NRC, and Americans were evenless pleased with the utility (General Public Util-

ities) and the plant designer (57). One observerhas described public reaction to the accident as“essentially a crisis in confidence over institu-tions” (30). A feeling that the nuclear utility wasbeing dishonest helped spark the first referendumto shut down Maine Yankee, and events at DiabloCanyon led to nationwide doubts about the credi-bility of the nuclear industry (see Case Studies).

A final societal change that has been closelyintertwined with negative attitudes toward nucle-ar power is the growth of the women’s move-ment. Public opinion polls over the past 20 years

have shown a strong correlation between genderand attitudes toward nuclear power: Womenare consistently more opposed (41). While thestrength of this correlation is well-known, thereasons for it are not clear. Environmental valuesand having young children have been linked withwomen’s opposition to nuclear power (46). Soci-ologist Dorothy Nelkin argues that women’s dis-trust of nuclear power cannot be attributed to agreater aversion to risk in general. Instead, Nel-kin’s analysis of women’s magazines and the fem-inist press suggests that women’s opposition be-gins with the specific risk of cancer in the event

of a major radioactive release from a reactor. per-sonal value priorities, including some women’sview of themselves as nurturers or “caretakersof life, ” also lead them to oppose what they viewas a life-threatening technology (49).

These connections have helped to bring nu-clear power as an issue into the mainstream ofthe women’s movement. Women’s magazinesranging from Redbook to Ladies Home Journal

to Ms. have questioned nuclear safety, and thenational Young Women’s Christian Association(YWCA) took a public stand against nuclear pow-

er in 1979. The League of Women Voters has de-veloped a national policy favoring only limitedconstruction of new reactors, and the League’slocal affiliates have taken even stronger anti-



Ch. 8—Public Attitudes Toward Nuclear Power . 231

nuclear stands. In 1980, the National Organiza-tion for Women (NOW) recommended a resolu-tion opposing the “use of nuclear power in favorof safer energy methods” (49).

While changing value priorities appear to havecontributed to increasing public concern about

nuclear power over the past 20 years, those pri-

THE ROLE OF

Both the amount and type of news coveragehave played an important role in shaping publicattitudes toward nuclear power. As noted pre-viously, people tend to overestimate the prob-ability of certain hazards, including nuclearpowerplant accidents, in part because thesehazards are discussed frequently in the media.

The Extent of Media Coverageof Nuclear Power

The most detailed analysis of print media cov-erage of nuclear power currently available isbased on the number of articles on the subjectindexed in the yearly Readers’ Guide to Period- ical Literature. In this study, sociologist Al IanMazur compared trends in media attention withtrends in public opinion as revealed by nationalopinion surveys, numbers of plant interventions,and size of protests. This analysis suggested the

following three hypotheses (41):1. The greater the national concern over a ma-

jor issue that is complementary to a partic-ular protest movement, the more easily re-sources can be mobilized for the movement,and therefore the greater the activity ofprotesters.

2. As the activity of protesters increases, massmedia coverage of the controversy increases.

3. As mass media coverage of the controversyincreases, the general public’s oppositionto the technology increases.

At the time of the first citizen interventionagainst a nuclear plant in 1956, there was a greatdeal of positive mass media coverage of presidentEisenhower’s “Atoms for Peace” program. In the

orities could change again in the future. In addi-tion to values, knowledge about accidents at nu-clear plants has played a major role in shapingpublic attitudes. In the future, new informationon improved management of nuclear powercould lead to a reversal of the current trend ofincreasing opposition to the technology.

THE MEDIA

early 1960’s, most coverage was still positive, buta few protests against local plants—particularlythe large demonstration at a proposed nuclearplant site on Bodega Bay, Calif., in 1963–re-ceived national publicity. During the mid-1 960’s,there was a decrease in both the number of peri-odical articles on nuclear power and in public

opposition as measured in opinion polls. This de-cline reflected a shift in public concern and mediaattention away from nuclear issues and towardcivil rights and other domestic issues. Beginningin 1968, magazine articles on nuclear power in-creased to cover local plant siting disputes. Printmedia coverage rose even higher in 1969, andopinion polls showed a similar peak of opposi-tion the following year.

From 1974 to 1976, anti-nuclear activism andmedia coverage again increased, with a great dealof national publicity given to the 1976 California

referendum. After 1976, both negative publicopinion and media coverage fell off, then roseslightly in 1978 and early 1979 and finally rosemassively following the accident at TMI in thespring of 1979. Trends throughout 1979 appearedto confirm the linkage between media coverageand public opinion: Public opposition rose sharp-ly immediately following the accident, subsidedwithin 2 months as media attention diminished,and then increased slightly during October andNovember, coinciding with media coverage ofthe final Kemeny Commission report (41).

Mazur argues that opposition to a technology

such as nuclear power will snowball with in-creased media coverage, whether that coverageis positive or negative (40). The fluoridation con-troversy of the 1950’s and 1960’s, like the cur-



rent nuclear power debate, involved complexscientific judgments and pitted the “establishedorder” against advocates of local self-control.During this period of public debate, persons ex-posed to both positive and negative argumentsabout fluoridation were more likely to oppose thepractice than persons who had heard neither ar-gument, and communities where there had beenheated debate were most likely to defeat fluorida-tion in a referendum. The prominence given todisputes between technical experts over the risksof a technology appears to create uncertainty inpeople’s minds, which in turn raises concern and

opposition, regardless of the facts under discus-sion. If this is true, the media play a key role inencouraging public opposition by giving exten-sive coverage to the experts’ disputes.

Analysis of the extent of television news

coverage of nuclear power has been much morelimited than analysis of print media coverage.Television nightly newscasts made relatively lit-tle mention of nuclear power over the decadepreceding the accident at TMI. Within the overalllow level of reporting, the trends were somewhatsimilar to those in the print media: Coverage in-creased in 1970, and then dropped off again un-til 1976, with greater coverage between 1976 and1979 (70).

The Content of Media Coverage

Just as there can be little doubt that mediacoverage influences public opinions toward nu-clear power, there also is little doubt that jour-



Ch. 8—Public Attitudes Toward Nuclear Power Ž 233

nal ists, l ike most Americans, are ambivalent

about this technology. In a 1980 survey, similarpercentages of media personnel and the public(about 55 percent) viewed the benefits of nuclearpower as greater than the risks, while other “lead-ership groups” were much more supportive ofnuclear energy (59). In another study, attitudes

toward nuclear power were measured on a scaleranging from -9 to +9, with a higher score in -

dica ting g reater supp ort for nuclear power. Whilescientists were quite supportive of the technologywith an average score of 3.34, science journalistswere much more skeptical, with an average scoreof 1.30, and journalists reporting on general issuesfor major national newspapers were slightly lesssupportive of nuclear power than science journal-ists, with an average score of 1.16 (62).

Following the accident at TMI, the KemenyCommission found that the public’s right to in-

formation had been poorly served. Confusionand uncertainty among the sources of informa-tion combined with a lack of technical under-standing by the media personnel were identifiedas contributing to the problem. Many of the re-porters “did not have sufficient scientific andtechnical background to understand thoroughlywhat they heard.” As a result of these difficultiesin reporting on emergencies, the commission rec-ommended that all major media outlets hire andtrain nuclear energy specialists and that reporterseducate themselves about the uncertainties andprobabilities expressed by various sources of in-

formation (31).

The me dia ’s need for balance in coverage ofmany issues, including nuclear power may leadto understatement of the scientific consensus thatthe technology is acceptably safe. Media person-nel are expected to bring various viewpoints be-fore the public, and in the case of a controver-sial technology such as nuclear power, this gen-erally means quoting both an advocate and a crit-ic in any given story. One analysis of televisionnews coverage showed that over the decade priorto Three Mile Island, most news stories dealing

with nuclear power began and ended with “neu-tral” statements (70). However, among the “out-side experts” appearing most frequently in thestories, 7 out of 10 were critics of nuclear power.Thus, while meeting the requirement of present-

ing opposing views, these stories may have over-simplified complex issues and failed to conveythe prevailing consensus among scientists andenergy experts. Psychiatrist Robert DuPont, afterviewing the same 10 years of television storiesused in this analysis, suggested that fear, especial-ly of nuclear accidents, was the underlying motif

in all of the stories (16). Another study of 6 yearsof television news stories about various energysources found that the risks and problems of nu-clear power were emphasized, coal was givenneutral treatment, and solar power was treatedeuphorically (56).

While these studies suggest that television cov-erage of nuclear power emphasizes the risks ofthe technology, there is no evidence that mediapersonnel deliberately bias their coverage ofnuclear power due to personal convictions. TheKemeny Commission found that overall coverage

of the TMI accident was balanced although attimes confused and inaccurate. One of the big-gest factors in inaccurate reporting at TMI wasfound to be the lack of reliable information avail-able to the media. For example, national reportsthat the hydrogen bubble inside the reactor couldexplode within 2 days were an accurate reflec-tion of the views of NRC’s Washington office.Reporters, trusting these views and wanting to“scoop” other reporters, tended to disregard theonsite NRC officials who argued that the bubblecould not possibly explode. However, overall,

the Commission found a larger proportion of re-

assuring than alarming statements in both televi-sion and newspaper reporting of the accident.

Media coverage of nuclear power maybe influ-enced by the fact that journalists are trained tobe skeptical of news sources, including the nu-clear establishment. Informed critics have beensuccessful in publicizing many cases in which thenuclear industry, the Department of Energy, andthe NRC have not been completely open aboutsafety problems. For example, during the first 2days of the accident at Three Mile Island, Metro-politan Edison withheld information on the situa-

tion from State and Federal officials as well as thenews media (72). According to the Kemeny Com-mission, the utility’s handling of information dur-ing this period “resulted in the loss of its credibili-ty as an information source” (31). Experiences




such as this have led reporters to be particularlyskeptical of nuclear industry sources and look tothe critics for the other side of any given story.

Proponents of nuclear power are likely to viewmedia treatment of nuclear plant safety issues asbiased because of the inherent complexity of

those issues. It is important that problems suchas construction errors, skyrocketing costs, andoperating difficulties be reported to the public.However, since few people (including reporters)understand nuclear technology well, problemsmay appear more threatening than they actuallyare. Considerable expertise is needed to sift thefacts and accurately interpret them to the public.By comparison, the media are not consideredanti-airplane, even though most coverage of that

industry focuses on crashes. Because the publicis unlikely to view a single plane crash as an in-dication that the entire airline industry is unsafe,the airline industry is confident that all airplaneswill not be grounded. With no such assurancesfor nuclear power, the nuclear industry may viewcoverage of accidents as a threat to its survival.

Finally, it is important to note that journalistsdid not create the nuclear controversy. Duringperiods of greatest public concern, their cover-age of nuclear power has increased, which in turnhas contributed to still greater public uncertain-ty. If the media are more critical of nuclear powernow than they were in the 1950’s, they may bereflecting public opinion as well as influencing it.

WHAT WOULD IT TAKE TO INCREASEPUBLIC ACCEPTANCE OF NUCLEAR POWER?

It is unlikely that utility executives will orderany new reactors as long as they believe that amajority of their customers oppose nuclearenergy. However, a societal consensus on thenecessity for and benefits of the technology maybe very difficult or impossible to attain. The pre-vious analysis indicated that the general publicand the staffs of some public interest groups areconcerned about the possibility of a catastrophicreactor accident. They perceive that the technol-ogy offers few or no benefits compared to theserisks. In addition, many Americans’ personal val-ues contribute to their skepticism of the tech-nology and its managers. These value conflictsmay prevent a total resolution of the current con-troversy. However, attitudes might change eitheras a result of external events (e.g., another oil em-bargo or new research findings on the environ-mental impacts of coal burning) or because of im-provements made internally by government andthe nuclear industry. External events cannot becontrolled, but it is up to the nuclear establish-

ment to demonstrate the safety and economic at-tractiveness of nuclear power.

Assuming that major improvements were madein management of nuclear power, it would stillbe difficult to communicate them to the public

because of the present lack of trust in governmentand industry. There are some extremists on bothsides of the nuclear controversy whose opinionswill not change, regardless of the evidence placedbefore them. Even more moderate citizens, whoare willing to change their opinion on the basisof new evidence, are influenced strongly by pre-existing attitudes and values so that they may“filter out” or wrongly interpret new evidence.Finally, for the majority of the public, new infor-mation on improvements in utility managementof nuclear power will be viewed skeptically unlesspresented in a manner that arouses trust and in-terest. However, while better communicationsare needed, the first and most important step isto make concrete improvements responding topublic concerns.

Enhance Nuclear Advantages

Research conducted in the United States andthe Netherlands suggests that people’s judgment

of a technology or activity is influenced as muchby their assessment of its potential benefits as bytheir view of its risks. There are at least threepotential benefits of nuclear power that could beperceived by the public: 1) its contribution to na-





Protectors’ tent at a demonstration against the Seabrook nuclear plant in May 1977

tional energy and electricity supply, 2) its poten-

tial cost advantage relative to other energy op-tions, and 3) the fact that safely operated nuclearplants produce no fossil air pollution.

It is difficult for many Americans to see a needfor nuclear powerplants at a time when electricitydemand has slowed. While this slow growth isexpected to continue over the next several years,new powerplants of some kind still will beneeded in the years ahead. Regions experienc-ing rapid economic and population growth willneed new capacity sooner than others. Planscould be developed at a regional level to evaluate

the alternatives to meet demand growth. Theplanning process itself could become a vehiclefor public participation, and any long-term costadvantages of nuclear power could be most clear-ly demonstrated to the public this way.

Under some conditions, nuclear electricity can

be cheaper than its major competitor: electrici-ty from coal combustion. Standardized plant de-signs, increased predictability in the licensingprocess, and improved management of operatingreactors all could help to realize the technology’seconomic potential. New rate regulation systemsalso could be used to reduce the initial costs ofnew nuclear and coal powerplants to the con-sumer. Assuming all of these changes took placeand nuclear electricity did indeed offer long-termcost advantages, public opposition to new plantsin hearings before State PUCs very likely wouldbe reduced.

However, coal is not the only alternative tonuclear power in meeting national energy needs.Conservation, oil shale, and renewable energyresources all can be used to match energy sup-




ply and demand but widespread application ofthese technologies could be expensive. In Maine,public rejection of a 1982 referendum to shutdown Maine Yankee was based in part on recog-nition that conservation and renewable energycould not quickly make up for the inexpensivenuclear power lost by the shutdown (see CaseStudies).

Paradoxically, accelerated R&D on these alter-natives might enhance the image of nuclear pow-er and could confirm that they may never bewidely competitive. As part of accelerated R&Don alternative energy sources, the environmen-tal costs and benefits of each source should beexamined. Environmental groups currently areamong the leading critics of nuclear power. Thesegroups also are very concerned about the adverseimpacts of coal combustion and other energysources, and are monitoring research into thoseimpacts. If this research indicated that acid rainwas a more serious problem than presently per-ceived or that carbon dioxide buildup would re-sult in near-term climatic changes, some environ-mentalists might become less negative about nu-clear relative to coal. This shift, in turn, couldchange attitudes among the broader public.

Public relations or educational programs areunlikely to increase public awareness of nuclearpower’s potential benefits until those benefits areapparent. This might result either from events out-side the industry’s control which decrease avail-ability of alternative energy sources (e.g., an oildisruption) or from improvements in manage-ment of the technology. Without such actions,public relations programs such as the currentCommittee for Energy Awareness campaign mayhave little impact, and possibly even a detrimen-tal effect on public opinion. The response to thiscampaign from critics may increase public uncer-tainty and skepticism. Even programs viewed asunbiased by all sides, such as the League ofWomen Voters Education Fund’s (LWVEF) “Nu-clear Energy Education Program” carried out in

1980 and 1981, may do little to increase publicacceptance until the costs of new nuclear pow-erplants are better controlled (34). Nevertheless,the low level of public understanding of nucleartechnology does indicate a need for more infor-

mation, and a number of organizationsvolved in public awareness campaigns.

Reduce Concerns OverNuclear Accidents

are in-

While increased awareness of nuclear power’s

benefits might decrease concerns about risk, oneof the most favorable things that could happento the nuclear industry over the next 10 yearswould be an increasing output of nuclear elec-tricity along with an absence of events causingbad publicity. Presently, both TMI-type accidentsand incidents such as the failure of the safety con-trol system at the Salem, N. J., plant are viewedby the public as precursors to a catastrophe.Given the slow rate at which public support fornuclear power has declined, an extended periodof quiet, trouble-free operations could have verypositive impacts on public attitudes. Chapter 5

identifies a number of approaches to improvedutility management of nuclear power, which, ifimplemented, could help to assure that neithermajor accidents nor precursors take place.

While a period of uneventful operation of nu-clear plants is necessary to restore public con-fidence, it probably is not sufficient. MaineYankee has had very high reliability but Statevoters have twice come close to shutting it down(see Case Studies). In addition, critics probablywould remain skeptical. It would be importantto demonstrate to them that the period of quiet

operation was a result of real improvements andthe beginning of a new trend, rather than justluck. However, given the present level of distrustbetween interveners, the NRC, and the nuclearindustry, it might be very difficult to do this.

Several steps could be taken to improve com-munications between the nuclear communityand public interest groups critical of nuclearpower. An effort might be made to identify theconcerns of particular groups and respond to thesubstance of those concerns. For example, somegroups currently are concerned about insurance.

A compromise on this issue might not decreasethe groups’ fundamental criticism of nuclear safe-ty, but it could improve the climate and allowfurther negotiations to take place. If the currentheated debate could become a reasoned ongo-




ing dialog, the public might be less likely to view

the technology as unsafe. As discussed previous-

ly, the prominence and stridency of the debate

currently increases public uncertainty and en-

courages opposition.

As part of this effort, the Federal Government

could actively encourage involvement of respon-

sible interveners in both regulatory proceedingsand long-range planning efforts through funding

and other support. In Ontario, Canada, the inde-

pendent Porter Commission funded knowledge-

able nuclear critics to conduct studies and par-

ticipate in extensive hearings as part of its long-

term electricity planning. In the Commission’s

interim report, health problems caused by im-

proper disposal of uranium mine tail ings were

identi f ied, and environmental groups were ac-

knowledged for bringing the issue to the public’sattention. Similarly, based on testimony fromleading critics, the commission found that theprobability of a loss of coolant accident causing ameltdown at Ontario Hydro’s heavy water reac-tors was much greater than the Canadian nuclearindustry had claimed. Because the Commissionnot only sought critics’ concerns but also ac-knowledged and responded to them, the processhad the effect of moderating some groups’ anti-nuclear positions (see vol. II).

Previous U.S. efforts to involve government, in-

dustry, and environmentalists in dialog or “en-vironmental mediation” provide another modelfor improved communication. Nonprofit organi-

zations such as the Conservation Foundation inWashington, D. C., as well as several private firmshave brought all three parties together to discusstopics such as radioactive waste disposal andchemical waste management. By careful staffpreparation and beginning the discussions witha common objective (e.g., safe disposal of toxicwastes), these forums have succeeded in devel-oping preliminary agreements on Federal and in-dustry policy.

Nuclear regulators and the industry can in-crease their credibility with both interveners and

the public by emphasizing candor in their publicinformation programs. Prior to the accident atThree Mile Island, the nuclear establishmentcreated the impression that such an accident was

so unlikely as to be “impossible.” As a result,when the accident did happen, it greatly reducedthe credibility of the regulators and the industry.The nuclear establishment should acknowledgethat both operating events and more serious ac-cidents can occur, attempt to educate the publicabout the difference between the two, and dem-

onstrate its preparedness to deal with accidents.For example, TVA immediately reports to themedia any event that could be considered news-worthy. This very open approach increases theutility’s credibility with both the media and thepublic. Another positive example is offered bya Midwestern utility that encountered quality-as-surance problems during construction. Once thecompany had greatly increased its constructionmanagement capabilities, it launched a public re-lations effort to educate the public about theproblems and the steps it had taken to overcomethem. These efforts appear to have increased

local trust in the utility. (See Case Studies.)

Two approaches to siting policy might helpalleviate the public’s safety concerns. Both re-spond to the public’s opposition to constructionof new plants near where they live. First, asdiscussed earlier, some people living near nuclearplants tend to view them as less risky than peo-ple who are less familiar with the technology.While some polls show increasing opposition tonearby plants since the accident at Three MileIsland, support for other plants has remainedhigh. This fact has led Alvin Weinberg and others

to promote a “confined siting” policy, underwhich most new reactors would be added to ex-isting sites, rather than creating new sites. Thisapproach has been used successfully in Canada(see vol. Ii) and is supported by some U.S. en-vironmental groups. It is most attractive in theEast, where high population density makes re-mote siting infeasible.

The second approach to dealing with local op-position to new construction is to site new reac-tors at remote locations. This approach could in-corporate “confined siting, ” with new reactors

clustered at existing remote sites. Alternatively,new sites in remote areas could be identified. Ineither case, public opposition to such plantscould be expected to be much less than opposi-




promise might seem to offer little future fornuclear power, it did allow construction of sixnew nuclear plants, with the result that over halfthe country’s electricity is now nuclear. In theUnited States, a compromise under which regu-lators and nuclear critics agreed to encouragecompletion and operation of units currently

under construction or planned might be prefer-able to the current impasse, especially in termsof financial return to investors. It has been sug-gested that Americans might reach consensus ona 150-gigawatt nuclear program (29). Similarly,the advocacy arm of the League of Women Vot-ers has adopted a national policy calling for a con-tinuation of nuclear power in its current percent-age of national energy supply. As energy andelectricity demand grow in the future, this policywould allow some growth in the nuclear pro-gram. Any such compromise or cap would haveto allow for adjustments as nuclear and com-peting technologies are improved and economicschange. In addition, regional differences in theUnited States might make a State-by-State ap-proach more feasible than a national referendumas in Sweden.

While all of the approaches discussed abovemight decrease public concerns about currentreactors, it is possible that public skepticism aboutthe technology is so great that these changeswould have little impact. In this case, other reac-tor concepts with inherent safety features mightbe considered. Several alternatives, such as the

high temperature gas-cooled reactor (HTGR), theheavy water reactor (HWR), and an improvedlight water reactor (the PIUS reactor) are dis-cussed in chapter 4. A substantial federallyfunded R&D effort on one or more of these alter-natives might meet with public acceptance, par-ticularly if demand for power picks up over thenext two decades. The inherent safety featuresof the chosen design might appeal to the generalpublic and the choice of design could be usedas a vehicle for much greater involvement of nu-clear critics. By bringing critics into the R&D pro-gram and addressing their specific concerns, con-sensus might be reached on an acceptable designfor future reactors.

Minimize Linkage BetweenNuclear Power and Weapons

Another issue that should be addressed in pol-icy decisions about nuclear power is the connec-tion between weapons development and civiliannuclear energy. Given the level of national con-

cern over the arms race, public acceptance ofnuclear power cannot be expected to increasesubstantially until the two nuclear technologiesare separated in people’s minds. This report hasnot analyzed the impact of policies that mightminimize the linkages between nuclear weaponsand power, but the effect on public opinion couldbe positive. For example, one action that mightincrease public acceptance by reducing the per-ceived linkages would be to remove nuclearweapons development from the jurisdiction ofthe Department of Energy. Another step wouldbe to legislate a ban on commercial fuel reproc-

essing. Many critics are more concerned aboutreprocessing than about reactors because plutoni-um separated from the spent fuel might be stolenand used to construct a bomb or to threaten thepublic. A legislated moratorium on reprocessingmight have greater impact on these concernsthan the executive orders imposed by PresidentsFord and Carter that were later revoked by Presi-dent Reagan. Such a ban might be especially ef-fective if imposed in conjunction with limits onthe total growth of the program, as discussedabove. In addition, it might be best to keep in-

dustry and military waste disposal strictlyseparate, although some public interest groupssupport joint disposal, because it encourages ac-tion on military waste that has been allowed toaccumulate for 40 years (53).

Policy makers also could take action to reducethe possibility of weapons proliferation throughcareful management of international nuclearpower development. A previous OTA analysisidentified weaknesses in the existing internationalnonproliferation regime (52). Recognizing the im-pact these weaknesses have on public percep-tions of nuclear power, Alvin Weinberg hasargued, “We must strengthen the Nonprolifera-tion Treaty (NPT) regime and take the next steps,



240 q Nuclear Power in an Age of l.uncertainty

which involve both a reduction in nuclear arma-ments and a strengthening of the sanctions thatcan be imposed on those who would violate theNPT” (75).

The link between nuclear weapons and nuclearpower also might be reduced in people’s mindsif more proponents of nuclear power who op-pose the continued buildup of nuclear weaponsstated their beliefs publicly. For example, HansBethe, a prominent nuclear physicist who hasbeen active in the arms control movement and

supportive of civilian nuclear power, reaches anaudience who might otherwise reject nuclearpower along with weapons (48).

In conclusion, current public attitudes towardnuclear power pose complex problems for thenuclear industry and policy makers. However,technical and institutional steps could be takenthat might lead the public to view nuclear poweras an important and attractive energy source inthe years ahead. Constructive leadership andimagination will be required to start this process.



[page omitted]This page was originally printed on a dark gray background.

The scanned version of the page was almost entirely black and not usable.




CHAPTER 8

1. “A Fresh Portrait o f the A nti-Nuclea r Move me nt,”

Groundswe/ / , vol . 5, No. 4, July/ Aug ust 1982.2. Aldrich, David, et al., Sandia National Labora-

tories, Technical Guidance for Siting Criteria De-velopment, NUREG/CR-2239 (Washington, D. C.:

U.S. Nuclear Regulatory Commission, December,1982).

3. Alexanderson, E. P., Fermi-7: New Age for Nuc/earPower (La Grange Park, Ill.: The A merican Nuclea r

Society, 1979).

4. Anthony, Richard, “ Trends in Public Op inion on

the Environment, ” Environment, vol. 24, No. 4,May 1982.

5. Atomic Energy Commission, Theoretic/ Possibil-ities and Consequences of Major Accidents in Large Nuclear Power Plants, WASH-740 (Wash-ington, D. C,: U.S. Atomic Energy Commission,1 957).

6. Barbash, Fred, and Benjamin, Milton R., “States

Can Curb A-Plants,” Washington Post, Apr. 21,1983.

7. Beardsley & Haslacher, Inc., Nuc/ear Attitudes After the Three Mi/e /s/and Accident (Portland,Oreg.: Pacific Gas & Electric, May 1979).

8. Buss, David M., et a l., “Perc ep tions of Tec hno -logic al Risks and Their Manag ement,” pap er pre-

sented at the Sym po sium o n Environmenta l Haz-

ard and People at Risk, 20th International Con-gress of Ap plied Psyc holo gy , Ed inburg h, July 26,1982.

9. Cambridge Reports, Inc., American Attitudes Toward Energy Issues and the Electric Utility in-dustry (Washington, D. C.: Edison Electric Institute,1983).

10. Cambridge Reports, Inc., 1982 Polls conducted for

the U.S. Committee for Energy Awareness (unpub-lished).

11. Comey, David, “ The Inc ide nt a t Brown’ s Ferry,”Not Man Apart, September 1975.

12. Committee on Nuclear and Alternative Energy Sys-

tems, Nationa l Research Counc il,Energy in Tran- s i t i o n 1985-2070 (Washington, D. C.: National

Ac ad em y of Sc ienc es, 1979).13. Delco igne, Georges, “Educ ation and Public Ac -

ceptance of Nuclear Power Plants, ” Nuc/ear Safe- ty, vol. 20, No. 6, November/December 1979.

14. “ Despite Nuc lea r Pow er’s Prob lem s, The Pub licis Convinced,” Nucleonics Week, Ma y 5, 1983.

15. Dona ldson, Thom as, “ Ethica lly, ‘Soc iety Expe c ts

More From a Corporation’, ” U.S. News & Wor/d

Report, Sept. 6, 1982.

REFERENCES

16.

17.

18.

19.

20

21.

22.

23.

24.

25 .

26 ,

27.

28,

29.

30 .

DuPont, Rob ert, “Nuc lea r Phob ia–Phobic Think-

ing About Nuclear Power” (Washington, D. C.:The Media Institute, 1980).DuPont, Robert, “ The Nucle ar Power Phob ia,”Business Week, September 1981.

“Energy-guzzling: Most Consumers Are Cured,”Harris poll reported in Business Week, Apr. 4,

1983,

“Expert Knowledge,” Public Opinion, vol. 4., No.6, October/November 1981.

Firebaugh, Morris, “Public Educa tion and Atti-tudes,” An Acceptable Future Nuclear Energy Sys- tem, M. W. Firebaugh and M. J. Ohanian (eds.)

(Oak Ridge, Term .: Institute for Energy Analysis,1980).Fischhoff, Baruch, et al., “Lay Foibles and ExpertFables in Judgments About Risk,” Progress in Re-

source Management and Environmental Planning,VOI. 3, 1981.

Ford, Daniel, “ The Cult of the Atom -l, ” The New Yorker, Oct. 25, 1982.Ford, Daniel, “ The Cult of the Atom -l I,” The New

Yorker, Nov. 1, 1982.Freudenburg, William, and Baxter, Rodney, “Pub-lic Attitud es Toward Loc al Nuc lea r Power Plants:A Reassessment,” paper presented at the 1983 an-

nual meetings of the American Sociological Asso-

ciation, Detroit, Aug. 31 -Sept. 4, 1983.

Fuller, John G., We A/most Lost Detroit (New York: Reader’s Digest Press, 1975).Gomberg , H. J., et al., “ Rep ort on the Possible Ef-fects on the Surrounding Population of an As-

sumed Release of Fission Products From a300-MW Nuclear Reac tor Loc ated at Lag oonaBeach,” APDA-120 (Atomic Power DevelopmentAssoc iate s, 1957).inglehart, Rona ld, “ Post-Materialism in an En-vironment of Insecurity, ” American Po/itica/Science Review, vol. 75, No. 4, December 1981.Institute of Nuclear Power Operations, Review of NRC Report: “Precursors to Potential Severe Core Damage Accidents: 1969-1979: A Status Repott, ” NUREG/CR-2497 (Atlanta, Ga.: INPO, 1982), p,

21.

Kasperson, J. X., et ai. , “Institutional Responsesto Three Mile Island , ” Bu//etin of the Atomic Sci- entists, vol. 35, No. 10, December 1979.Kasperson, Roger, et al., “Public Opposition to

Nuclear Energy: Retrospect and Prospect,”Science, Technology, and Human Values, vol. 5,No. 31, spring 1980.

25-450 0 - 84 - 17 : QL 3




66.

67.

68,

69.

70 .

71 ,

to r Accident,” F/uc/eus, vol. 4, No. 4, winter 1983,p. 4.

Slovic, Paul, et al., “Ima ges of Disaster: Percep -tion and Acceptance of Risks From NuclearPower,” Proceedings of the 15th Annual Meeting,National Council on Radiation Protection and Measurements: Perceptions of Risk (Bethesda,

Md.: NCRP, 1980).Slovic, Paul, et al., “ Perceived Risk and Quanti-

tative Safety Goa ls for Nuclea r Powe r, ” Transac- tions of the American Nuc/ear Society, No. 35,1980.

Slovic, Pau l, et al., “ Ra ting the Risks: The Struc-ture of Expert and Lay Perceptions,” Environment,vol. 21, No. 3, April 1979.“Spectral Lines: Opinions on Nuclear Power, ”IEEE Spectrum, vol. 16, No. 11, November 1979.

Theb erge , Leonard, Television Evening News Looks at Nuc/ear Energy (Washing ton, D. C.: The

Media Institute, 1981).

U.S. Congress, House of Representatives, Com-

mittee on Interior and Insular Affairs, Reactor Safe- ty Study (Rasmussen Report) (Washington, D. C.:U.S. Government Printing Office, June 1976).

72. U.S. Congress, House of Representatives, Com-mittee on Interior and Insular Affairs, 97th Cong. ,

1st sess., Reporting of Information Concerning the Accident at Three Mile Island (Washington, D. C.:

U.S. Government Printing Office, 1981).73. Van Loon, Eric, “Executive Director’s Message, ”

Nuc/eus, vol. 4, No, 4, winter 1983.74 . Vlek, Charles, and Stallen, Pieter-Jan, “Judging

Risks and Benefits in the Small and in the Large,”

Organizational Behavior and Human Perform- ance, No, 28, 1981.

75. Weinberg, Alvin (cited in M. W. Firebaugh and

76

M. J. Ohanian (eds.), An Acceptable Futu;e Nu-c/ear Energy System (Oa k Ridg e, Term.: Institutefor Energy Analysis, 1980).

Weinberg, Alvin, “Nuclear Safety and Public Ac-

ceptance” Nuc/ear News, October 1982,



Chapter 9

Policy Options

Photo credit: Ontario Hydro

Pickering nuclear station has four identical CANDU units built in 1973 and four more scheduled for completion in1985. Standardization helped reduce construction costs and improve operator training



Contents

Page

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..., . . . . . . . , , . ..................251

Polic y Goa ls and op tions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ...252

Goal A: Reduce Capital Costs and Uncertainties . . . . . ..........................252

Goal B: Improve Reactor Operat ions and Economics . . . . . . . . . . . . . . . . . . . . . . . , . . .255Goa l C: Red uc e the Risk o f Ac c idents That Have Public Safetyo r

Utility Financ ial Im pa c ts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .................256Goal D: Alleviate Public Concerns and Reduce Political Risks . . . . . . . . . . . . ........260

Major Federal Policy Strategies and the Likelihood of More Orders for Nuclear Plants. .263Ba se Ca se: No C ha ng e in Fed era l Polic ies . . . . . . . . . . . . . . ......................263Base Case Variation: Let Nuclear Power Compete in a Free Market . ....,.........270

Strategy One: Remove Obstacles to More Nuclear Orders.. . ....................271Strateg y Two : Provide Mode rate Stimula tion to More Orde rs . . . . . . . . . . ...........273Strategy Two Variatio n: Sup po rt the U.S. Nuc lea r lndustry in Future World Trad e ...275

Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .277

Chapter 9 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..............,...279

Tables

Table No. Page

34. Sum ma ry o f Polic y Options . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .................25235. Ma jor Policy Strate g ies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .....................264

36. Four Scenarios Affecting the Future of the Nuclear Industry . . . . . . . . . . . . . . . . ....26537. Four Com bina tions of Ec ono mic and Mana ge me nt Sc ena rios Unde r

the Base Case.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ...............267

38. Alternative Nuclear Futures Under Strategy One: Remove Obstacles

to More Nuclear Orders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ........27239. Alternative Nuc lea r Futures Und er Strategy Two : Provid e a Modera te Stimulus

to More Nuc lea r Orde rs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..........273



Chapter 9

Policy Options

INTRODUCTION

The previous chapters have painted an un-promising picture of the future of nuclear power

in the United States. Projections for new centralstation generating capacity over the next 20 yearsare much lower than those of just a few years ago.The high financial and political risks involved withnuclear plants suggest that any central stationcapacity that is added would be coal-fired. Underexisting conditions, there are few incentives forutilities to select nuclear plants and many reasonsto avoid them.

It can be highly misleading, however, to fore-cast future decisions on the basis of existing con-ditions. Some of the problems that appear so for-

midable now will diminish. The plants under con-struction now were designed according to con-cepts developed 10 to 15 years ago. Any futureplants can be expected to incorporate majorchanges that have been backfitted onto existingplants and other changes that have been sug-gested to improve operation. In addition, muchhas been learned about how to construct plantsmore efficiently, While these and other changeswould go far toward eliminating the large costoverruns some utilities have experienced, theyprobably are not sufficient to restore confidencein the financial viability of the technology. Other

concerns exist that these changes will not ad-dress. Therefore, it is probable that additional ini-tiatives, including Federal actions, will be re-quired if the country concludes that nuclear en-ergy is to continue to grow past the plants nowordered.

As recounted in chapters 1 and 3, there are rea-sons why the Nation could decide that it wouldbe in the national interest to maintain a domesticnuclear option. Nonfossil fuel energy sourcesmay be urgently required for environmental rea-sons within several decades, and nuclear energycould be the most economical source that canbe readily deployed. Even if such environmen-tal conditions do not materialize, it could beeconomically prudent to retain a generating

source other than coal. The energy outlook forthe early 21st century, when oil and gas reserves

will become seriously depleted, is very uncertain.If it is reasonably possible that nuclear power willbe seen as very desirable or even indispensablewithin 20 or 30 years, it probably is more effi-cient to have a continuous learning curve thanto try to put the industry back together when itis needed.

There are, of course, reasons for opposing thesearguments. Even if it is conceded that it wouldbe in the national interest to have the nuclear op-tion, that does not mean it is the responsibilityof the Federal Government to ensure it. The eco-

nomic penalty for not having more nuclear plantswould not be crippling (though the total dollarpenalty could be quite high) (2,4), and as shownin chapters 3 and 5, unless nuclear plants are builtand operated well, they are not the most eco-nomic choices. If any more serious accidents oc-cur, forcing long shutdowns and expensive back-fits, the economics of nuclear power will be veryhard to defend. Thus, it could be more produc-tive economically to concentrate on the develop-ment of alternative energy sources.

Therefore, policy options presented here arenot intended to prop up a terminally ill industry,

but to cure the problems of an industry that issalvageable and which the Nation decided wasneeded. I n addition, some of the options can be

of importance for ensuring that existing reactors

operate safely and economically regardless of the

choice about the industry’s future.

The next section presents a series of policygoals and options that might be considered byCongress. For some of these, a lead congressionalrole would be needed. For others, congressionalaction may be no more than general policy set-ting and oversight because the main initiative

must arise elsewhere.One of the difficulties facing policy makers is

that few if any of these options will be very effec-

251




tive by themselves. Actions need to be taken ona broad front, but the responsibility for these ac-tions is diffuse.

The third section, therefore, groups the optionsaccording to three different strategies: first, nochange in Government policy as it currently ap-

pears; second, remove obstacles in the way offurther orders for nuclear plants; third, stimulatemore nuclear orders. These strategies correspond

to different levels of involvement to which pol-icymakers may want to commit the Government.

The success of these strategies depends in turnon two other factors: the need for nuclear powerand how well the industry manages its presentreactors. Therefore this section also includes

economic and industry management scenariosthat are combined into four different futures tohelp evaluate the strategies.

POLICY GOALS AND OPTIONS

In order for nuclear power to become moreacceptable in general, progress must be made inseveral different areas. Reactors must be moreaffordable, operations of existing nuclear plantsmust be improved, concerns over potential ac-cidents must be alleviated, and public acceptance

must be improved. This section discusses the spe-cific policy initiatives that would contribute tothese goals. The goals and options are listed intable 34. Under each goal the options are listednot by importance, but in order of ease of imple-mentation according to the strategies discussedlater.

Goal A: Reduce Capital Costsand Uncertainties

At present, nuclear reactors pose too great afinancial risk for most utilities to undertake. Fewutilities can support such a great capital cost for

the length of time required to build a nuclearplant, even if Iifecycle cost projections show thatit would be the cheapest power source over thelifetime of a powerplant. Not only are capital costestimates high, but the actual cost could be muchhigher if designs continue to change during con-struction. As discussed in chapters 3, 4, and 5,

Table 34.—Summary of Policy Options

Strategya Congressional roleA.

B.

c .

D.

Reduce capital costs and uncertainties1. Revise the regulatory process for predictable licensing . . . . . . .

2. Develop a standardized, optimized design . . . . . . . . . . . . . . . . . . .

3. Promote the revision of rate regulation . . . . . . . . . . . . . . . . . . . . .

Improve reactor operations and economics1. R&D programs to improve economics of operations . . . . . . . . . .2. Improve utility management of nuclear operations . . . . . . . . . . .3. Resolve occupational exposure liability . . . . . . . . . . . . . . . . . . . . .

Reduce the risk of accidents that have public safety orutility financial impacts1. Improve confidence in safety2. Certify utilities and contractors . . . . . . . . . . . . . . . . . . . . . . . . . . . .3. Develop alternative reactors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .4 Revise institutional management of nuclear operations. . . . . . . .

Alleviate public concerns and reduce political risks1. Accelerate studies of alternative energy sources . . . . . . . . . . . .2. Address the concerns of the critics . . . . . . . . . . . . . . . . . . . . . . . .

3. Control the rate of nuclear construction ., . . . . . . . . . . . . . . . . . .4. Maintain nonproliferation policies . . . . . . . . . . . . . . . . . . . . . . . . . .

One

OneTwoTwo

Base CaseOneTwo

Base CaseTwoTwoTwo

Base CaseOne

OneTwoTwo5. Promote regional planning for electric growth . . . . . . . . . . . . . . .

Oversight, legislation

Moderate R&D funding (design)Major R&D funding (demonstration)Inquiry; FERC regulation

Minor R&D fundingNRC oversightLegislation

NRC oversight; minor R&D fundingLegislationMajor R&D fundingInquiry, oversight

Minor R&D fundingOversight, legislation

LegislationOversight of legislationLegislation

astrategies incorporating these policy options are described later in the chapter: Base Case, Strate9Y One, and Strate9Y TWO





A2. Develop a Standardized LWR Design, Opti- mized for Safety, Reliability, and Economy

For a variety of reasons discussed in chapters4, 5, and 6, the reactor plant designs currentlyavailable could be significantly improved. An ef-fort that rethinks the concepts by which reactorshave been designed could result in light waterreactors (LWRs) that are cheaper, safer, moreoperable, and perhaps smaller than the presentgeneration. This effort would draw on all that hasbeen learned about the characteristics of good,safe reactors and integrate the best features intoa package that would represent the best of tech-nology. The design philosophy would emphasizeresiliency and passive safety features as well asaffordability and economy. The system would besubjected to intensive analysis from every possi-ble perspective to ensure that, insofar as possi-ble, all contingencies had been covered.

To a degree, the Westinghouse effort on theadvanced pressurized water reactor described inchapter 4 meets these objectives. The rationalefor a Government role is that a complete reac-tor and plant design is extremely expensive, andno corporation is likely to be able to finance itunless it sees a major market, which is not nowthe case. In addition, there are several technicalquestions such as the unresolved safety issues re-quiring additional R&D that is best funded by theFederal Government. A Government-initiatednonproprietary design could more easily draw onthe work of more than one vendor or architect-engineer as well as a coordinated R&D program,and be available to more producers. Therefore,a national design could have a better chance ofbeing truly optimized. The safety analysis alsomight be more convincing since it would be donein a more open atmosphere, with direct feedbackto the design to improve safety to the maximumextent possible. There is also a growing feelingthat current reactors have overshot the ideal size.U.S. vendors are unlikely to be in a position toredesign their new reactors to be smaller.

There are several advantages to a standardized

design. The cost would be much more predict-able, since most of the regulatory and construc-tion uncertainties could be cleared up beforeconstruction started. Costs also could be lowered

by incorporating improved construction tech-niques. It should be cheaper to operate becauseit would be designed to operate at a higher ca-pacity factor, lower fuel cycle costs and loweroperator exposure. Even if the technology weresimilar to present reactors, this new designpackage might represent a major improvement

in the acceptability of nuclear power.There are also disadvantages, however. It

would be at least as expensive for the Govern-ment to sponsor such a design as it would for acorporation—perhaps several hundred milliondollars if a demonstration were required. In ad-dition, a Government lead in developing a newdesign might imply to some groups a dissatisfac-tion with present designs serious enough that ex-isting reactors should be shut down.

A3 . Promote the Rev is ion o f Rate Regu la t ion

The process of rate regulation in most Stateswas designed for an era of relatively small capitalcost increments and declining costs per kilowatt-hour. High interest rates and high capital costsfor new generating capacity have strained the sys-tem so seriously that changes may have to bemade before utilities resume ordering new cen-tral station capacity. The current overcapacitygives utilities a welcome respite, but large con-struction programs will be needed once again.

Regulatory changes that should be consideredhere are: 1) rate base treatment of utility assets

that takes inflation into account, 2) some con-struction work in progress (CWIP) to be includedin the rate base, and 3) real rates of return onequity appropriate to the actual investment risk.These changes and others are discussed in detailin chapter 3. Their general intent is to even outrate increases and provide greater financial stabil-ity for utilities and their customers.

A difficulty for Federal policy in this area is thatrate regulation is the prerogative of the States. IfFederal action is to be acceptable, it must betaken in a way that makes it in the interest of theStates. Federal encouragement of long-range re-

gional planning and regulation (see option D5)may be useful since many States are finding thattheir regulatory programs are encountering in-creasing difficulties in forging satisfactory com-



Ch. 9—Policy Options q 255

promises. To some extent, regulation of whole-sale power sales by the Federal Energy RegulatoryCommission (FERC) influences State regulation.In the summer of 1983, there was extensive con-gressional debate on legislation restricting FERCallowance of CWIP. Consumer opposition toCWIP has been intense because it allows pay-ment for facilities before they are of any use tothe ratepayers. Some States, however, do havepartial CWIP allowances.

Goal B: Improve ReactorOperations and Economics

Decisions on the desirability of future reactorswill be based not just on capital cost projections(as improved under goal A) but also on the per-formance of existing reactors. The low reliabilityexperienced at some plants negates their poten-

tial economic benefits and raises concerns oversafety. Investors, critics, and the public will beopposed to more orders if some plants are notice-ably unreliable. Thus, it is in the interests not onlyof the specific utility involved but of the industryas a whole to improve operations at all plants.Other means for improving the economics of ex-isting reactors could also improve the outlook fornuclear power as a whole.

The specific options toward this goal follow.

B 1. Support R&D Programs to Improve the Eco-

nomics of Operation DOE and most of the nuclear steam supply sys-

tem (NSSS) vendors have modest programs fordeveloping extended burnup for fuel elements.There would be a national benefit from expan-sion of these programs. Fuel cycle costs could bereduced slightly, and the volume of spent fuel ac-cumulation would be decreased considerably(perhaps by 40 percent). This latter factor, byitself, could justify a significant Federal effort. Sav-ing 40 percent of the spent fuel would not reducethe spent fuel problem proportionately, but itwould ease the total burden considerably in thelong term. The objection generally voiced to aFederal program is that private industry couldhandle it. While this is probably true, a Federalrole would expedite matters and improve our in-

ternational competitive position. A long-termR&D program could provide further benefits.

B2. Improve Ut i l i ty Management of Nuclear Power Operations

None of the policy options discussed in this

chapter will do as much to improve the attrac-tiveness of nuclear power for all the parties to thedebate as improved utility management. Manyutilities were unprepared for the complexities ofnuclear power and the dedication required. Thissituation was perhaps unavoidable, given theoverenthusiasm gripping the nuclear supply in-dustry and the Federal nuclear promoters. Bynow, however, we are in a period of operation,not expansion. Utilities now have the primaryresponsibility, and all utilities responsible fornuclear plants must be adept at carrying it out.

it is important to recognize that much is beingdone to improve the quality of operations as dis-cussed in chapter 5. The Institute for NuclearPower Operations (IN PO) was set up for preciselythis purpose and has developed a large numberof specific programs. The NRC has shifted someof its scrutiny from the plants to the organizationsrunning them. It is not yet clear whether theseefforts will be sufficient.

Specific areas for attention are training and or-ganizational structure. Both previously had beenleft to the discretion of the utility but are now be-ing addressed by both the NRC and IN PO.

Requirements for training show a remarkablevariation. Good training programs are expensive,and qualified instructors are in limited supply. Itis important to set standards for training and es-tablish reasonable programs for achieving them.INPO is beginning to do this by establishing atraining accreditation program. It also may benecessary for NRC to impose these standards toachieve the optimum progress. The NRC prob-ably has the statutory authority to do this, but acongressional directive would expedite NRC ac-tions. Careful observation of the results of INPO’s

efforts is important to see whether additional NRCaction is necessary.

Criteria for organizational structure will beharder to define. One factor that is apparent,




however, is that the highest levels of the utilitymanagement must be involved with the plantsand committed to their good operation. Again,the NRC probably has the authority to commandattention at the utility headquarters, and is at-tempting to do so. Still greater resolve seems tobe in order at some utilities, however, and con-gressional encouragement of NRC to make thisa high priority item would help.

Both the NRC and INPO know which utilitiesare most in need of upgrading their management.All utilities are strongly influenced by the ex-periences of these few. Strong measures may berequired to get the operation of these plants upto minimally acceptable levels. Congressional ex-presson of the importance of a strong manage-ment commitment would be a significant incen-tive for the NRC and the utilities.

B3. Resolve the Financial Liability for Occupa- tional Exposure

The weapons testing program has focused at-tention on compensation for injuries arising fromexposure to radiation. New approaches are be-ing developed for compensating test participantsand downwind residents, and the industry is con-cerned that these plans will be applied arbitrari-ly to commercial nuclear plants (and possibly themedical industry). The proposals under consid-eration for the weapons tests plaintiffs link radia-tion exposure to the probability of contracting

cancer, and then award compensation based onthat probability. With this approach, claimantswho receive the most exposure also receive thegreatest rewards. Recent legislation in Congressproposes awarding $500,000 to a claimant if thereis at least a 50-percent chance that the cancerdeveloped from the radiation exposure. At lowerlevels of exposure that may only result in a 10-to 20-percent chance of cancer, the claimantcould receive $50,000. This proposal is controver-sial for two reasons. First, the nuclear industrycontests the relationship between low radiationdoses and cancer since there is insufficient scien-

tific or technical basis to support it. In addition,many claimants who were exposed to low doseswould receive compensation for cancers thatwere not produced by radiation but by other

causes. Critics would argue that excluding suchcases would deprive a large number of potentialvictims of just compensation.

Exposure levels during the weapons tests wereconsiderably higher than expected occupationalexposures at nuclear reactors. Some workers will,

over their lifetimes, nevertheless accumulate ahigh enough dosage to qualify for awards if thefloor is at the 10- to 20-percent level. Hospitalsalso may find themselves liable for the exposurefrom X-ray machines and nuclear medicine. Com-pensation for test victims is an important socialissue. It also is important to recognize that it hasimplications for the nuclear industry that couldbe serious if the awards are large.

Goal C: Reduce the Risk of AccidentsThat Have Public Safety or

Utility Financial ImpactsNuclear reactor safety is a function of the de-

sign of the plant, the standards by which it is built,and the care with which it is maintained and op-erated. If any of these are deficient, safety willbe compromised, perhaps seriously, and costsmay well escalate unexpectedly. Option A2 hasdiscussed how to improve the designs of the nextgeneration of LWRs, but this alone may not beadequate. It would not affect existing plants, andit may not go far enough in assuring safety in fu-ture plants. Without a consensus that nuclear re-

actors now are safe enough, there are unlikelyto be any more. Therefore, ways to improve thesafety of both present and future reactors are ex-plored under this goal.

The quality of the people involved appears tobeat least as important as the design of the plant.Option B2 discusses how to improve utility opera-tion, but again this may be inadequate by itself.Some utilities simply may be unable to improvetheir performance sufficiently. Others may thinkthey have done so but experience the same dif-ficulties in construction when they order another

plant. Two options discussed under this goal canbe considered if utility improvement is inade-quate. Construction permits and operating li-censes could be reserved for utilities and contrac-




tors that can demonstrate the commitment to

build and operate the plants to the exacting stand-

a rds requ i red . Sec ond , d i f fe ren t i nst itu ti ona l

arrangements might be considered to replace util-

i ty ma nag em ent o f reac tors, This op tion a lso

could be effective in stimulating further growth

of nuclear power if utilities are reluctant to order

more.

Cl. Improve Confidence in the Safety of Existing and Future Reactors

As discussed in the options above, there has

been a continual evolution in designs because

of frequent discoveries of inadequacies with re-

spect to safety or operation. As our understand-

ing of the technology has improved, formerly un-foreseen accident sequences or conditions arerecognized. Unquestionably, the technology ismaturing, but there is considerable dispute over

how much farther it has to or can go.Part of the problem has been the partitioned

nature of the safety analysis both in the industryand the NRC. Eac h system m ay b e thoroug hly

scrutinized, but the entire plant is not viewed as

a system, and responsibility for analyzing its over-all safety appears to be lacking.

No amount of analysis will uncover all poten-tial problems, but an intense analysis of eachplant could identify design or operating flawsbefore they caused problems. These studies areexpensive, but a few utilities already are under-

taking them in their own interests. The intent isto discover weak points in the design and developmeasures to address them, whether by changingplant equipment or modifying operations.

Other efforts to improve safety could focus onimproving the analytical techniques. As has beenstated above, probabilistic risk assessment is auseful tool that is still imprecise. Developmentof this technique would be beneficial for bothsafety and economics. This will involve mainlyimproving the data base for failure rates andanalyzing the human element, as is done in the

aircraft industry.The existence of unresolved safety issues, and

the probable introduction of more as new con-cerns are developed, undermines confidence in

safety. Resolving them expeditiously would elim-inate some safety concerns, demonstrate a com-mitment to maximum safety on the part of theNRC, and permit more stable cost projections forfuture plants. Resolution of some of the issuesmay call for modifications on existing plants.While the utilities would not welcome such ex-penditures, the overall reduction of uncertaintiesand the gains in safety would be useful.

C2. Certify Utilities and Contractors

It is readily apparent that some nuclear plantsare not being built and operated skillfully enough.As discussed earlier, all plants may be hostageto the weakest because an accident, or even poorperformance, reflects on all. If the persuasive ap-proach of option B2 is insufficiently effective inimproving nuclear plant management, more dras-

tic steps could be warranted.For existing reactors, the NRC evidently already

has the power to suspend an operating licenseif a utility is incapable of managing a reactor safe-ly. Few people expect the NRC to do this withoutthe most compelling evidence of incompetence.If higher standards are to be enforced, it prob-ably will only be with congressional legislation.Such improved standards would be in the bestinterests of the industry even though their imple-mentation could be traumatic. Even if this author-ity were never invoked, it could be a strong in-

centive to utilities to improve their performance.The result would be greater confidence in thesafety and operability of reactors.

Future reactors present a slightly different pic-ture. Utilities have Iearned that building reactorsis very difficult, and few, if any, would embarkon a new construction program unless they wereconfident they had the ability. Even then, how-ever, other parties of concern may not share thatconfidence. Certification of utilities as having thenecessary ability and commitment to build andoperate a reactor to high standards would ensure

that many of the expensive mistakes of the pastwere not repeated. This would reassure many ofthe critics of nuclear power as well as investors,utility commissions, and the public. It also mightbe necessary to eliminate from contention con-




tractors who had not demonstrated their capabili-ty of meeting the exacting standards required fornuclear construction. Presumably utilities wouldknow better than to select these contractors, butsome past experiences have been so poor thatmaking it official would increase confidence.

Even though this option is not likely to preventany plants from being built, it would be viewedby the industry as another set of regulations tomeet in what they consider to be an already over-regulated enterprise. The utilities also may resenthaving a Federal agency judge utility manage-ment quality. An independent peer body analo-gous to that being set up for review of medicareinpatient treatment might meet with better ac-ceptance.

There are no clear criteria as to what constitutesgood management concerning construction of anuclear powerplant. Nevertheless, as part of astrategy to rebuild confidence in the technology,this option clearly bears further examination.

C3. Develop Alternative Reactors

DOE has carried on a modest program for R&Don the high temperature gas-cooled reactor(HTGR). Given a higher commitment, the HTGRmight develop into a superior reactor. In par-ticular, it has inherent safety features that at leasttemporarily would shield it from some of the safe-ty concerns of the LWR. Further, if problems de-velop with the LWR that are too difficult to solve

economically, the HTGR probably would be thenext available concept in this country. An en-larged R&D program could prove vital in main-taining the nuclear option.

On the negative side, it has to be noted thatgas reactors have not been a great success any-where, and most countries have turned to theU.S. developed LWR technology. Estimates offuture costs and reliability are much more con-

jectural than for the LWR. Many utilities wouldbe reluctant to turn to a less familiar technologythat might turn out to be subject to many unfore-

seen problems. Such uncertainties will only beresolved by a substantial R&D program. To agreater degree than for the standardized LWRdiscussed above, a thorough demonstration of

the entire HTGR concept, including licensabili-ty, costs, operability, and acceptability would berequired. This would necessitate an increased de-velopment program at DOE.

Even if the HTGR is not seen as a replacementfor the LWR, there are still several reasons for sup-

porting an R&D program paced to make it avail-able early in the next century. it would use urani-um more efficiently than LWRs, has relatively be-nign environmental impacts, and could be usedfor industrial process heat. A small, modular formalso has been proposed that could have majorsafety and financial advantages and be particular-ly well suited to process heat applications.

It is harder to see a role for heavy water reac-tors (CANDU) in this country. CANDUs are work-ing extremely well in Canada. At least some ofthat success, however, is due to the managerialenvironment in which the nuclear industry oper-

ates in Canada. Transplanting it to this countrycould lose these advantages, and would neces-sitate industry learning and investing in a quitedifferent technology. While the technology canbe mastered, a significant research programwould be necessary to adapt CANDUs to our reg-ulatory requirements, or vice versa. It is not clearthat this effort is warranted compared to otheralternatives such as the HTCR or improved LWRs.

The final alternative reactor discussed in chap-ter 4 is the PIUS, which was conceived largelyto meet safety objections to the LWR. While rad-

ically different from the LWR in some ways, it stillis an LWR. Therefore it has an element of famili-arity that the others do not. The concept, or atleast some features of it, appear promising, butonly a significant research effort will confirm thefeasibility of the design since it is still a paper reac-tor concept, There is great uncertainty over thisconcept, but if the research program does proveout the expectations of the developers, the reac-tor could be deployed rapidly. PIUS could beperceived as much safer by the public and critics.

Development of new technology will not by

itself solve the problems of the industry.However, it will play a vital role in an overall

upgrading, whether the end result is an improvedLWR or an alternative concept.




C4. Revise the Institutional Management of Nuclear Power if Necessary

If a utility has its license revoked as in option

C2, or such action seems likely, it might think of

turning the plant over to a different operating

agent instead of just shutting it down. Utilities

already use a large number of consultants andservice companies for specific tasks. Operatingservice companies, discussed in chapter S, couldbean extension of these, or they could be otherutilities that have established good records andare prepared to extend their expertise to otherreactors. The NRC would be satisfied that theplant was being given the management attentionrequired, the utility would have its plant operatingagain, probably at higher availability than before,and the public would have greater assuranceabout the commitment to safe operation.

There are potentially serious liabilities to theidea, however. No utility would like to admit thatit is incapable of operating its plant safely andwould be reluctant to turn to another operatingcompany except under extreme conditions. Thecontract between the two would have to be care-fully worked out to determine who would payfor modifications and maintenance. If a seriousaccident did occur, plant restoration costs andliability for offsite damages would have to bespelled out. Premature plant closure due to unex-pected deterioration could be another problem.

There do not appear to be any legal impedi-ments to the idea that would require legislation.However, Congress might want to encourage theNRC, and perhaps the Justice Department, toundertake further analysis.

Alternative institutional arrangements alsocouId be formed to encourage nuclear orders i nthe future. If individual utilities are unable toundertake the risk, consortia of utilities, possiblyincluding vendors and architect-engineering firmsetc., might be able to do so. Alternatively, Gov-

ernment-owned power authorities might be theonly wa y to ma intain the nuclea r op tion. These

concepts are explored briefly in chapter 5.

Goal D: Alleviate Public Concernsand Reduce Political Risks

The issue of public acceptance has permeatedthis report for good reason. If the long-term trendin public opinion toward increasing opposition(described inch. 8) is not reversed, there will be

few, if any, more orders for nuclear plants.Many of the options discussed above are rele-

vant to this goal. Nuclear energy will not be ac-ceptable so long as there are spectacular ex-amples of out-of-control cost escalations and acontinuing series of alarming operating events.A major accident involving offsite loss of lifewould almost certainly preclude future plants andquite likely close many operating reactors. There-fore, almost any action to improve operations andsafety will pay dividends in public acceptance.The options discussed here are intended to re-

duce the controversy or to confine a role fornuclear power.

01. Ac celerate Stud ies of Alternative Energy Sou rce s

One of the major factors affecting public opin-ion against nuclear power is the feeling that therisks associated with it outweigh the benefits. Aslong as other energy sources are available thatare perceived to be both more economical andacceptable, there is little incentive to favornuclear energy with its more controversial risks.Therefore, as more information is developed onthe resource base, costs and impacts of these al-ternatives, better decisions can be made on therelative merits of nuclear energy.

The major competitor of nuclear power for newcentral station plants is coal. Yet coal is arousingconcerns (e.g., carbon dioxide and acid rain) thatmay exceed those of nuclear. Significant researchis going on in these areas, and the answers arecrucial for nuclear power. The sooner they be-come available, the easier it will be to make in-formed decisions.

Some analysts feel that natural gas resourceshave been greatly underestimated. This cannotbe confirmed for many years, but there is an im-



portant data-gathering role for the Government.If gas remains plentiful and is permitted as a boilerfuel, it will reduce the competitiveness of nuclearenergy. From a different perspective, it also mightbe useful to expand R&Don the solar energy op-tions that appear promising. Some of the euphor-ia about solar energy has withered under the hardlight of costs, but some technologies such asphotovoltaics are still candidates. Acceleratingthese technologies actually could be beneficialto nuclear power. If they ultimately prove to be

not widely competitive with nuclear energy, wewould know that sooner. If they are reasonablycompetitive, then the Nation has another option.

None of these proposals is particularly contro-versial, though some might be expensive. In gen-eral, decisions on these options will be made on

a basis other than one’s attitude toward nuclearpower. The outcome, however, could be veryimportant to the future of nuclear power.

D2. Ad dress the C onc erns of the Critics

Critics of nuclear power have long felt a deepdistrust of the industry and the NRC. They feelthat their concerns have been ignored or down-played while the Nation plunged ahead to buildmore reactors. The mistrust is mutual. The indus-try feels that nothing would change the mind ofthe critics.

Bridging this distrust will be difficult at best. Forthose critics who do not want nuclear powerunder any conditions and for those in the nuclearindustry who refuse any concessions, resolution




probably is impossible. However, opposition tonuclear power is not monolithic. Many criticshave specific concerns over the technology andits implementation. These critics tend to be tech-nically knowledgeable and respected in the envi-ronmental and anti-nuclear communities. Fur-ther, some in the industry are aware of these con-

cerns and appear to be willing to engage in a use-ful discussions. It is with these groups that abridge might be constructed.

One important step is to resolve the safety con-cerns that have already been identified, as dis-cussed under option Cl. Further steps may berequired to convince critics that every effort wasbeing made to identify previously unrecognizedsafety concerns and implement solutions at ex-isting reactors.

The most straightforward way of providing thisassurance is to involve critics directly in the reg-ulatory process. This might be done through con-tracts to supply specific information or review ma-terial (intervener funding), by including techni-cally knowledgeable critics on the Advisory Panelfor Reactor Safeguards, or by creating an officewithin the NRC that would serve as a liaison tothe critics.

Few proposals generate as much controversyas this one. Industry sees it as opening the flood-gates to implacable opposition that would makeany progress impossible. Utilities see it as presag-ing a steady stream of new backfit orders and un-

necessary regulations. Much of the NRC sees itas an unwarranted infringement on its process.If it is to be implemented, congressional direc-tion will be needed. It may be worth the effort.Nuclear technology is still imperfect, and thesooner problems are discovered, the sooner thetechnology can be improved. Involving the criticsis likely to speed this process. I n addition, im-provements in public support is a prerequisite formore orders; public support is unlikely to im-prove as long as the controversy over safety isso bitter; and this controversy is unlikely to diedown until most of the concerns of the criticshave been addressed. Given the current impassethere may be little to lose by trying this approach.

D3. Control the Rate of Nuclear Construction

Many of the concerns over nuclear power origi-nated during the late 1960’s and early 1970’swhen projections of very high growth seemed tobe on the way to being realized. People becamealarmed over the thought of 1,000 reactors or

more around the country, many reprocessingplants with spent fuel and plutonium shipmentsrequiring security that disrupts normal transpor-tation and threatens civil liberties, and the everpresent possibility of accidents. The industry itselffound the rapid expansion more than it couldadequately manage.

Lower projections of nuclear growth have re-duced some of these concerns. However, a re-sumption of orders might rekindle the fears ofanother “too rapid” expansion.

Establishing a controlled growth rate may give

assurance that the early. concerns about overex-pansion would not recur. As discussed in chapter8, the 1980 Swedish national referendum limitingthe total number of reactors appeared to quellthe political controversy.

Controlling the growth rate might realize thisimprovement in public acceptance without pre-cluding all future orders. The limit could be inthe form of capacity that could be granted con-struction permits in a year, or a sliding scale toallow nuclear construction to remain at a roughlyfixed fraction of total new capacity. If utilities’ in-terest in new nuclear orders revives to the extentthat the growth rate could be exceeded, the NRCwould allocate the permits using criteria of re-gional need and ability to manage the technologyas discussed under A3 above. There is no intrin-sic difficulty in the Government allocating limitedpermits (e.g., airline routes were limited for manyyears).

This policy option would be controversial, atleast at first. The industry would argue that itwould constitute unwarranted interference in themarketplace and would distort economic deci-sions. In particular, if nuclear reactors turn out

to be the most economic form of electric power

when manag ed carefully and if fully redesigned,



262 ŽNuclear Power in an Age of Uncertainty

controlled growth could result in a misallocationof resources. However, until public acceptanceimproves noticeably, no utility is going to orderany reactors. If controlled growth were instru-mental in improving public acceptance to thepoint that orders resumed, it would be of majorassistance to the industry. Furthermore, the bur-

den of proof should be on the industry that itcould manage a rapid increase of orders, sincemany of the present problems came from the lastsurge. The NRC has the authority to prevent sucha surge by insisting on preapproved designs andevidence of utility capability, but a congressional-ly imposed limit would be more convincing tothe public and the critics.

D4. Ma in ta in a St r ic t Nonpro l i fe ra t ion Stanc e

Proliferation has been one of the major con-cerns of the critics and the public. All known re-

actors could be used in some fashion to facilitatethe production of nuclear weapons. If nuclearpower is to regain public trust, this linkage mustbe minimized.

One step is to keep the U.S. weapons programsand power programs sharply distinct, both tech-nically and institutionally. Separate waste dispos-al programs could be one step, even though thematerial is not much different. Consideration alsomight be given to removing the weapons pro-gram from DOE though that would be a compli-cated decision beyond the scope of this study.

A related suggestion is to consider a ban or ex-tended moratorium on reprocessing. Reprocess-ing is the focus of much of the opposition tonuclear power. A long-term legislated moratori-um, perhaps coupled with the controlled growthin option D3 and the extended burn up of optionB1, would eliminate many of the major causesof concerns.

D5. Promote Regional Planning for Electric Power Capacity

One of the major reasons for the poor public

opinion of nuclear power is the perceived lackof need for it. As discussed in chapter 3, this needis unlikely to be readily apparent before the late

1990’s. By then, many utilities may be findingtheir own capacity fully committed and bulkpower purchases less available. Without majorchanges in the way we generate and use elec-tricity-changes that are highly speculative now —the Nation will need substantial new generatingcapacity to come online by 2000, and perhapssooner. Some regions with high growth rates frompopulation shifts and economic changes will ex-perience the need earlier.

Planning for electric growth can make clearwhat the choices are and what the consequencesmight be. These plans could help build a consen-sus on the necessary additional capacity and loadmanagement. At the least, plans would providea format for discussion. in conjunction with thecontrolled construction rates discussed above,such plans could be quite effective.

National planning is probably too large a scaleto be useful. State planning may be too small con-sidering the growing regional nature of powerwheeling. Regional planning appears best to cap-ture the commonality of interests, This might becombined with regional rate-setting as mentionedin chapter 3.

insofar as nuclear power is concerned, this pro-posal might not make any difference. It wouldnot by itself eliminate any barriers to new reac-tors and might even raise an additional layer ofregulation. On the whole, however, it should

allow utilities that decide they should build areactor to make a stronger case for it by show-ing how it will benefit the customers in the longrun.

This policy option would be implemented bysetting up regional planning authorities, possiblywith ratemaking authority, that are agreed to bythe States. It is important that these authoritiesalso have authority to determine power needs.Such responsibility should not go by default toFederal agencies such as the NRC, which are notwell equipped to make such determinations. The

concept of regional authorities appears promis-ing, but it has not been studied in detail in thisproject.



Ch. 9—Policy Options q 26 3

MAJOR FEDERAL POLICY STRATEGIES AND THE LIKELIHOODOF MORE ORDERS FOR NUCLEAR PLANTS

It should be clear from the other chapters inthe study that none of the individual policy op-tions described earlier in the chapter is sufficientby itself to improve significantly the prospectsfor more nuclear orders. There are too many dif-ferent problems that have to be addressed beforenuclear power can be again considered a viableenergy option for the future.

If several of these policy options are coupledin an overall strategy, however, they may be col-lectively much more effective. These strategiesshould include options directed toward each ofthe four policy goals described earlier in the chap-ter: reduce construction costs, improve reactoreconomics, reduce the risks of accidents, andalleviate public concerns. Most of the policy op-

tions will be controversial to some extent. For thisreason, it is likely to be necessary to take stepsthat meet the concerns of several different groupsat once: utility executives, critics, investors, reg-ulators, and the public. The divergence amongthe views of these groups should be clear fromthe rest of the study.

In the face of the controversies surrounding nu-clear power and the uncertainties surroundingits future, one obvious Federal course is to makeno changes in Federal policy. If such is the case,

future nuclear orders will be heavily influenced

by two sets of conditions outside the direct con-trol of the Fed eral Government: econo mic c ondi-

tions and improvement in nuclear industry man-

agement. In the section that follows, the pros-pects for new nuclear orders in the absence ofnew Federal policies, the Base Case, are exam-ined for each of four nuclear futures that assumedifferent sets of economic conditions and industrymanagement success.

The two other strategies described here assumevarious degrees of Federal intervention on severalfronts. The first of these, Strategy One, would

merely remove obstacles to further nuclear or-ders. A more active approach, Strategy Two,would go further and attempt to stimulate morenuclear orders. The four futures described underthe Base Case also are examined for each strategy

to help evaluate how successful the strategiesmight be under different conditions.

There are also two variations on these strategies

that are not analyzed in detail in this study butare worthy of consideration. One of these, a vari-ation on the Base Case, would make severalchanges in Federal policy to encourage moremarket competition between nuclear power andother generation (and load management) tech-nologies. The other, a variation on Strategy Two,would consider nuclear power, not so much asan important aspect of U.S. energy policy but asa key element of U.S. industrial and world tradepolicy.

The Base Case and two strategies are outlined

in table 35 with the policy options, discussed inthe previous section, listed for each strategy.There is also a brief description of the two varia-tions with a general description of the probablepolicies under each.

Base Case: No Change inFederal P o l i c i es

There are several reasons why policy makersmight choose a strategy that avoids any majorchanges in the current Federal laws and regula-tions affecting nuclear power. Some policy makersmay view nuclear energy as unimportant or un-desirable, while others may feel that the FederalGovernment already has done enough for the in-dustry, making further legislation unwarranted.Still others may not wish to take any action rightnow. At present there are many uncertaintiesabout future electricity demand, the environmen-tal impacts of coal combustion, and the poten-tial of conservation and renewable resourceswhich will affect the necessity for and attrac-tiveness of nuclear power. Policy makers mayprefer to wait 5 or 10 years to see how these

uncertainties are resolved before revising currentnuclear energy policies. Finally, policy makersmay feel that Federal legislation would have lit-tle impact on the industry and that economicforces will ultimately determine its fate.




Table 35.–Major Policy Strategies (and the policy options included in each)

Strategy Policy Options Included

Base Case: No change in Federal nuclear policy: Three noncontroversial policies that would be usefuleven in the absence of more orders

Goals Policy Options Improve reactor economics . . . . . . . . . . . . . . . . . . . . . . . . . (B1) R&D to improve fuel burnupReduce accident risk . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (Cl) Improve analysis of reactor safetyAlleviate public concern . . . . . . . . . . . . . . . . . . . . . . . . . . . (D1) Accelerate studies of alternative energy sources

Variation: Sharpen market competition of nuclear power This strategy, not analyzed in detail, would include some or all of steps towards: reduction or removal of Federal subsidiesfor nuclear and alternatives; marginal cost pricing; deregulation; full costing of external impacts

Strategy One: Remove obstacles

Goals Reduce capital cost barrier . . .

Improve reactor economics . . .Alleviate public concerns . . . .

to more nuclear orders: Three policies above plus five others

Policy Options . . . . . . . . . . . . . . . . . . . . . . (Al) Revise regulation

(A2) Assist funding of standardized optimized LWR design. . . . . . . . . . . . . . . . . . . . . . (B2) Improve utility management. . . . . . . . . . . . . . . . . . . . . . (D2) Address concerns of critics

(D3) Control the rate of nuclear construction

Strategy Two: Provide a moderate stimulus to more nuclear orders: Eight policies above plus eight others

Goals Reduce capital cost barrier . . . . . . . . . . . . . . . . . . . . . . . . . (A2)

(A3)Improve reactor economics . . . . . . . . . . . . . . . . . . . . . . . . . (B3)

Reduce accident risk . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (C2)(C3)(C4)

Alleviate public concern . . . . . . . . . . . . . . . . . . . . . . . . . . . (D4)

Policy Options Assist funding of a demonstration of new LWR designsPromote the revision of rate regulationSolve occupational exposure liability

Certify utilities and contractorsDevelop alternative reactorsRevise institutional management of nuclear operationsMaintain nonproliferation policies

(D5) Promote regional planning for electric growth

Variation: Support the U.S. nuclear industry in future world trade This strategy, not analyzed in detail, would support industry and utility R&D and export financing policies aimed at obtaininga major share of the future world market in nuclear and other advanced electrotechnologies.SOURCE Office of Technology Assessment.

A “no change” strategy would continue thecurrent Federal policy toward nuclear power.DOE could continue to fund R&D of both theLWR and alternative reactor types at about cur-

rent levels. Although current NRC efforts toreduce backfit orders and streamline the licens-ing process would continue, there would be nomajor legislation and no fundamental changes inpresent regulatory procedures.

This strategy does assume continuation of twofairly controversial Federal policies. One assump-tion is that Congress will renew with no majorchanges the Price-Anderson Act, limiting the lia-bility of plant owners and constructors in theevent of an accident (described in ch. 3). Part of

the act expires in 1987 and if it were not renewed,it could have a significant impact on the nuclearpower industry, although how much and whatkind of impact has not been analyzed in this re-

port. A second assumption is that the NuclearWaste Policy Act of 1982 is implemented success-fully and the feasibility of safe waste disposal willbe demonstrated.

The strategy also assumes that three noncon-troversial policy options (actually expansions ofexisting efforts) discussed in the preceding sec-tion could be implemented: (B I ) R&D for higherburnup and other improvements to reactoreconomics would be funded; (Cl) Safety con-cerns would be addressed more vigorously; and



Ch. 9—Policy Options . 265

(D1) research into problems and opportunitiesfor alternative sources of electricity generationwould be accelerated.

The likely outcome if Federal policy is notchanged will depend on two major factors—the success of industry efforts to eliminate cur-

rent problems, and the economy. Two alter-native sets of external economic conditions areconsidered here. One would result in a relative-ly high demand for new central station generating

plants and the other in low growth. Similarly, thepotential range of results of current industry ef-forts to improve the viability of nuclear energyare represented by two different outcomes: rela-tively successful and only moderately successful.These outcomes, or scenarios, are summarizedin table 36, and will affect the impact of each ofthe other two strategies described in this reportas well as the Base Case results. These scenarios

are not predictions or projections of the future,but instead brief sketches of a few of the possi-

ble combinations of events that could make nu-clear power more or less attractive to utilities overthe next 10 years.

Economic Conditions: Two Scenarios

The major economic factors that will affect fu-

ture demand for nuclear power are the rate ofgrowth in electricity demand, the price andavailability of alternative energy and electricitysources, and inflation and interest rates. All ofthese factors are discussed in greater detail inchapter 3 and summarized only briefly here.

Economic Scenario A: More Favorable to Nu-clear Orders. —As shown in table 36, EconomicScenario A includes a combination of thoseeconomic factors that could be expected to makenuclear power more attractive. In this scenario,rapid price increases for oil and gas might ac-

celerate the shift to electricity helping create amoderately high increase in electricity demand

Table 36.—Four Scenarios Affecting the Future of the Nuclear Industry

Economic Scenarios Affecting the Nuclear Industry

Variable Scenario A: Favor more orders Scenario B: Hinder more orders

Electricity demand (average annualgrowth rate 1983-93 . . . . . . . . . . . . . . .

New capacity needed in 1995 (GW)would have to be ordered in late1980’s a . . . . . . . . . . . . . . . . . . . . . . . . . .

Additional capacity needed between1995 and 2000 at same demandgrowth rate (GW) would have to be

ordered by 1992 . . . . . . . . . . . . . . . . . . .Price of alternative fuels:

Oil and gas . . . . . . . . . . . . . . . . . . . . . .

Renewable . . . . . . . . . . . . . . . . . . . . . . .

Inflation rates and interest . . . . . . . . . . .

Environmental impacts . . . . . . . . . . . . . .

3.5%

161

218

Real price increases faster than priceof electricity

Real price remains higher than price ofelectricity

Low

Concerns about acid rain, global CO2

increase

Industry Improvement Scenarios

1 .5 ”/0

o

84

Real price increases at same rate aselectr icity

Price becomes competitive withelectricity

High

No constraints on fossil

Variable Scenario A: Major improvements Scenario B: Modest improvements

Average construction time . . . . . . . . . . . 7 years 12 years

Operation of existing reactors . . . . . . . . 70% availability 60°/0 availability

Safety risks . . . . . . . . . . . . . . . . . . . . . . . . Currently operating reactors shown Little progress on unresolved safetymuch safer issues

Reportable operating events. . . . . . . . . . Almost none over decade; management Continue at current rate; much mediaimprovements obvious coverage

s e e ch. 3 for a complete discussion of assumptions used in capacity projections; GW as used here means GWe.Possible factors In price increases: limited gas resemes; tight international oil market; increased environmental controls on coal burning.csteady resemes of oil and g=; continued consewation eases demand.


25-450 0 - 84 - 19 : QL 3




(3.5 percent per year). This rate of growth in de-mand, coupled with a moderate need to replaceaging powerplants, is expected to create a needfor 161 gigawatts (GW)* of new central stationgenerating capacity by 1995 and an additional 218GW by 2000. Given the time required to com-plete new generating plants, utilities would be ex-

pected to order this much capacity in the 1983-93decade.

Under this scenario, increasingly stringent en-vironmental restrictions could make new coalplants very expensive. If this price increase oc-curred at the same time as the projected growthin electricity demand, utilities would be facedwith the need for new capacity while their mostimportant fuel was becoming considerably harderto use. As a result, nuclear power would appearmuch more attractive to utilities placing power-plant orders. If the inflation rate and prevailing

interest rates were relatively low, capital costs ofnuclear plants would be more manageable andpredictable for utilities. Low inflation and thedecreasing construction costs over the next fewyears will stabilize rates to consumers, very like-ly lessening hostility to utilities. In addition, thebenefits of nuclear power would grow in the eyesof the public as electricity demand increases.

Economic Scenario B: Less Favorable toNuclear Orders. —If the economy follows thispath, nuclear power remains relatively less attrac-tive. In this scenario, moderate price increasesof gas and oil slow the shift to electricity. In ad -

dition, renewable energy sources become morecompetitive with central station electricity. As aresult, there is only slow growth (1.5 percent) inaverage annual electricity demand, and no newgenerating capacity must be completed in thedecade. However, even at this slow rate ofgrowth, if moderate numbers of existing plantsare retired, about 84 GW of new generating ca-pacity would be needed by the year 2000, andthis capacity would have to be ordered in the1983-93 time period. With relatively small in-creases in the price of coal, and high interest rates

driving up the capital costs of nuclear plants, util-ities would be more likely to invest in coal-firedplants.

*One gigawatt equals 1,000 MW (1 ,000,000 kW) or slightly lessthan the typical large nuclear powerplant of 1,100 to 1,300 MW.

in Economic Scenario B, rapid inflation overthe next few years would cause continued priceincreases and continued high interest rates dur-ing completion of the 30 nuclear plants now un-der construction. Utilities would be forced to re-quest large rate increases from utility commissionsas the plants are finished. These rate increases,

combined with the slow growth in electricity de-mand, would cause consumer opposition and in-creased public skepticism about nuclear power.All of the assumptions included in Economic Sce-nario B would be expected to make new nuclearplants less attractive to utilities.

Management Improvement Conditions:Two Scenarios

Industry and utility success or failure to makesubstantial improvements in the management ofthe nuclear enterprise will be reflected in several

indicators: Ieadtime to build nuclear plants; aver-age plant availability; progress on unresolvedsafety issues; and frequency of precursor events.These subjects were discussed in chapters 4 and5.

Management Scenario A: Major lmprove-ment.—ln Management Scenario A, the nuclearindustry would be very successful at overcom-ing some of its current difficulties without govern-ment assistance. Utilities currently operating reac-tors would overcome operating and safety prob-lems, creating a steady improvement in reliabili-ty of operating reactors. Improved operation ofexisting plants and projections of reduced con-struction costs wouId make nuclear power moreeconomically attractive to investors and publicutility commissions as well as to consumers.

Management Scenario A assumes that operat-ing plants and those completed over the nextdecade would be shown to be much safer thanpresently assumed because of improved opera-tions and better understanding of the technology.Improved analysis and information (e.g., theongoing research into “source terms”) coulddemonstrate other safety characteristics (as

discussed in ch. 4).

Two consequences would follow from thesesafety improvements. First, the managementchanges would greatly reduce major events, suchas the failure of the automatic shutdown system




at the Salem, N. J., reactor, which are viewed bythe public as precursors to a major accident. Sec-ond, the new information on the small amountof radioactivity released in the event of an acci-dent would temper the reaction to the few oper-ating events that did occur over the decade.These safety gains should be helpful in reduc-ing public opposition to the technology and fur-ther increasing investor confidence.

In addition to the improvement in utility man-agement of operating reactors, the nuclear supplyindustry would be offering improved standard-ized LWR d esign s suc h a s the APWR c urrently

being developed by Westinghouse and Japan.

Under current regulatory policy, the NRC couldgive licensing approval for a complete design,and, if a plant were built exactly to the design,there would be few regulatory changes duringconstruction. Thus, the regulatory environment

would become somewhat more predictablewithout any major Federal legislation.

Management Scenario B: Minor lmprove-ments.—ln Management Scenario B, some of theweaker utilities would fail to improve their per-formance despite the efforts of INPO and theNRC. Average availability for operating plantswould be only about 60 percent, and there wouldbe little progress in solving unresolved safetyissues. poor management of operations wouldcontinue to cause precursors to serious accidents,and the media wouId continue to give extensive

coverage to these near-accidents and major con-struction problems. One operating event mightbe so significant that a plant would have to be

shut down for several months to a year for repairs.Without an adequate insurance pool, this wouldcause a major rate hike to cover purchased pow-er. The long construction periods, continuingoperating problems, and rate hikes due to out-ages would increase investors’ and consumers’skepticism of the technology.

without Government intervention, this sce-nario could be expected to have very negativeconsequences for the industry regardless of ex-ternal economic events.

Four Nuclear Futures Under the BaseCase: No Policy Change

The two sets of economic conditions describedabove combine with the two management sce-narios to form four futures that illustrate the rangeof possibilities for more nuclear orders. FutureOne is a combination of favorable economic con-ditions (Scenario A) and major improvementsin management (Scenario A). Future Four com-bines the least favorable scenarios. Futures Twoand Three are intermediate. The discussion thatfollows describes the factors under each futurethat would affect decisions on new nuclearplants. The four futures and their likely outcomesare summarized in table 37.

Future One.—Under the assumptions of FutureOne, there is a clear need for more generatingcapacity, the cost of other fuels is rising rapidly,

operating plants are performing well, and industryoffers improved, standardized designs. As dis-cussed in chapter 8, public acceptance of nuclear

Table 37.—Four Combinations of Economic and Management Scenarios Under the Base Case

Management Scenario A: Majorimprovement7-year construction time; 70% peravailability; safer reactors; few precur-sor events

Management Scenario B: Minorimprovement12-year construction time; 60°/0availability; little progress on safety;continued precursors.

Economic Scenario A: More favorableFairly rapid growth in demand; utilityalternatives costly; low interest/inflation.

Future One

Some further orders possible, especial.Iy if standardized preapproved designsare available.

Future Three

A few well-managed utilities may orderplants over the next decade.

Economic Scenario B: Less favorableSlow growth in demand; alternativeenergy available; high interest/inflation

Future Four

More orders very unlikely before 2000;a few possible after 2000.





technology is influenced by perceived benefitsas well as by perceived risks. Therefore, the in-creased electricity demand in Economic ScenarioA, combined with the safety improvements envi-sioned in Management Scenario A could be ex-pected to increase support for nuclear energy.Utilities would be more confident that plantscould be completed without unreasonable delaysdue to changing regulations or intervention.

Under these circumstances, and if the 7-yearconstruction period for some recent plants ap-pears achievable for new plants as well, someutilities might be willing to order new nuclearplants even without changes in Federal policy.However, most utilities would still be deterredby uncertainties and risks, especially regulatorydelays and costs.

Those utilities with a need for new capacity

might choose to share these risks by forming aconsortium. This consortium could order severalplants based on the best current design and sharethe necessary startup costs with component man-ufacturers. Under current regulatoy procedures,the NRC could grant simultaneous constructionpermits for three or four new plants. The SNUPPSconsortium in the early 1970’s is a prototype ofsuch an effort (see ch. 5). If the plants were builtin strict accordance with the complete design,there would be only minimal requirements forchanges during construction.

A problem-free licensing and constructionprocess would demonstrate to other utilities thebenefits of standardization and show that at leastsome utilities were committed to the technology.Given the need for additional power and theconsortium’s expected success, nuclear ordersmight “snowball” without any major Federalaction.

Future Two.–The safety and management im-provements, reduced construction times, and ex-pected increased public acceptance under Man-agement Scenario A would make nuclear powera more attractive option. However, the assumed

slow growth in electricity demand in EconomicScenario B would make utility investment in anytype of new powerplant unattractive throughoutmost of the decade.

As discussed earlier, Economic Scenario B en-visions that 84 GW of new electric-generatingcapacity would have to be ordered only at theend of the 1983-93 decade. By then, the absenceof orders and slowdown in construction wouldhave eliminated many suppliers of nuclear plantcomponents and services, increasing the cost of

a new nuclear plant. This cost increase, com-bined with high inflation and interest rates, mightoffset the savings from reduced constructiontimes envisioned in Management Scenario A, fur-ther discouraging nuclear orders. Given the over-all risks and uncertainties, it is unlikely that a util-ity would order a nuclear unit unless its projectedcosts were much lower than a similar coal plant.This is unlikely, however, because poor businessprospects would keep nuclear companies frominvesting heavily in the design and analysisneeded to reduce costs.

At most, only one or two utilities could be ex-pected to order a nuclear plant under FutureTwo, given no major change in Federal policy.Any orders that did occur probably would comefrom a very experienced nuclear utility and mostlikely would be in the form of initiating or com-pleting construction at a currently inactive plantsite. However, it is more likely that there wouldbe no orders at all over the next decade underthese circumstances. Given some utilities’ cur-rent problems with nuclear energy, no utilitywould want to be the only company venturinginto a new nuclear effort.

In the 1990’s, with few or no new orders, theU.S. nuclear industry would lose most of its ex-pertise in plant design and construction, andsuppliers of key components would drop out. De-spite these problems, some utilities could be in-terested in ordering reactors toward the end ofthe century, especially if demand growth startsto increase. By then, the utilities would probablyfind a Japanese, French, or West German designpreferable to outdated U.S. plant designs. If U.S.companies wanted to offer the most current reac-tor designs, they might have to license them from

foreign companies, perhaps the very ones theyhad licensed to build LWRs in the first place. Ineither case, as discussed in chapter 7, U.S. com-panies probably would still have the resources




to tailor the designs to American needs and tobuild the plants.

Future Three.-The poor management condi-tions of Management Scenario B, under a policyof no change in Federal policies, would createserious tensions between the need for nuclearpowerplants and continued opposition (by invest-

ors, critics, and the public) to their high costs andrisk. As discussed previously, this scenario envi-sions constraints on fossil fuel combustion, en-couraging utilities to consider purchasing nuclearplants, Those utilities with successful experiencein building and operating nuclear plants wouldnot be deterred directly by the poor experiencesof others. Indirectly, however, all utilities are“tarred by the same brush” aimed at the weakerutilities by the NRC actions, the public, investors,and others. Thus, the lack of major industry im-provements envisioned in Management ScenarioB would cause present uncertainties to continue.

Moderate interest and inflation rates assumedunder economically favorable Economic ScenarioA might help to offset the cost escalation resultingfrom the lengthy average construction period ex-pected in Management Scenario B. This wouldmoderate the “rate shock” as new plants cameon line, reducing consumer opposition slightly.Public perception of increased electricity demandcould be expected to offset concerns about safetyrisks slightly, perhaps returning public opinion toa 50-50 split on the technology.

The net effect of favorable economic condi-tions combined with little change in the stateof the industry and no major Federal policychange would be that, at most, only a fewutilities would order new nuclear plants over thenext decade. Despite the constraints on fossil fuelcombustion, most new plants would be coal-fired, with environmental controls to meet cur-rent regulations. With only a few new orders,subsequent events would follow the path out-lined for Future Two. In essence, the U.S. nuclearindustry wouId slowly decline, and any new nu-clear plants ordered after 1995 might be of foreign

design.

Future Four.–With the combination of bothManagement Scenario B (Minor Improvement)and Economic Scenario B, there is little prospectfor any nuclear orders with no change in Federalpolicies. Continued management and safetyproblems at plants currently operating and underconstruction, slow growth in electricity demand,

and increasing competitiveness of other fuels,create a climate in which no utility could be ex-pected to order a new nuclear plant. In the in-flationary environment expected under EconomicScenario B, utilities would be very reluctant toinvest in capital-intensive plants of any type. In-stead, utilities could be expected to match supplyand demand through load management, efficien-cy improvements to existing coal plants, andcogeneration, contributing to the slow overall rate(1.5 percent) of growth in electricity demand.

Near the end of the 1983-93 decade, some util-ity executives might order relatively inexpensivecombustion turbines to supply the 84 GW ofnew capacity needed by 2000. Although Eco-nomic Scenario B expects moderate gas prices,an increased reliance on gas and oil for electricitygeneration might drive up prices, resulting inmuch higher electricity prices. Despite high in-terest and inflation rates, a few utilities mightorder coal- fired plants in the early 1990’s. Theelectricity from these plants would be rather ex-pensive because of the high capital costs, further-ing dampening growth.

Most designers and equipment suppliers would

leave the business, leaving a much smaller U.S.industry. Because public opposition would behigh under this combination of scenarios, utilitieswould be unlikely to order new plants fromabroad, and no new nuclear plants would be builtin the United States before 2000.

Even in Future Four, however, there is apossibility that new nuclear units could be builtafter 2000. As discussed in chapter 7, the U.S.nuclear industry is very resilient. By maintainingits expertise with fuel service and waste disposalbusiness, the industry still should be capable of

building a reactor (probably based on foreign




Another problem with the current system is thatsome of the costs of different sources of electricityare not fully reflected in the private cost of suchsources. Nuclear power receives some direct orindirect Federal assistance of several kinds: 1 )Federal limits on public liability following a nu-clear accident (the Price-Anderson Act), 2) Fed-

eral subsidies for uranium enrichment, and 3) theFederal cost of nuclear safety regulation andnuclear R&D. Coal-fired electricity also receivesFederal assistance from: 1) black lung payments(currently about $1 billion/year); 2) Federal coalmine regulation; and 3) no charge for air pollu-tion within regulatory limits.

investment comparisons for nuclear power andcompeting generating technologies would bemore accurate if these subsidies were eliminated.This study has not analyzed the consequencesof reducing Government support, such as R&Dor the effects of eliminating the Price-Andersonliability limitations. It does seem clear, however,that such acts would be viewed by both the in-dustry and the public as signaling a lessening ofthe Government’s commitment to nuclear pow-er. At this point, the industry can ill afford suchsignals. Therefore, it should be recognized thattaking these initiatives, whatever their overallmerits, may well be tantamount to ending the nu-clear option, at least for the foreseeable future.

Strategy One: Remove Obstaclesto More Nuclear Orders

The intent of this strategy (see table 35) for alist of policy options) is to establish an environ-ment in which utilities would be more likely toconsider nuclear reactors if demand does pickup over this decade and, at the same time, to es-tablish policies that would win the support of nu-clear critics and the public.

The policy options included here work togetherto achieve these ends. Two of the options, (Al)revise regulation and (A2) develop standardizedoptimized LWR designs, are closely linked. Theseoptions wouId eliminate some of the major con-

cerns utilities have over nuclear. It is more dif-ficult to predict how to gain the necessary publicsupport. Strategy One includes three policy op-tions designed to reduce the controversy over nu-

clear power and the concerns of the public: (62)improve utility management of nuclear plants,(D2) address the concerns of the critics, and (D3)establish limits on future nuclear constructionwithin the context of a balanced energy program.

Of these policy options, the easiest to imple-ment is the involvement of the NRC in upgrading

utility mana gem ent. Such a program a lread y ex-

ists. The c halleng e is to ma ke it mo tivate utilities

to excellent performance (as is discussed in ch.

5) rather than merely to avoid getting in troublewith the NRC. This is probably only possible ifthe NRC program is developed in close coopera-tion with IN PO. There are several possible waysdiscussed earlier to involve interveners moreclosely in monitoring and improving nuclear plantsafety. This policy option is likely, however, tostimulate substantial opposition from utilitiesunless it is clear that it is closely coupled with

licensing and backfit reform.The effect of Strategy One on utilities’ percep-

tions of costs and schedules would be mixed. Li-censing and backfit reforms would help assureutilities that they could build the plant as designedonce a construction permit were approved. How-ever, opening the process more to the critics in-troduces another element of uncertainty, espe-cially for the first few orders. Furthermore, theseproposed reforms would not necessarily dramat-ically reduce capital costs, so electricity fromaverage cost nuclear plants still could be

perceived as more costly than electricity fromaverage cost coal plants in most U.S. regions (seech. 3).

The impact of Strategy One on orders for newnuclear plants also would vary sharply for eachof the four nuclear futures described above forthe Base Case. As can be seen in table 38, theimpact of this strategy under each of the fourfutures does not differ greatly from the impact ofthe Base Case. In light of the considerable dif-

ficulty that would be involved in implementingthe five policy options, this finding suggests that

Strategy One maybe only marginally effective.

Future One.—Under these conditions that arerelatively favorable for more nuclear orders, Strat-




Table 38.-Alternative Nuclear Futures Under Strategy One: Remove Obstacles to More Nuclear Orders

Management Scenario A: Majorimprovement

Management Scenario B: Minorimprovement


egy One would increase the

Economic Scenario A: More favorable Economic Scenario B: Leas favorable

Future One

A few orders likely if utilities are will-ing to be the first; a consortium and/orturnkey contracts could initiateordering.

likelihood relativeto the Base Case. Utilities would have greaterconfidence that a new reactor could be builtclose to the projected cost and schedule. Mostmajor design questions would have been workedout before construction had started. The NRCessentially would have approved the design and

apparently would be ready to move an applica-tion through expeditiously. Critics would havebeen given ample opportunity to critique the de-sign. Controversy over nuclear power would benoticeably lower because the last of the presentreactors under construction would be completewithout a continuation of the cost overruns theyare now experiencing, and operating reactorswould show considerably improved perform-ance,

Under such circumstances, vendors might bewilling to encourage the first orders by offering

a fixed price “turnkey” contract. The first fewplants might not be much less expensive thanpresent plants under construction, but simpli-fied engineering and cumulative construction ex-perience would be expected to cut the cost ofa typical plant 20 to 30 percent from currentlevels (see chs. 3 and 5). Utilities might hesitateto order the first few plants largely because ofdoubts that cost and regulatory problems reallyhad been solved. Sharing the risk with the ven-dors would do much to alleviate these concerns.If the experiences of the first few orders werefavorable, further orders could be expected in

line with demand growth.Intervener involvement in the licensing process

for standard designs combined with the effects

of good management on plant operations shouldreassure nuclear critics and reduce the reasonsfor opposition to nuclear power in licensing pro-cedures and electricity rate hearings. The strongneed for more powerplants coupled with the rela-tively high prices and environmental problemsof coal in Economic Scenario A increase the rel-ative advantage of nuclear and also reduce op-position at State regulatory hearings.

Future Two.– If economic and industrymanagement conditions are less favorable thanthey are in Future One, the policies of StrategyOne will not succeed in stimulating very manyorders. In Future Two, industry management isassumed to improve substantially but electricitydemand grows slowly and inflation and interestrates are high, discouraging capital expenditures.Because of fewer precursor incidents, public ac-ceptance grows, but there is little obvious needfor nuclear powerplants.

With only 84 GW to be constructed by 2000,there would be little pressure to diversify intonuclear. With few prospective orders, vendorsand architect-engineers would be unlikely to takethe risk of offering turnkey projects. A few util-ities, seeking to preserve the option, might makea point of at least considering a few nuclearplants. The new standardized designs, especialyif smaller than current designs, and streamlinedlicensing could make reactors competitive with

coal. Under Future Two, Strategy One has asomewhat better chance than the Base Case ofleading to a few more orders by the end of thecentury.



Ch. 9—Policy Options . 273

Future Three. -Under these conditions, Strate-

gy One actually could reduce the prospects fornew nuclear orders from what they are underthe Base Case. With continued poor plantoperating performance and continued precursorevents, nuclear critics will not be satisfied thatadequate progress is being made. With inter-

venors more closely involved in safety regulation,the lack of progress on resolving safety concernscould increase the time devoted to particular safe-ty issues. Even with high growth, the utilities are

likely to regard more nuclear orders as too risky,

not only from technical problems raised by the

NRC and the critics, but also from the financial

impact of public opposition on rate hearings and

investor decisions.

Future Four.– Finally, the policy options ofStrategy One could diminish still further the pros-pects for nuclear orders under the dismal condi-

tions of Future Four which combines both ad-verse economic conditions and little industry im-provement. The combination of intervener in-volvement with poor industry improvement andlittle apparant need for nuclear power couldcreate conditions of public opposition thatwould make orders unlikely even after 2000.

Strategy Two: Provide ModerateStimulation to More Orders

Strategy Two builds on Strategy One. It assumesthat efforts to remove obstacles to more nuclear

orders would be inadequate, largely because util-ities still would be unconvinced that the prob-lems were resolved. As in Strategy One, policy

measures to reduce capital and operating cost ofnuclear plants are combined with policy meas-ures to make nuclear power more acceptable tonuclear critics and the public.

The first policy option is a demonstration of thereactor(s) developed under A2 as discussed inStrategy One. The main purposes would be toshow that the reactor could be built accordingto design and that the licensing process couldhandle it as expected. It could be expected thatprivate industry would fund most of this projectsince technological feasibiIity is not i n question,but significant Federal participation could berequired.

A second step is (A3) Stimulate improved rateregulation treatment of capital-intensive projects.No Federal budget is required for this policy op-tion but sensitive Federal leadership is neededsince rate regulation traditionally has been leftto the states.

Strategy Two adds a politically controversialstep to improve operating economics with (64)Reduce uncertainty about occupational exposureliability. Clarifying occupational exposure con-ditions and ranges of possible payments to ex-posed workers with health problems might addexpense to the operation of nuclear plants butwould reduce the uncertainty which accompa-nies an unspecified liability (which could be thesubject of private lawsuits, such as is now hap-pening with asbestos exposure liability).

Strategy Two includes three additional optionsto reduce the risk of a reactor accident. (C2) would

Table 39.—Alternative Nuclear Futures Under Strategy Two: Provide a Moderate Stimulus to More Nuclear Orders

Economic Scenario A: More favorable Economic Scenario B: Less favorable

Management Scenario A: Major Future Oneimprovement

Some plants ordered by about 1990;more by the end of the century.

Management Scenario B: Minor Future Three Future Fourimprovement

New orders Iikely only if Government Government actions to improveoptions to improve management have a management and R&D funding of newbig impact on utilities and public opin- reactor types should improve pros-ion, resulting in Future One; new reac- pects for more orders after 2000.tor types would be useful.

SOURCE: Office of Technology,




require the NRC to restrict construction licensesfor nuclear plants to qualified utilities and con-struction contractors. Substantial Federal funding,up to several billion dollars over 10 years or more,would be required for a second step: (C3) Funda major R&D program on alternative reactortypes. If this option is pursued at a high level, the

demonstration LWR probably would be deleted.For a third option, (C4) Revise institutional man-agement of utilities, there could be controversialchanges in Federal and State utility regulation,antitrust law and other long-standing institutionalguidelines. These steps should ensure the avail-ability of reactors and operators in which the pub-lic could have great confience.

Substantial changes in public acceptance of nu-clear power are needed to accumulate the politi-cal capital for either large Federal budget expen-ditures or for Federal efforts to change utility rateregulation or utility institutional management.Two options included in Strategy Two go furtherthan the steps in Strategy One to alleviate thepublic’s concerns about nuclear power andsharpen the basis for judgment about the long-term need for nuclear power. (D4) reduces con-cerns over the linkage between nuclear powerand nuclear weapons by such steps as banningreprocessing. (D5) encourages or requires the useof regional planning and perhaps rate regulation,to improve the consensus on the long-term needfor capital intensive technologies such as nucle-ar power.

Future One.–Strategy Two would have a bigimpact on conditions such as those in Future Onewhich hypothesizes rapid demand growth andsuccessful industry improvement, but also util-ity reluctance to place the first nuclear order aftera hiatus in orders and substantial public contro-versy.

Perhaps the best vantage point from which toappreciate the possible need for Strategy Two isfrom a look forward to 1993 under the assump-tions of Future One if there have been no nuclearorders even though Strategy One has been

implemented. It is likely that the main ob-stacle to any nuclear orders would be utility un-certainty that all the problems had actually beenresolved, despite the steps in Strategy One. By

1993, more coal fired capacity would have beenordered over the previous 10 years than existedin 1983, but concerns over the environmental im-pacts would be rising sharply. Stringent new emis-sion regulations would appear probable, butwould be very expensive. The costs of other fuelsalso would be rising rapidly. Other industrial

countries, especially France, Japan, West Ger-many, Great Britain, and Canada would havemaintained their nuclear programs, and theircheaper electricity could give them a competitiveedge in certain areas of international trade. Util-ities would have raised their standards of nuclearoperation such that mishaps rarely would occurand reliability would be high. Americans in 1993might well wonder what went wrong with ournational decision making,

What would Strategy Two have accomplishedin the same period? How would utility uncertain-ty have been reduced? Under the conditions ofStrategy Two, utilities would have a clear dem-onstration of Federal government commitmentto resolving the problems with nuclear power andgood reason to expect that nuclear plants def-initely would be cheaper than coal plants in mostregions of the country.

By 1990, the standardized design would be suf-ficiently well proven that interveners, NRC andnuclear designers would be satisfied that it wasvery resilient to any accident sequences anyonesuggests. It would be clear to all that construc-tion of the first standardized plant was proceeding

smoothly on schedule and within budget. TheNRC would offer a construction permit based onits previous review of the existing design, withonly site specific characteristics to be approved.The financial burden on utilities would be less-ened because of reduced construction cost andchanges in rate regulation, and, perhaps, becausesmaller reactors would be available. Alternatively,the HTGR or one of the other reactors could bein an advanced stage of development.

Given the strong demand, low interest ratesand relative unattractiveness of other fuels in

Economic Scenario A, there is little doubt thatutilities would order nuclear plants if they wereconvinced costs would come down and publicacceptance would be sufficient to avoid serious




economic risk to the plants. Nuclear critics and

the public would have been reassured by a dec-

de of steadily improving nuclear plant perform-

nce, the closer involvement of interveners in

l icens ing, a nat iona l ly agreed l imi t on fu ture

nuclear construction and the several other meas-

res of Strateg ies One and Two . In ad dition , the

obvious need for generating capacity would leadto considerably greater publ ic acceptance for

more nuclear plants by 1993.

Under these conditions, some plants wouldbe ordered in the early nineties and probablya larger number in the late nineties. Strategy Two

thus makes probable what is only a possibilityunder the same favorable conditions of StrategyOne. While it is difficult to be precise, this policypackage could lead to more nuclear orders evenunder conditions somewhat less stimulating thanin Economic Scenario A. A growth rate in elec-tricity demand averaging 2.5 percent might havethe same probability of initiating nuclear ordersas 3.5 percent did under Strategy One.

Future Two.– If utility management were im-

proved but economic conditions were much less

favo rab le to nuc lea r orders, as in Future Two ,

much of the effort going into standardized design

could be wasted. When the time came for neworders n-ear or past 2000, a foreign design, analternative reactor type, or possibly photovoltaicsmight appear more appropriate. Thus, this effortmight be dropped if demand growth stays low.An effective program to develop alternative reac-

tor technology could prove to be the crucial re-assurance to utilities contemplating new ordersafter 2000. Utilities that might be unwilling to bethe first or second to order a new reactor typemight be willing to be the third or fourth.

The other steps of Strategy Two (coupled withtwo decades of good management) would beuseful in changing the climate of public opinionto be more receptive to nuclear power in the longterm. With demand growth of only 1.5 percentannually, such changes are likely to be impor-tant even when there has been good utility

management.Future Three. -Strategy Two might not be fea-

sible if utility management of nuclear reactors im-proved very little, as is assumed under Future

Three and Future Four. The consensus that wouldbe required for the large Federal budget expend-itures and the changes in Federal-State relationson utility rate regulation wouId not emerge.

For Future Three, with favorable economicconditions for nuclear orders, Strategy Two mightbe implemented in stages beginning with options

to restrict construction licenses to qualified util-ities (C2)—and perhaps revoke operating licensesas appropriate,—and changes in utility institu-tional structures (C4), allowing utility service com-panies to take over poorly run plants. In effect,these actions would change the situation to thatof Future One.

If these actions are insufficiently successful,future orders probably would be contingent onthe development of inherently safe reactorswhich, by allaying many safety concerns, wouIdrestore some degree of public acceptance.

Future Four.–Strategy Two would be veryhard to justify for Future Four with both little in-dustry improvement and unfavorable economicconditions. The options to improve utility man-agement, however, would help improve publicacceptance for after 2000, which is the earliestnuclear orders might be placed. It also is possi-ble that a modest level of investment in new reac-tor types could lead to a more ambitious effortin the nineties when public acceptance wouldbe improved as a result of NRC efforts to improveutility management.

Strategy Two is clearly a high-risk, high-gainstrategy. Under the circumstances of Future Oneit could assure the future of the U.S. nuclear in-dustry, although it might not even be necessaryto stimulate nuclear orders if the industry itselftakes sufficient steps to improve the technologyand if demand for electricity grows rapidly. Underother circumstances, Federal efforts and fundscould produce little of value (Future Two) or beimpossible to carry out (Futures Three and Four).

Strategy Two Variation: Support

the U.S. Nuclear Industry inFuture World Trade

In this va riation o f Strateg y Two , the Fed era l

Government also would play an activist role in



276 Ž Nuclear Power in an Age of Uncerta inty

sup po rting the U.S. nuc lea r industry. The ration-

ale for this variation would come, however, froma completely different source. Rather than regardnuclear power as an element of U.S. energy pol-icy, the strategy would treat the nuclear industry,and related electrotechnologies, as key elementsof an emerging U.S. industrial policy.

Several assumptions about the current natureof world competition in nuclear technologiesunderlie this approach. One is that the advancedreactor designs now underway with joint U. S.-Japanese teams will establish a new standardLWR for the 1990’s that will make more modestimprovements obsolete. These designs—probablyto be licensed jointly by Japanese and U.S.vendors—may well account for any nuclear or-ders in the 1990’s. These designs should give boththe United States and Japan a very strong posi-tion in world nuclear trade.

The second assumption underlying this ap-proach is that there will beat least one more gen-eration of non breeder reactors after these ad-vanced LWRs become available. The slowdownin plant construction, potential for extendedburnup and new uranium discoveries all wouldmake it likely that nonbreeder reactors will becompetitive for several more decades. The stand-ardized LWR discussed in option A2 or one ofthe advanced alternative reactors in C3 could bethe choice.

The question is: should the U.S. Government

encourage a strategy of R&D that leads to the nextstages of reactor development beyond the jointJapanese-U.S. vendor projects? Several elementsof a general industrial policy could be applied tonuclear power: 1 ) support for R&D into productdevelopment, 2) relaxation of antitrust prohibi-tions on cooperation among businesses duringthe product development stage of R&D, and 3)financing assistance in export promotion throughthe Export-Import Bank (7,8).

Such a strategy might be especially productivebecause of the historically low spending on R&Dby the electric utility industry. The $250 millionbudget of the Electric Power Research Institute(EPRI) represents only about 0.25 percent of elec-tric utility sales each year. General manufactur-ing industries spend about 2 to 3 percent of sales

on R&D and high technology industries mayspend up to 10 to 15 percent.

A major increase in R&D for the electric utilityindustry could be allocated among:

q elect rotechnologies for industry such asplasma reduction of iron or microwave heat-

ing;q commercialization of photovoltaics and the

solid-state control technology needed to in-tegrate them in the grid; and

q a more advanced nuclear reactor for beyond2000.

It seems likely that a balance among support fordifferent advanced generating technologies, in-cluding photovoltaics as well as advanced nuclearreactors, will be necessary to get widespread sup-port. Utilities also are likely to favor diverse R&Dbecause of the financial risk inherent in relyingon single technologies.

One approach to obtaining funding for suchan R&D effort is to make the treatment of R&Din utility rate regulation more attractive. Thetelephone industry traditionally has includedR&D in the rate base in most States.

Federal action in support of this policy strategywould have several elements:

1.

2.

3.

4.

5.

6.

Selection of these technologies as a signifi-cant element of U.S. long-term industrial pol-icy, and expansion of these R&D programs.Legislation providing incentives or requiringStates to allow R&D expenditures in the cost-of-service.Increased Federal funding for some jointEPRI-DOE projects,Elimination of antitrust penalties for vendorcooperation on an advanced reactor design.Consideration of long-term loans such as theJapanese government made available to thesemiconductor industry.Attractive financing to foreign customers fornuclear technology exports-through the Ex-port-import Bank.

Implementation of such a strategy would givethe U.S. a headstart in world competition justwhen U.S. orders could be expected to pick upbecause of load growth and replacement of ex-




isting plants. It could result in a healthy nuclear

industry with brighter export prospects, and moreattractive options of both supply and demand forthe electric utility industry. Electricity consumingindustries would also benefit, some quite signifi-cantly.

There are some obvious difficulties with thisstrategy. Given the Federal system, it may be verydifficult to get State support for more favorabletreatment of utility R&D expenditures. Althoughsome R&D will be carried out by manufacturers,a comprehensive strategy will not work unless amajor part of the impetus comes from the elec-

tric utilities. Utilities should derive a major partof the benefit, aside from being able to buy thedeveloped product. Another difficulty is that elec-trical demand technologies and advanced gen-erating technologies may not seem as importantas other U.S. industries in claiming a role in ageneral high-technology industrial strategy. Cur-

rent strategies tend to focus on computer, semi-conductor manufacture, and biotechnology rath-er than traditional manufacturing industries. TheR&D effort would also have to be designed withcare to ensure that the funds are spent produc-tively.

CONCLUSIONS

The scenarios and discussions of the previous

section underscore the conclusion that there isno simple key to restoring nuclear power as a na-tional energy option. Nor is there any assurancethat even a rigorous set of policy initiatives woulddo so. Uncertainties over the future growth rateof demand for electricity are at least as impor-tant as the policy options analyzed here. At thevery least, a moderate demand growth rate andmoderate improvements in management of exist-ing plants are prerequisites for creating the con-ditions under which utilities could consider order-ing additional nuclear plants.

The problems and uncertainties are great butnot insurmountable. If the technology and itsmanagement improve sufficiently that confidencein both safety and economics is much higher, ifnuclear regulation shows a parallel improvement,and if financial risks are shown to be less thanfinancial rewards, then nuclear power would bea logical part of our energy future. To see howthis might be so, consider the seven sides to thenuclear debate discussed in chapter 1:

qThe nuclear industry would have a productthat was thoroughly analyzed, demonstrably

safe and economically competitive.q The NRC would have exhaustively examined

the design and be confident that few addi-tional issues would be raised once a con-struction permit had been granted.

q

q

q

q

q

PUCs would have more confidence that a

utility would not bankrupt itself with a newreactor because costs would be predictableat the start due to matured designs and reg-ulatory stability, and operation of existingplants had proven to be in the best interestsof consumers.Investors could expect more favorable reg-ulatory treatment from the PUCs as well ashaving more confidence in the economic at-tractiveness of nuclear power.Critics would have far fewer specific con-cerns with safety and the overall threat of a“nuclear economy. ” This would lower theintensity of the controversy even if few criticschanged their minds on the inherent desir-ability of nuclear power.The public would be more supportivebecause of the lowered controversy, the im-proved operating records of existing plants,and the more visible benefits.Finally, utilities, or generating entities set upto replace or supplement them, would seethis improved environment for nuclearpower, the predictability of costs and opera-tions, and the affordability and com-petitiveness of the new plants, and would bemuch more receptive to proposals for newnuclear plants.

The purpose of the policy options discussedabove is to assist in the transition from the pres-





Demonstration in front of the Capitol after the accident at Three Mile Island

ent situation to this future. However, it is impossi-ble to say with any certainty how much improve-ment in any one factor is necessary, or how mucheach policy initiative would contribute. Theseproblems are interdependent: none can besolved in isolation, but the progress on each willassist in the solution of others.

The future listed above corresponds to FutureOne in the previous section, coupled with asmany policy initiatives as it takes to start nuclearorders. Strategy One might be adequate. StrategyTwo probably would be, but we cannot say forsure. The uncontrollable uncertainties are simply

too great. For instance, future plants can be de-signed to be safer than existing plants. industryalready is working on this. The national designof Strategy One would improve on this and bemore convincing to critics because of the open-ness of the design effort. The inherently safe reac-

tors of Strategy Two would be more reassuring,but would introduce cost and operational uncer-tainties. How far is it necessary to go to achievea consensus that reactors are safe enough? Itseems reasonable to assume that greater safetywould result in greater acceptance, but there islittle direct evidence to support that view or toquantify the relationship.

Any policy strategy should be flexible enoughto try things that may not work or may be foundinappropriate to changing conditions, and to dis-card the less successful initiatives in favor of thebetter ones. The strategy should include elements

dealing with the technology, its management, nu-clear regulation, financial risk, and public accept-ance.

Additional nuclear plants essentially have beenrejected by the American people because of per-




ceptions of the current technology and its man-

agement. Major improvements will have to bemade to restore their faith. To be successful, astrategy must recognize the different concerns

and try to balance the interests. In particular, the

role of the critics in any nuclear resurgence willbe crucial. Critics have been the messengers to

the public of many of the real problems with nu-clear power. They will continue to play that roleuntil they have been shown that the problemshave been solved or rendered inconsequential.They also c an be the messengers, even if onlyby losing interest, when they are convinced thatnuclear power as a whole is a minor problemcompared to other societal concerns. It is difficult

to conceive of how public acceptance can be im-

1.

2.

3.

4.

5.

6 .

7.

8 .

CHAPTER 9

Bauer, Douglas C., “Deregulat ion: Motivat ions,Modalities and Prospects, ” paper presented at the

conference on E/ectr/c Power Generation o r g a -

nized by the Energy Bureau, Inc.

Committee on Nuclear and Alternative Energy Sys-

tems, Energy in Transition 1985-2000 (Washing-

ton, D. C.: National Academy of Sciences, 1979. )

Fitzgibbons, Harold E., “A European Perspective

on U.S. High Technology Competition, ” in “High

T ech n o lo g y : public Policies for the 1980’ s,” ~a -

tiona/ Journal, March 1983.

Ford Founda t ion , Nuclear Power Issues and Choices (Cambridge, Mass.: Ballinger, 1977).Golub, Bennet W., and Hyman, Leo na rd S., “TheFinanc ial Diffic ulties and Co nseq uenc es of Dereg-

ulat ion Throug h Divestiture, ” Pub/ic Uti/ities Fort-night/y, Feb. 17, 1983.Graha m, A. J., “ Elect ricity Pricing: An Ind ustrialView, ” printed in Energy Technology V. Deregu- lation.Krist, William R., “ The U.S. Respo nse to ForeignIndustrial Policies, “ in “High Tec hno log y: Pub lic

policies for the 1980’ s,” Nat/ona/Journa/, March

1983.

Merrifield, D. Bruce, “Forces of Change Affect-ing High Tec hno log y Ind ustries, ” in “ High Tec h-

proved significantly while knowledgeable criticsstill are voicing real concerns.

The outlook for the nuclear supply industry isbleak but not hopeless. New policy initiatives canset the stage for a turnaround. Without appro-priate action, it is likely that the option of

domestically produced nuclear reactors graduallywill fade away. If such is to be the future, thedecision should be made consciously, with theknowledge that even the nonnuclear futuresavailable to the Nation contain risks anddrawbacks. Nuclear power can be a significantcontributor to the Nation’s energy future, butonly if the efforts to restore it are undertaken withwisdom, humility, and perseverance.

REFERENCES

no logy : Public Policies for the

Journal, March 1983.980’s, ” Nationa

9. Meyer, Micha el B., “ A Mod est Prop osal for thePartial Deregulation of Electric Utilities, Pub/ic

Utilities Fortnjghtly, Apr. 14, 1983.10. Ma ssa c huset ts Institute of Tec hno log y, Deregu/a-

11

tion o~Electrjc Power: A Framework for Analysis,Phase I rep ort, DOE/ N BB-0021, Sep te mb er 1982,

pub lished as: Pa ul Joskow and Richard Sc hm a len-

see, Markets for Power: An Analysis of Electric W/ity Deregu/atjon (Ca mb ridge, Ma ss.: MIT Press,1983).

Office of Tec hno logy Assessme nt, / ndustria/ and

Cornmercia/ Cogeneratjon, OTA-E-1 92, February1983.

12. Oosterhuis, Paul, “ High Tec hno logy Ind ustriesa nd Ta x Policy in the 1980’ s,” in “ High Tec hno l-og y: Pub lic Policies for the 1980’ s,” Natjona/ journal, March 1983,

13. Radford, Bruce C., “ Ma rginal Co st Pric in~ Consid-

14

ered i n Rec ent Elec tric Rat e C ases, ” Pub/ic Uti/ ities

Fortnight/y, Oct. 9, 1980.

Samuels, Warren J., “ Prob lems of Marginal Cost

Pric ing in Public Utilities, ” Pub/ ic Uti/ itie s Fort-

night/y, Jan. 31, 1980.



Appendix



Appendix

List of Acronyms and Glossary

List of Acronyms

ACRS —Advisory Committee on ReactorSafeguards

AE —a rc hitec t-eng ineers

AEC –Atomic Energy CommissionAFUDC –allowance for funds used during

constructionAIF –Atomic Industrial Forum

AGR —advanc ed g as reac tor

ANS –Ame ric an Nuclea r Soc iety

ASME –America n Soc iety of Mec hanica lEngineers

BWR —bo iling wa ter rea cto rCAA –Clean Air Act

CEGB –Central Electricity Generating Board

COL —construction and operating license

CP —construction permit

CPCN –Certificate of Public Convenience andNecessity

CRBR –Clinch River Breeder Reactor

CRGR –Committee for Review of GenericRequirements

CRS –Congressional Research ServiceC WA –C le an Wa te r Ac tCWIP —construction work in progress

DOE –Department of EnergyDRI –Da ta Re so urc e s, In c .

EIS –Environm enta l Imp a c t Sta tem ent

EPRI –Electric Power Research InstituteERDA —Energy Research a nd Development

Administration

ER –Environmental Report

FEMA –Fed eral Emergenc y Mana ge ment Ag enc yFERC —Federal Energy Regulatory CommissionFSAR –Final Safety Ana lysis Rep ortGA P –Government Accountability ProjectGCR –gas cooled reactorGNP –gross national product

GPU –General Public UtilitiesHTGR —high te mp erature ga s-coo led reac tor

HW R –heavy water reactorIDCOR –Industry Degraded Core Rulemaking

ProgramIN PO —Institute of Nuclear Power Operations

JCAE –Joint Committee on Atomic EnergyKWU —Kra f twork Un ion

LMFBR –liquid metal fast breeder reactorLWR –lig ht wa te r re a c to rs

MITI –Ministry of Interna tiona l Trad e a nd In-dustry (Japan)

NAS –Nat iona l Academy o f Sc iencesNASA –National Ae rona utics an d Spa ce

Administration

NCRP –National Commission on RadiologicalProtection

NEIL –Nuclear Electric Insurance Ltd.

NEPA –National Environmental Policy Act

NERC –Northeast Electric Reliability Council

NML –Nuc le a r M utua l Ltd.

NPCC –Northeast Power Coordinating CouncilNPDES —Nat iona l Polluta nt Disc ha rge Elimina tion

Syste m p ermitNPRDS –Nuclea r Plant Reliab ility Da ta Syste mNPT –Non -Proliferat ion Trea ty

NRC —Nuclear Regulatory Commission

NSAC —Nuclea r Sa fety Ana lysis Ce nte rNSSS —nuclear steam supply system

NTOL —nea r-te rm op era ting licensesNUTAC –Nuc lea r Utility Task Ac tion Co mm itteeOECD –Organization for Economic Cooperation

and Development

OL —operating licenseO TA –Offic e o f Tec hno log y Assessme nt

PCRV –prestressed concrete reactor vesselPG&E –Pacific Gas & Electric Co

PHWR –pressurized heavy water reactorPlus –process inherent ultimately safe reactorPRA –probab ilistic risk assessmentPSAR –Preliminary Safety Analysis ReportPUC —public utility commissionPURPA —Pub lic Utility Reg ula to ry Po lic ies Ac t

PWR –pressurized water reactorRNPA –regional nuclear power authorityRNPC –regional nuclear power companySAI –Sc ienc e Ap p lica tions, Inc .

SEE-IN –Sign ifica nt Events Eva lua tion and Infor-mation Network

SER –Sa fet y Eva luation Rep ort

SNUPPS–Sta nd ardized Nuc lea r Unit Pow er PlantSystem

SPDS –Safety Parameter Display Systems

SPP –Sou thw est Pow er Pool

SRP –Sta nd ard Review PlanTM I –Three Mile IslandTVA –Ten nessee Va lley Authority

WPPSS –Washington Public Power SupplySystem

283




Glossary

absorption, neutron: Any reac tion in which a freeneutron is ab sorbed by a nuc leus, including c ap -ture and fission.

Allowance for Funds Used During Construction(AFUDC): An account in the income statem ent of

a utility in which interest is accumulated on the con-struction expenditures for construction work inprogress that has not been entered into the utility’srate ba se a nd is therefore not yet e arning a ca shreturn on investment. The a c c umula ted interest isthen ad ded to the ac tual construction expenditureswhen the plant enters the rate base.

base loaded: Keep ing a po we r station c ontinuouslyloaded at the maximum load because it is one of

the lowest cost power producers on the system.boiling water reactor: A reactor cooled by water that

is allowed to boil as it passes through the core. Thiscoolant is used directly to produce the steam whichgenerates electricity.

capacity factor: Ratio of average plant electricalenergy output to rated output.chain reaction: The continuing process of nuclearfis-

sioning in which the neutrons released from a fis-sion trigger at least one other nuclear fission.

cladding: The term used to d esc ribe any m ate rial thatenc loses nuclear fuel. In a wa ter-co oled pow er reac -

tor this is the fuel rod tube.Construction Work in Progress (CWIP): An account

on the asset side of the utility’s balance sheet thatincludes all construction expenditures for plant andequipment on p lant that has not yet been p lac edi n service .

construction Ieadtime: The time required to c om plete

construction of an electric generating plant, usual-ly defined from either da te of reac tor orde r or co n-struction permit to commercial operation.

containment building: A thic k c onc rete structure sur-rounding the pressure vessel and other reac tor co m-ponents. I t is designed to prevent radioactivematerial from being released to the atmosphere inthe unlikely event that it should escape from thepressure vessel.

control rods: Long thin rods that are positioned

am ong fuel rods to regulate the nuc lear c hain reac -t ion. Control rod s are c omp osed of m aterial that a b-

sorb s ne utrons rea d ily. The y interrupt o r slow d ow na chain reaction by capturing neutrons that would

otherwise trigger more fissions.coolant: Fluid that is circulate d throug h the co re o f

a reactor to remove the heat generated by the f is-sion process. In rea c tors that have mo re than onec oo lant system , the fluid whic h p a sses throug h the

co re o f a reac tor is known a s the p rima ry co olant.It absorbs heat in the core and then transfers it toa sec ond ary co olant system.

core: The region of a rea ct or in which the nuc lea rchain reaction is initiated, maintained, and con-trolled. Coolant is constantly circulated through the

core to remove heat produced by the fissionprocess.

decay heat: The heat p roduc ed by radioac t ive dec ayof materials that are primarily the remnants of thechain reaction.

deplete: To red uc e the fissile c ontent of a n isotop icmixture, particularly uranium.

elasticity: The ratio of c hange in de ma nd for a p rod -uct (in this ca se e lec tricity) to c hange in a c ateg oryof prices, or to change in income.

emergency core cooling system: Any engineeredsystem for c oo ling the co re in the event of failureof t he ba sic c oo ling system , suc h a s c ore sprays orinjectors.

enrichment: The p roce ss of increa sing the c onc en-tration of one isotope of a given element.

fabrication: The final step in prepa ring n uc lear fuelfor use in a reactor.

fast breeder reactor (FBR): A reactor cooled by liquidsodium rather than waste. In this type of reactor,

the transformation of uranium-238 to plutonium oc-c urs read ily. Since p luton ium fissions ea sily, it c a n

be rec ycled a nd used as fuel for a breede r reac tor.The c on version o f uranium to p luton ium is so e ffi-

cient in an FBR that this reactor creates more fuelthan it consumes.

feedwater: Water, usually from a condenser, supplied

to replenish the water inventory of componentssuch as boilers or steam generators.

fertile: Ma terial co mp osed of ato ms which readily ab -sorb neutrons to produce fissionable materials. Onesuch element is uranium-238, which becomes plu-

tonium-239 after it absorbs a neutron. Fertilematerial alone cannot sustain a a chain reaction.

fissile: Material composed of atoms which readily fis-sion when struck by a neutron. Uranium-235 andplutonium-239 are examples of fissile materials.

fission: The p roc ess b y wh ich a ne utron strikes anucleus and splits it into fragments. During the proc-ess of nuclear fission, several neutrons are emittedat high spee d, and heat a nd radiation are released .

fission products: The sma ller ato ms create d w hen anuc leus fissions. The ma ss o f the fission p rod uc ts

is less tha n tha t o f the origina l nuc leus. The d if-ference in mass is released as energy.

fossil plant: A p owe rplant fueled b y c oa l, oil, or gas.fuel: Basic chain-reacting material, including both

fissile and fertile materials.



Appendix: List of Acronyms and Glossary Ž 285

fuel cycle: The set o f che mica l and p hysic al op era-

tions needed to prepare nuclear material for use inrea cto rs and to d ispo se o f or rec ycle the m ate rialafte r its rem ova l from the reac tor. Existing fuel c yc lesbegin with uranium as the natural resource andc rea te p lutonium as a byp rod uct. Some future

cycles may rely on thorium and produce the fissile

isotope uranium-233.fuel rod: An assembly consisting of a capped zircalloyor stainless steel tube filled with fuel pellets.

half life: The p eriod required for an unsta ble rad io-active element to decay to one-half of its initialmass.

heat rate: A me asure of the am ount o f fuel used toproduc e e lec tric and / or thermal energy.

total heat rate refers to the am ount o f fuel (in Btu)required to produce 1 kilowatt-hour of elec-tricity with no credit given for waste heat use.

incremental heat rate is calculated as the addi-

tional (or save d) Btu to p rod uce (or not pro-duce) the next kilowatt-hour of electricity.

ne t heat rate (also measured in Btu/kWh) creditsthe thermal output a nd d enotes the energyrequired to produce electricity, beyond whatwouId be needed to produce a g iven quant i-ty of thermal energy in a separate facility (e.g.,

a boiler).interest coverage ratio: The ratio o f a firm’ s ea rning s

to its current interest obligations.isotopes: Atoms having the same number of protons,

but a different number of neutrons. Two isoto p esof the same atom are very similar and difficult to

sep arate by ordinary che mica l mea ns, Isotop es c anhave very different nuc lear prop erties, howe ver. Forexample, one isotope may fission readily, while

ano ther isoto pe of the sam e a tom ma y not fissionat all.

light water reactor: A general term that refers to allnuclear reactors which use ordinary water as a cool-ant. This includes pressurized water reactors andboiling water reactors, which are the predominantreactors in the United Sta te s.

load: The demand for electric or thermal energy atany particular time.

base load is the normal, relatively constant de-mand for energy on a given system.

peakload is the highe st d ema nd for energy from

a supplying system, measured either daily,

seasonally, or annually.

intermediate load fa l ls between the base andpeak .

load factor is the ratio of the average load overa d esignated time p eriod to the pea k load oc-

curring during that period. Also used as asynonym for capacity factor.

load eye/e pattern is the va riation in dem and over

a specified period of time.loss-of-coolant accident (LOCA): A reac tor ac cident

in which coolant is lost from the primary system.

market potential: The n umb er of insta nc es in which

a technology will be sufficiently attractive-all thingsconsidered—that the investment is likely to bem a d e .

market-to-book ratio: The ratio o f the m a rket p ric eof a firm’s stock to its book value.

MWe: Megawatts of electrical energy.MWt: Mega wa tts of thermal energy.

moderator: A c omp onent (usually wa ter, heavy w ater,or grap hite) of som e nuc lear rea c tors that slows neu-trons, thereby increasing their chances of being ab-sorbed by a fissile nucleus.

neutron: A ba sic ato mic p article that ha s no electric alcharge. Neutrons and protons, which are positive-ly charged particles, form the central portion of the

atom known as the nucleus. Negatively chargedelectrons orbit the nucleus at various distances. Thechemical and nuclear properties of an atom aredetermined by the number of its neutrons, protons,and electrons.

neutron poison: The ge neral name g iven to ma terialstha t ab sorb n eu trons. These ma te ria ls eithe r in-terfere with the fissioning process or are used to

control it.nuclear island: The b uildings and eq uipm ent tha t

co mp rise the reac tor and a ll its eme rgenc y and aux-ilia ry system s.

nuclear steam supply system (NSSS): The b asic reac -tor and support equipment, plus any associated

equipment nece ssary to produc e the steam thatdrives the turbines.

once-through fuel cycle: A nuclear system whereinnuclear materials are introduced into a reactor onlyonce ; they are not recycled.

plutonium: An element that is not found in nature,but can be produced from uranium in a nuclearreactor. Plutonium fissions easily, and can be usedas a nuclear fuel.

power density: The po we r ge nerated pe r unit volumeof the core.

pressure vessel: A heavy steel enclosure around the

co re o f a rea ct or. It is de signe d to withstand highpressures and temperatures to prevent radioactive

material from escaping from the core.pressurized water reactor: A reactor cooled by water

that is kept at high pressure to prevent it from boil-ing. Primary coolant passes through the core of a




PWR, and then transfers its heat to a secondary

c oo lant system . Stea m is prod uc ed from the hea tedwater in the secondary system.

primary coolant: The fluid u sed to c oo l the fue l

elements. It may be liquid or gas.qualifying facility: A cogenerator or small power pro-

ducer that meets the requirements specified in thePub lic Utility Reg ulatory Polic ies Ac t o f 1978–in theca se of a co gene rator, one that p rod uce s electric i-ty a nd useful therma l energy for industrial, co mm er-

cial, heating, or c oo ling purpo ses; that mee ts theop erating req uirem ents spe c ified b y the Fed eral En-ergy Regulatory Commission with respect to suchfactors as size, fuel use, and fuel efficiency); and

that is owned by a person not primarily engagedin the generation or sale of electric power (otherthan cogenerated power).

radioactive decay: The p roc ess b y which a nuc leusof one type transforms into a nother, ac c omp aniedby emission of radiation.

radioactive waste: Waste materials, solid, liquid, or

gas, that are produced in any type of nuclear facility.rate base: The ne t va luation of utility p rope rty i n serv-ice, consisting of the gross valuation minus accrueddepreciation.

reactor: A fac ility that c onta ins a c ontrolled nuc lea rfission cha in reac tion, It m ay be used to ge nerateelectrical power, to conduct research, or exclusivelyto produce plutonium for nuclear explosives.

reactor containment boundary: The p ressure enve-lope in which a reactor and its primary cooling sys-

tem are loca ted.reactor vessel: The c onta iner of the nuc lear core or

c ritica l assem b ly; ma y be a steel p ressure ve ssel, aprestressed concrete reactor vessel (PCRV), or a

low-pressure vessel (e.g., a c aland ria or sod ium p ot).reprocessing: Chemical treatment of spent reactor fuel

to separate the plutonium and uranium from the fis-sion products and (under present plans) from eachother.

safeguards: Sets of reg ulat ions, proc ed ures, andequipment designed to prevent and detect thediversion of nuclear materials from authorized

channels.safety system: A me c hanic al, elec tric al, or instrumen-

tation system or any c omb ination o f these, who se

purpose is the safety of the reactor or of the public.scram: The rap id shutdo wn, via introd uc tion of

neutron absorbers, of the chain reaction.

seismic load: The stresses impo sed on a c om po nen tby a seismic shock.

shutdown: The ac t of stopp ing p lant ope ration for an y

reason.spent fuel storage pool: The p oo l of d em ineralized

water in which spent fuel elements are stored pend-ing their shipment from the facility.

spent nuclear fuel: Material that is removed from areactor after it can no longer sustain a chain reac-tion. Spe nt fuel from a ligh t wa ter rea c tor is c om -posed primarily of uranium and contains someradioactive materials, such as fission products. Spentfuel also contains some valuable nuclear materials,such as uranium-235 and plutonium.

steam generator: The m a in hea t exch a ng ers in apressurized wa ter or ga s-co oled reac tor po we rplantthat generates the steam that drives the turbine gen-

erator.thermal efficiency: In a powerplant, the ratio of net

electrical energy produced to total thermal energyreleased in the reactor or boiler.

thermal load: The stresses impo sed on a c om p one ntdue to restrict ion o f thermal growth c aused by tem-perature cha nges.

thermal neutron: A neutron whose energy level hasbeen lowered sufficiently so that upon collision with

another atom it will cause the atom to split andrelease energy. Neutron energy levels can be

lowered by recoil off moderating atoms.

thorium-232 (Th232): A fetile, naturally occurring iso-top e from which the fissile isoto pe uranium-233 ca nbe b red .

turbine generator: The a ssem b led stea m turbine

coupled to an electric generator that produces theelectric po wer in a pow erplant.

uranium: A metallic element found in nature that iscommonly used as a fuel in nuclear reactors. As

found in nature , it conta ins two iso top es—

uranium-235 and uranium-238.

uranium-233 (U233): A fissile isotope bred by fertile

thorium-232. It is similar in weapons quality to plu-tonium-239.

uranium-235 (U235): The less ab und a nt uraniumisoto pe , ac c ounting for less than o ne pe rcent of na t-ural uranium. Uranium-235 splits, or fissions, whenstruck by a neutron. When uranium is used as a fuelin a nuclear reactor, the concentration of urani -

um-235 is often increased to enhance the fissionp roce ss. For exam ple, the fuel for light wa ter rea c -

tors contains about 30/0 uranium-235.uranium-238 (U238): The m ore a bund ant uranium

isotope, accounting for more than 99 percent of nat-

ural uranium. Uranium-238 tends to absorb neu-trons rather than fission. When it absorbs a neutron,

the uranium atom changes to form a new element-plutonium,

water hammer: The shoc k loa d imp osed on a flow -ing pipeline by the rapid closure of a shutoff valve.



Index



Index

Advisory Council on Historic Preservation, 149Advisory Committee on Reactor safeguards (ACRS), 145,

164, 165Alternative reactor systems, 83-109, 258 (see also

nuclear powerplant technology)advanced light water reactor design concepts, 94-96basics in nuclear powerplant design, 84

heavy water reactors, 96-99 high temperature gas-cooled reactor (HTGR), 99-102,103

inherently safe reactor concepts, 102-105light water reactors, safety and reliability of, 87-94,

102comparison of fossil units to all nuclear units, 89examples of specific concerns, 90overview of U.S. reactors, 87reliability concerns, 88safety concerns, 87

probabilistic risk assessment, 88unresolved safety issues, 91

small reactor, 105-107standardized reactor, 107

American Electric Power Co., Inc., 136

American Nuclear Society, 88, 214American Physical Society, 218American Society of Mechanical Engineers (ASME), 182architect-engineers (AEs), 15, 22, 87, 107, 114, 135,

179, 183Argentina, 200, 202, 203Arizona Public Service Co., 115, 125, 127ASEA-ATOM, 103, 105Atomic Energy Commission, 101, 144, 218, 227

WASH-740 study, 218Atomic Industry Forum, 214Audubon, 32Australia, 197

Babcock & Wilcox Co., 87

Baltimore Gas & Electric Co., 5Bechtel Corp., 127Blue field Water Works and Improvement Co. v. West

Virginia Public Service Commission, 51Bonneville Power Administration (BPA), 138Boston Edison Co., 118Brazil, 202, 203Brown & Root, Inc., 127Browns Ferry, 6, 87, 122, 123, 153, 213Buss, David, 228

California, 136, 151, 216, 231Calvert Cliffs plant, 5, 122Canada, 17, 18, 23, 71, 96, 98, 99, 109, 138, 179, 182,

191, 200, 203, 237

Carolina Power & Light Co., 118Carter administration, 203case studies, 240-244Central Electricity Generating Board (CEGB), 96

Certificates of Public Convenience and Necessity(CPCN), 151

C. F. Braun, Co., 136Cincinnati Gas & Electric Co., 116, 157Colorado Public Service Co., 101combined construction and operating license (COL),

165, 168, 169

Combustion Engineering, Inc., 87Committee for Energy Awareness, 214, 236Committee for Review of Generic Requirements

(CRGR), 130, 157, 159Commonwealth Edison, 59, 66, 67, 126, 148, 168Connecticut Mutual Life Insurance Co., 217Connecticut Yankee, 136Congress:

Congressional Research Service (CRS), 201House Committee on Science and Technology, 8House Interior Committee, 220House Subcommittee on Energy and Environment,

219Joint Committee on Atomic Energy, 144Senate Committee on Energy and Natural Resources,

8

Conservation Foundation, 237construction permit (CP), 145, 148, 151, 157, 158, 160,

161, 163, 165, 168, 170Consumer Product Safety Commission, 164Council on Environmental Quality, 149Critical Mass Energy Project, 214

Data Resources, Inc. (DRI), 33, 34Decision Research, 222, 225Department of Agriculture, 149Department of Defense, 136, 149Department of Energy (DOE), 33, 38, 44, 46, 60, 61,

65, 102, 136, 149, 154, 157/ 158, 159, 161, 162, 165,168, 169, 170, 172, 179, 185, 228, 233, 239Office of Policy Planning Analysis, 33

policy options, 253Department of Housing and Urban Development, 149

701 Comprehensive Planning Assistance Program, 151Department of the Interior, 149Department of Justice, 132, 171, 260

Attorney General, 145Department of Transportation, 149Detroit Edison’s Fermi Breeder reactor, 213Diablo Canyon, 116, 153, 228, 230

case study, 242Donaldson, Thomas, 230Duke Power Co., 5, 60, 66, 67, 136DuPont, Robert, 222, 233

Ebasco Services, Inc., 127

Edison Electric Institute, 32, 214Eisenhower administration, 144Electric Power Research Institute (EPRS), 46, 62, 63, 73,

88, 115, 118, 127, 182

289




Energy Information Agency, 32Environmental Impact Statement (EIS), 145, 148, 168Environmental Protection Agency, 149, 220Eugene, Oreg., 221

Federal Aviation Agency, 149Federal Energy Regulatory Commission (FERC), 138, 255Federal Power Commission v. Hope Natural Gas Co.,

51

Federal Register, 166, 167Federal Trade Commission, 163financial and economic future, 13-15, 29-76

cost of building and operating nuclear powerplants,57-71

cost of electricity from coal and nuclear plants, 64financial risk of operating a nuclear powerplant, 68

risk of public liability, 69future construction costs of nuclear powerplants, 66impact of delay on cost, 63impact of risk on the cost of capital, 70increase in nuclear construction Ieadtimes, 62rapid increase in cost, 58reasons for increased construction costs, 60

electricity demand, 13, 29, 31-42

sources of uncertainty, 33-41estimated impact of industrial electrotechnologies in

the year 2000, 38

nuclear power in the context of utiltity strategies,71-76

alternative utility construction plans, 73estimated cost of nuclear plants under construction,

76implications for Federal policies, 74

plant construction, 14, 29rate regulation and powerplant finance, 13, 46-57

accounting for capital costs of powerplants underconstruction, 50, 64

allowance for funds used during construction(AFUDC), 47, 50, 51, 52, 57

changes in rate regulation, 52construction work in progress (CWIP), 50, 52, 56,57, 71

history of the deterioration in the financial health ofelectric utilities, 1960-82, 48

impact of changes in rate regulation in electricityprices, 57

implications of utilities’ financial situation, 50market-to-book ratios and dilution of stock value,

47obstacles to a long-term commission perspective, 56pay-as-you-go inflation schemes, 56phased-in rate requirements, 52public utility commissions (PUC), 50, 51, 52, 56, 57utilities current financial situation, 46utility accounting, 53

recent past, 29reserve margins and retirement, 42-46

economic obsolescence, 43loss of availability of generating capacity, 44need for new powerplants, 45retirements due to age, 43

Florida, 59, 72, 92Florida Power & Light Co., 59, 72, 115, 136Ford, Daniel, 230Font St. Vrain, Colo., 9, 101, 102, 128France, 22, 23, 67, 189, 191, 196, 199, 200, 202, 203

backfits, 197siting of plants, 198

Fuel Supply Service, 136

Garrett, Pat, 240General Accounting Office (GAO), 171General Atomic Co., 101General Electric Co., 87, 94, 180, 218General Public Utilities (GPU), 68, 136

Georgia Power Co., 40Great Britain, 22

Great Lakes Basin Commission, 149gross national product (GNP), 29, 32, 33, 34, 36, 40, 41

Gulf Power Co., 40

Hodel, Donald, Secretary of Energy, 162Houston Lighting & Power Co., 127Hutchinson, Fla., 115

Illinois, 56, 136impasse, 5

India, 202, 203Indian Point Station, 150

Indiana Public Utility Commission, 56Industry Degraded Core Rulemaking Program, 88Inglehart, Ronald, 229

Institute for Nuclear Power Operations, 9, 19, 20, 25,128, 130, 131, 133, 134, 137, 185, 219, 255near-term operating licenses (NTOL), 132

Significant Events Evaluation and Information Network(SEE-IN), 129, 130

Systematic Assessments of License Performance

(SALP), 131

Japan, 22, 23, 102, 191, 194, 199, 200, 203Ministry of International Trade and Industry (MITI),195, 198

siting of plants, 198

Kasperson, Roger, 222

Kemeny Commission, 231, 233Komanoff, Charles, 58, 65Korea, 194, 200

League of Women Voters, 230, 236, 239legislation:

Administrative Procedures Act (APA), 164Atomic Energy Act, 21, 22, 143, 144, 151, 154, 160,

161, 164, 172, 203

Clean Air Act, 151

Clean Water Act (CAA), 151Economic Recovery Tax Act of 1981, 50, 51Energy Reorganization Act of 1974, 144Foreign Assistance Act of 1961, 203Fuel Use Act, 72

National Environmental Policy Act (NEPA), 151



Index . 291

Nuclear154

NuclearNuclear

154Nuclear

Licensing and Regulatory Reform Act of 1983,

Non-Proliferation Act of 1978, 202, 203Powerplant Licensing Reform Act of 1983,

Waste Policy Act of 1982, 264price-Anderson Act, 69, 144, 264Public Utility Regulatory Policies Act (PURPA), 138

Lifton, Roger, 222Lloyd’s of London, 69Long Island Lighting Co., 60Lonnroth, Mans, 190, 199, 200Louis Harris & Associates, 228Lovins, Amory, 229

Maine Yankee, 136, 222, 228, 230, 236case study, 240

management of nuclear powerplants, 18-20Massachusetts, 136, 215Massachusetts Institute of Technology, 220Mazur, AlIan, 231Mexico, 194Michigan, 45, 136Middle South Utilities, Inc., 136

Mitchell, Robert, 226

Nader, Ralph, 214, 215, 228National Aeronautics and Space Administration, 136,

182

National Interveners, 228National Oceanic and Atmospheric Administration, 149,

164National Opinion Research Center, 227National Organization for Women (NOW), 231National Pollutant Discharge Elimination System

(NPDES), permit, 151

National Technical Information Service, 10Nelkin, Dorothy, 230Netherlands, 231, 234

New England Electric System, 72New Hampshire public Service Co., 136New Jersey, 68, 118New York, 56, 135, 150New Zealand, 197Non-Proliferation Treaty (NPT), 202Northeast Electric Reliability Council (NERC), 44, 45Northeast Power Coordinating Council (NPCC), 44nuclear disincentives, 4Nuclear Electric Insurance Ltd. (NEIL), 68nuclear enterprise, management of, 113-139

assessments of efforts to improve quality, 132-134improving the quality of nuclear powerplant manage-

ment, 127-132classifications for INPO evaluations, 132creating the right environment, 129detecting and improving poor performance, 130institutional approach, 128technical approach, 127

management challenges, 118-127comparison of manpower requirements, 122

external factors, 121historical labor requirements, 123internal factors, 124technological factors, 118

new institutional arrangements, 134certification of utilities, 137Government-owned regional nuclear power

authority, 138larger role for vendors in construction, 135privately owned regional nuclear power company,

138service companies, 136

variations in quality of construction and operation,114-118performance of selected U.S. LWRs, 117

Nuclear Mutual Ltd. (NML), 68nuclear powerplant technology, 15-18

boiling water reactor, 84, 85, 86, 90, 92, 94heavy water reactor, 17, 84, 96-99, 175high temperature gas-cooled reactor (HTGR), 16,

17, 84, 87, 99, 100-102, 128, 175light water reactors (LWRs), 15, 16, 18, 83, 84, 85,

88, 93, 94, 96, 97, 98, 102, 113, 128, 156, 174safety and reliability of, 87-94

construction records, 115process inherent ultimately safe reactor, 17, 18, 103,

104, 105, 175pressurized water reactor, 84, 85, 86, 90, 94, 95, 96prestressed concrete reactor vessel (PCRV), 99, 100,

103

Nuclear Regulatory Commission (NRC), 4, 5, 6, 16, 18,19, 20, 23, 25, 26, 63, 69, 87, 90, 92, 93, 115, 118,120, 122, 123, 128, 133, 134, 137, 145, 150, 154,156, 157, 158, 159, 164, 165, 166, 168, 180, 213,233approval of site suitability, 170Committee for Review of Generic Requirements, 130,

157, 159

Office of Nuclear Reactor Regulation, 145, 148, 149,

155Performance Assessment Team (PAT), 131, 132

policy options, 253PSAR, 150Regional Administrators, 167Regulatory Reform Task Force, 158responsibilities in licensing, 146Safety Evaluation Report (SER), 145, 148Standard Review Plan (SRP), 145, 166

Nuclear Safety Analysis Center (NSAC), 129nuclear steam supply system (NSSS), 135, 145, 179,

180, 182, 188Nuclear Test Ban Treaty, 227Nucleonics Week, 181NUS Corp., 118

Ontario Hydro, 138operating license (OL), 148, 149, 151, 160, 161, 163,

165, 168Oregon, 151, 215, 217, 221




Organization for Economic Cooperation and Develop-ment, 189

Pacific Gas & Electric Co., 116Pahner, Philip, 222Pakistan, 202, 203Palo Verde plants, 125Pennsylvania, 68, 72, 213, 230Philadelphia Electric Co, 101

Philippines, 135policy options, 3, 25-26, 251-278

major Federal policy strategies, 263-277base case: no change in Federal policies, 263base case variation: let nuclear power compete in a

free market, 270economic conditions: two scenarios, 265four nuclear futures under the base case, 267management improvement conditions: two

scenarios, 266policy goals and options, 252-262

alleviate public concerns and reduce political risks,260

improve reactor operations and economics, 255reduce capital costs and uncertainties, 252

reduce the risk of accidents that have public safetyor utility financial impacts, 256strategy one: remove obstacles to more nuclear

orders, 271strategy two: provide moderate stimulation to moreorders, 273

Porter Commission, 237Portland General Electric, 217, 222Preliminary Safety Analysis Report (PSAR), 145President’s Commission on the Accident at Three Mile

Island, 166Princeton University, 220public attitudes toward nuclear power, 23-25, 211-244

actors in the nuclear power debate, 214expert’s view, 217-220factors influencing the public’s view on nuclear safety,

221

psychological factors, 222history of statewide referendum votes dealing with

nuclear powerplants, 212impact of public opinion on nuclear power, 215impact of risk assessments on public opinion, 220public acceptance of nuclear power, 234-244role of the media, 24, 231-234State laws and regulations restricting construction, 216values and knowledge, 226

Public Citizen, Inc., 214Public Service Electric & Gas Co., New Jersey, 118, 214purpose of study, 8

Quadrex Corp., 136

Rasmussen Report, 69, 218, 220Reactor Safety Study, 218, 219Reagan administration, 203regional nuclear power company (RNPC), 138

regulation of nuclear power, 21-22, 143-175backfitting, 154-160

definition, 158legislation, 160specific concerns, 155

Federal regulation, 143-150historical overviews of nuclear regulation, 144NRC responsibilities in licensing, 146role of emergency planning, 150

utility responsibilities in licensing and construction,147

hearings and other NRC procedures, 160-167changes in role of NRC staff, 164efficiency improvements, 162enforcement, 167improvements in public participation, 162management control, 165other NRC procedural issues, 164proposals for change, 161use of rulemaking, 166

issues surrounding nuclear plant regulation, 153-154licensing for alternative reactor types, 174other NRC responsibilities, 171safety goals, 172

State and local regulation, 150-153environmental policy, 151need for power, 151State siting activities, 153State siting laws, 152

two-step licensing process, 168-170Richland, Wash., 126River Basin Commissions, 149Rochester Gas & Electric, 213Rogovin Report, 164, 166Rumania, 194

Salem, N. J., 93, 214, 219, 221, 236Sandia Siting Study, 219Schumacher, E. F., 229

Science Applications, Inc., 219selected U.S. light water reactors, comparison of, 19Shadis, Ray, 240Shoreham plant, N. Y., 5S. M. Stoner Corp., 179Solar Energy Research Institute, 32South Africa, 202Southern Co., 136Southwest Power Pool (SPP), 44Spain, 200Standardized Nuclear Unit Power Plant System

(SNUPPS), 96Stanford University, 220Strauss, Lewis, 144survival of the nuclear industry, 179-206

effects in the U.S. nuclear industry of no, few, or

delayed new plant orders, 179architect-engineering firms, 183engineering manpower, 181impact on future new construction of nuclear

plants, 186



Index q 2 9 3

nuclear power in an age of uncertainty

Documents